**Part 3**

**Autonomic Dysregulation** 

190 Advances in Electrocardiograms – Clinical Applications

Yan, L., et al. (2005). Autophagy in chronically ischemic myocardium. *Proc Natl Acad Sci U S* 

Yu, Y., et al. (2005). Differential impact of prostaglandin H synthase 1 knockdown on

Yu, Y., et al. (2006). Genetic model of selective COX2 inhibition reveals novel heterodimer

Yu, Y., et al. (2007a). Targeted cyclooxygenase gene (ptgs) exchange reveals discriminant

Yu, Z., et al. (2007b). Protective effects of Shuangshen Ningxin capsule on miniature swine

after myocardial ischemia by intervention. *Zhongguo Zhong Yao Za Zhi,* Vol. 32, No.

platelets and parturition. *J Clin Invest,* Vol. 115, No. 4, pp.986-995

isoform functionality. *J Biol Chem,* Vol. 282, No. 2, pp.1498-1506

*A,* Vol. 102, No. 39, pp.13807-13812

16, pp.1695-1699

signaling. *Nat Med,* Vol. 12, No. 6, pp.699-704

**11** 

*Japan*

**The Emergence and Development** 

**of Newborn Infants in a** 

**Neonatal Intensive Care Unit** 

**of Physiological Regulatory Systems** 

Motoki Bonno, Esmot Ara Begum and Hatsumi Yamamoto *Fetal – Neonatal Physiology Research, Clinical Research Institute, National Hosiptal Organization, Miechuo Medical Center, Mie,* 

In the life of a human being, the early neonatal period is the most prone to life-threatening events. After birth, following the transition from the intra- to the extra–uterine environment, babies experience dramatic hemodynamic changes. Once babies begin life in the extrauterine environment, they must regulate their own homeostasis in order to survive. Preterm, low birth weight, newborn are even more vulnerable and require the mechanical support for proper tissue oxygenation and nutrition in order to grow and survive in the extra-uterine environment. The adaptation to extra-uterine life is a slow and difficult process for these

At this critical period, hypothermia, apnoea, respiratory distress, and cardiac instabilities such as bradycardia and hypotension are common features in these newborn babies (Bhatt et al., 2010; Di Fiore et al.,2001; Dransfield et al., 1983; Tirosh et al., 2010; Trevisanuto et al., 2005; Upton, et al., 1992), and the resulting hypoxia may lead to brain damage and cardiac arrest if the medical support of a special incubator equipped with a ventilator and systemic monitoring is not provided a as a "primary life support system". Therefore, the continuous monitoring of heartbeat, respiration, oxygen saturation, blood pressure and temperature has been integrated into the neonatal intensive care unit (NICU) as a mandatory tool to support

Monitoring data collected using non-invasive electrocardiograms have been used to understand the physiological regulatory system of these vulnerable infants. The measurement of heart rate variability has been widely examined using various analytic methods (Ardura et al., 1997; Baldzer et al., 1989; Katona et al.,1980; Patural et al., 2004; Yiallourou et al., 2010). Among them, circadian rhythms have been documented to be a prognostic marker of physiological stability and the maturation of the physiological regulatory system, which can be defined as the long-term regulatory system. The endogenous circadian rhythm is generated by the endogenous biological clock located in the anterior hypothalamic suprachiasmatic nuclei (Panda et al., 2002) and may be modulated by exogenous factors (Reppert & Weaver, 2002). It has been documented in the growing foetus

**1. Introduction** 

babies because of their prematurity.

the fragile clinical conditions of these infants.

### **The Emergence and Development of Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit**

Motoki Bonno, Esmot Ara Begum and Hatsumi Yamamoto *Fetal – Neonatal Physiology Research, Clinical Research Institute, National Hosiptal Organization, Miechuo Medical Center, Mie, Japan*

### **1. Introduction**

In the life of a human being, the early neonatal period is the most prone to life-threatening events. After birth, following the transition from the intra- to the extra–uterine environment, babies experience dramatic hemodynamic changes. Once babies begin life in the extrauterine environment, they must regulate their own homeostasis in order to survive. Preterm, low birth weight, newborn are even more vulnerable and require the mechanical support for proper tissue oxygenation and nutrition in order to grow and survive in the extra-uterine environment. The adaptation to extra-uterine life is a slow and difficult process for these babies because of their prematurity.

At this critical period, hypothermia, apnoea, respiratory distress, and cardiac instabilities such as bradycardia and hypotension are common features in these newborn babies (Bhatt et al., 2010; Di Fiore et al.,2001; Dransfield et al., 1983; Tirosh et al., 2010; Trevisanuto et al., 2005; Upton, et al., 1992), and the resulting hypoxia may lead to brain damage and cardiac arrest if the medical support of a special incubator equipped with a ventilator and systemic monitoring is not provided a as a "primary life support system". Therefore, the continuous monitoring of heartbeat, respiration, oxygen saturation, blood pressure and temperature has been integrated into the neonatal intensive care unit (NICU) as a mandatory tool to support the fragile clinical conditions of these infants.

Monitoring data collected using non-invasive electrocardiograms have been used to understand the physiological regulatory system of these vulnerable infants. The measurement of heart rate variability has been widely examined using various analytic methods (Ardura et al., 1997; Baldzer et al., 1989; Katona et al.,1980; Patural et al., 2004; Yiallourou et al., 2010). Among them, circadian rhythms have been documented to be a prognostic marker of physiological stability and the maturation of the physiological regulatory system, which can be defined as the long-term regulatory system. The endogenous circadian rhythm is generated by the endogenous biological clock located in the anterior hypothalamic suprachiasmatic nuclei (Panda et al., 2002) and may be modulated by exogenous factors (Reppert & Weaver, 2002). It has been documented in the growing foetus

The Emergence and Development of

system of physiological homeostasis.

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 195

rhythmicity of neonate as long-term regulatory system and HRV as short-term regulatory

Fig. 1. NICU LAN system for recording physiological parameters.

linear trend was removed using least square regression methods.

red line is the detected cycle superimposed on the original data.

**2.2.1 Data acquisition and trend removal** 

**2.2 Analysis of the heart rate circadian rhythm as a long-term regulatory system** 

For each patient, the heart rate was continuously recorded for 24 hours for the four postnatal periods (P): P1: 0-3, P2: 4-6, P3: 7-13, and P4: 14-21 postnatal days. Subjects with a continuous disruption in the data of more than 1 minute were excluded from the study. Any

Fig. 2. Determination of the dominant cycle using spectral analysis. *A*: Plot of original data for heart rate (HR). *B*: Periodogram intensities for HR. The largest peak of the periodogram was selected as the cycle component. *C*: The intensity of the cycle corresponding to the largest peak in the periodogram was reconstructed to fit the sinusoidal function. The bold

(Patrick et al., 1982) controlled by maternal circadian signal (Seron-Ferre M et al). In fullterm neonates, circadian rhythm are documented after birth, subsequently disappear and are not detectable within 3 or 4 weeks of postnatal age (Dimitriou et al., 1999; Mirmiran & Kok, 1991). However, in preterm neonates, the development of circadian rhythms is still controversial (Ardura et al., 1997; Begum et al, 2006; Bueno et al., 2001; Dimitriou et al., 1999; Korte et al., 2001; Mirmiran & Kok, 1991; Schimmel et al., 2002; Updike et al., 1985; Weinert et al., 1994) and the clinical relevance remains obscure.

Heart rate variability (HRV), derived from fast Fourier transform analysis of the R to R interval (RR interval) using an electrocardiogram, is another popular method to assess the autonomic nervous system (ANS) (Akselrod et al., 1981; Danguole et al., 2001; Finley et al., 1987; Kamen, 1996; Malik, 1996; Mazurak et al., 2010; Pontet et al., 2003; van Geijn et al., 1980). Continuous changes in sympathetic and parasympathetic neural impulses in the ANS induce changes in heart rate and cause oscillations around the mean, which is the HRV (Malik, 1996). In a frequency-based analysis of HRV, the low frequency (LF) band (0.04- 0.15Hz) derived from the frequency domain analysis expresses the activity of both the sympathetic nervous system (SNS) and parasympathetic nervous system, and the high frequency (HF) band (0.15-0.5Hz) expresses the parasympathetic nervous system (PNS) activity of the ANS (Akselrod et al., 1981). The LF to HF ratio reflects the sympathovagal balances. Reports regarding the relevance of HRV to cardiac-health-related events that contribute to the ANS in the foetus, such as asphyxia (Bocking, 2003), foetal distress (Karin et al., 1993); in infants, such as respiratory distress, apnoea, sepsis, bradycardia (Bennet & Gunn, 2009; Frasch et al., 2009; Li et al., 2005; Logier et al., 2008 ; Sampson et al., 1980), and hypotension (Di Fiore et al., 2001; Dransfield et al., 1983; Fairchild & O'Shea, 2010; Frasch et al., 2009; Upton et al., 1992; Wennergren et al., 1986); and adults, such as stroke and myocardial events (Faber, 1996; Korpelainen et al., 1999; Sosnowski et al., 2002), are numerous and well-documented. In foetuses and preterm infants, the ANS is highly dependent on the SNS while the PNS is immature (Chatow et al., 1995), and the role of the PNS increases with the increase in gestational age (Van Leeuwen et al., 2003). Evidence of the effect of HRV on the adaptation of the ANS in preterm, low birth weight infants during the early neonatal period is not widely available, although it is known that an immature ANS is one of the factors contributing to the vulnerability of these infants.

In this research review, developmental and adaptation changes will be described based on heart rate records from ECG monitors. This description will provide an important perspective on the developmental physiological homeostasis of preterm infants during the early neonatal period in the intensive care unit.

### **2. Utilisation of the electrocardiogram - analytical methods of heart rate assessment**

### **2.1 Monitoring and data recording through NICU local area network (LAN) system**

All infants hospitalised in the NICU are monitored via electrocardiogram (ECG) for heart rate (HR) and respiration rate (RR), with a pulse oxymeter on the wrist or foot for pulse rate (PR) and oxygen saturation by pulse oxymetry (SpO2), and arterial blood pressure (ABP) using a catheter manometer system occasionally. In our study, the monitored physiological information is transformed to measurement variables using the Wave Achieving System (WAS) or Clinical Database Engine (CDE) (Philips Electronics Japan, Tokyo, Japan) through NICU LAN system (Fig. 1) (Begum et al., 2006). In this report, we will explain the circadian

(Patrick et al., 1982) controlled by maternal circadian signal (Seron-Ferre M et al). In fullterm neonates, circadian rhythm are documented after birth, subsequently disappear and are not detectable within 3 or 4 weeks of postnatal age (Dimitriou et al., 1999; Mirmiran & Kok, 1991). However, in preterm neonates, the development of circadian rhythms is still controversial (Ardura et al., 1997; Begum et al, 2006; Bueno et al., 2001; Dimitriou et al., 1999; Korte et al., 2001; Mirmiran & Kok, 1991; Schimmel et al., 2002; Updike et al., 1985; Weinert

Heart rate variability (HRV), derived from fast Fourier transform analysis of the R to R interval (RR interval) using an electrocardiogram, is another popular method to assess the autonomic nervous system (ANS) (Akselrod et al., 1981; Danguole et al., 2001; Finley et al., 1987; Kamen, 1996; Malik, 1996; Mazurak et al., 2010; Pontet et al., 2003; van Geijn et al., 1980). Continuous changes in sympathetic and parasympathetic neural impulses in the ANS induce changes in heart rate and cause oscillations around the mean, which is the HRV (Malik, 1996). In a frequency-based analysis of HRV, the low frequency (LF) band (0.04- 0.15Hz) derived from the frequency domain analysis expresses the activity of both the sympathetic nervous system (SNS) and parasympathetic nervous system, and the high frequency (HF) band (0.15-0.5Hz) expresses the parasympathetic nervous system (PNS) activity of the ANS (Akselrod et al., 1981). The LF to HF ratio reflects the sympathovagal balances. Reports regarding the relevance of HRV to cardiac-health-related events that contribute to the ANS in the foetus, such as asphyxia (Bocking, 2003), foetal distress (Karin et al., 1993); in infants, such as respiratory distress, apnoea, sepsis, bradycardia (Bennet & Gunn, 2009; Frasch et al., 2009; Li et al., 2005; Logier et al., 2008 ; Sampson et al., 1980), and hypotension (Di Fiore et al., 2001; Dransfield et al., 1983; Fairchild & O'Shea, 2010; Frasch et al., 2009; Upton et al., 1992; Wennergren et al., 1986); and adults, such as stroke and myocardial events (Faber, 1996; Korpelainen et al., 1999; Sosnowski et al., 2002), are numerous and well-documented. In foetuses and preterm infants, the ANS is highly dependent on the SNS while the PNS is immature (Chatow et al., 1995), and the role of the PNS increases with the increase in gestational age (Van Leeuwen et al., 2003). Evidence of the effect of HRV on the adaptation of the ANS in preterm, low birth weight infants during the early neonatal period is not widely available, although it is known that an immature

ANS is one of the factors contributing to the vulnerability of these infants.

early neonatal period in the intensive care unit.

**assessment** 

In this research review, developmental and adaptation changes will be described based on heart rate records from ECG monitors. This description will provide an important perspective on the developmental physiological homeostasis of preterm infants during the

**2. Utilisation of the electrocardiogram - analytical methods of heart rate** 

**2.1 Monitoring and data recording through NICU local area network (LAN) system**  All infants hospitalised in the NICU are monitored via electrocardiogram (ECG) for heart rate (HR) and respiration rate (RR), with a pulse oxymeter on the wrist or foot for pulse rate (PR) and oxygen saturation by pulse oxymetry (SpO2), and arterial blood pressure (ABP) using a catheter manometer system occasionally. In our study, the monitored physiological information is transformed to measurement variables using the Wave Achieving System (WAS) or Clinical Database Engine (CDE) (Philips Electronics Japan, Tokyo, Japan) through NICU LAN system (Fig. 1) (Begum et al., 2006). In this report, we will explain the circadian

et al., 1994) and the clinical relevance remains obscure.

rhythmicity of neonate as long-term regulatory system and HRV as short-term regulatory system of physiological homeostasis.

Fig. 1. NICU LAN system for recording physiological parameters.

### **2.2 Analysis of the heart rate circadian rhythm as a long-term regulatory system 2.2.1 Data acquisition and trend removal**

For each patient, the heart rate was continuously recorded for 24 hours for the four postnatal periods (P): P1: 0-3, P2: 4-6, P3: 7-13, and P4: 14-21 postnatal days. Subjects with a continuous disruption in the data of more than 1 minute were excluded from the study. Any linear trend was removed using least square regression methods.

Fig. 2. Determination of the dominant cycle using spectral analysis. *A*: Plot of original data for heart rate (HR). *B*: Periodogram intensities for HR. The largest peak of the periodogram was selected as the cycle component. *C*: The intensity of the cycle corresponding to the largest peak in the periodogram was reconstructed to fit the sinusoidal function. The bold red line is the detected cycle superimposed on the original data.

The Emergence and Development of

**2.3.2.1 Time domain analysis** 

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 197

interval, the non-linear Poincare plot can be analysed (Brennan et al., 2002). In this study, we have used the Heart Rate Variability Software (Kubios HRV, version 2.0) to analyse the time

Time domain measurements can be easily described using the RR interval data. From the time domain analysis, the standard deviation of RR (SDRR), and the root mean square of

For the frequency domain analysis, a spectral analysis of the RR interval was performed using fast Fourier transformation (FFT). The frequency domain analysis is widely used and is well recognised for analysing HRV. The low frequency (LF: 0.04 - 0.15 Hz) band and the

Poincare plots were calculated for the geometric representations of the RR intervals. The non-linear Poincare plot presents a figure with each RR interval plotted against the previous RR interval (RRn+1, RRn) and provides a detail of beat-to-beat information for the total data

Fig. 4. An example Poincare plot: A standard Poincare plot of the R-R intervals of a healthy neonate born with 35 weeks of gestational age (GA). SD1 and SD2 represent the dispersion

The SD1, the width of the Poincare plot (the dispersion of points perpendicular to the line of identity), expresses the level of short-term heart rate variability, and SD2, the length (the dispersion along the line of identity), expresses the level of long-term heart rate variability

(Fig. 4) (Brennan et al., 2001). The equations for SD1 and SD2 are as follows:

1 1

*n n N RR RR*

( ) *<sup>N</sup>*

1 1 2 <sup>1</sup> <sup>1</sup> <sup>1</sup> ( ) *<sup>N</sup> n n n N RR RR*

domain and frequency domain analysis as well as Poincare plot also.

successive differences (RMSSD) were calculated as follows:

The root mean square of successive differences (RMSSD)

**2.3.2.2 Frequency domain analysis of the RR interval** 

high frequency (HF: 0.15 - 0.50 Hz) band were calculated.

along the minor and major axes of the fitted eclipse.

Standard deviation of RR interval (SDRR) ..

**2.3.2.3 Poincare plot of the RR interval** 

set (Brennan et al., 2002).

Standard deviation of RR interval (SDRR) <sup>1</sup> <sup>2</sup>

### **2.2.2 Analysis of circadian rhythm**

Circadian rhythmicity can be analyzed using various methods such as chi square periodogram, cosinor analysis.In our study, the existence of rhythmicity was analysed using power spectral analysis (periodogram) with SPSS 11.5 software (SPSS Inc. Chicago, IL) (Warner, 1998). Briefly, 24-hour sessions in 10-second intervals were run and aggregated into 1-minute time blocks. A periodogram analysis was performed using a time series of 1440 minutes. The Fisher test was used to assess the statistical significance of the cycle components (N = 1440, α = 0.05) (Russell, 1985). In our study, the cycle with the largest peak in the periodogram was considered to be the dominant cycle for each time series among the significant cycles. All dominant cycles were confirmed by Fourier analysis, further circadian cycles were confirmed by cosinor analysis with a significance of p < 0.05 by least squares analysis (Fig. 2) (Nelson, 1993).

### **2.3 Analysis of heart rate variability (HRV) as a short - term regulatory system 2.3.1 Data acquisition of RR interval and the detection of outliers**

To calculate R to R intervals in our study, for each patient, the heart rate wave was digitally transported to the CDE system, and the digital data for the heart rate was exported as a CSV file at a sampling rate of 500Hz. The RR interval was calculated by using MemCalc (GMS, Tokyo, Japan) which is statistical software specialised for entropy-based spectral analysis (Fig. 3).

Fig. 3. Transformation of the RR interval from the waveform of a heart rate signal.

The data set obtained for the RR interval was visually inspected to identify any abnormal R – waves or artefacts. The presence of outliers in the RR interval data set can adversely effect both the time and the frequency domain analyses of HRV. Data outside of three standard deviations from the mean of RR interval data set were assumed to be statistically irrelevant and were tagged as outliers unless they were part of a trend (Niles, 2003). The missing values were replaced by the mean of the total data set.

### **2.3.2 Methods for HRV analysis - time domain analysis, frequency domain analysis and Poincare plot of the RR interval**

Heart rate variability can be analysed in two ways: time domain and frequency domain analyses (Akselrod et al., 1981; Malik, 1996). For a geometric representation of the RR interval, the non-linear Poincare plot can be analysed (Brennan et al., 2002). In this study, we have used the Heart Rate Variability Software (Kubios HRV, version 2.0) to analyse the time domain and frequency domain analysis as well as Poincare plot also.

### **2.3.2.1 Time domain analysis**

196 Advances in Electrocardiograms – Clinical Applications

Circadian rhythmicity can be analyzed using various methods such as chi square periodogram, cosinor analysis.In our study, the existence of rhythmicity was analysed using power spectral analysis (periodogram) with SPSS 11.5 software (SPSS Inc. Chicago, IL) (Warner, 1998). Briefly, 24-hour sessions in 10-second intervals were run and aggregated into 1-minute time blocks. A periodogram analysis was performed using a time series of 1440 minutes. The Fisher test was used to assess the statistical significance of the cycle components (N = 1440, α = 0.05) (Russell, 1985). In our study, the cycle with the largest peak in the periodogram was considered to be the dominant cycle for each time series among the significant cycles. All dominant cycles were confirmed by Fourier analysis, further circadian cycles were confirmed by cosinor analysis with a significance of p < 0.05 by least squares

**2.3 Analysis of heart rate variability (HRV) as a short - term regulatory system** 

Japan) which is statistical software specialised for entropy-based spectral analysis (Fig. 3).

Fig. 3. Transformation of the RR interval from the waveform of a heart rate signal.

values were replaced by the mean of the total data set.

**and Poincare plot of the RR interval** 

The data set obtained for the RR interval was visually inspected to identify any abnormal R – waves or artefacts. The presence of outliers in the RR interval data set can adversely effect both the time and the frequency domain analyses of HRV. Data outside of three standard deviations from the mean of RR interval data set were assumed to be statistically irrelevant and were tagged as outliers unless they were part of a trend (Niles, 2003). The missing

**2.3.2 Methods for HRV analysis - time domain analysis, frequency domain analysis** 

Heart rate variability can be analysed in two ways: time domain and frequency domain analyses (Akselrod et al., 1981; Malik, 1996). For a geometric representation of the RR

To calculate R to R intervals in our study, for each patient, the heart rate wave was digitally transported to the CDE system, and the digital data for the heart rate was exported as a CSV file at a sampling rate of 500Hz. The RR interval was calculated by using MemCalc (GMS, Tokyo,

**2.3.1 Data acquisition of RR interval and the detection of outliers** 

**2.2.2 Analysis of circadian rhythm** 

analysis (Fig. 2) (Nelson, 1993).

Time domain measurements can be easily described using the RR interval data. From the time domain analysis, the standard deviation of RR (SDRR), and the root mean square of successive differences (RMSSD) were calculated as follows:

Standard deviation of RR interval (SDRR) <sup>1</sup> <sup>2</sup> 1 1 ( ) *<sup>N</sup> n n N RR RR*

The root mean square of successive differences (RMSSD)

$$=\sqrt{\frac{1}{N-1}}\sum\_{n=1}^{N-1} \left( RR\_{n+1} - RR\_n \right)^2$$

### **2.3.2.2 Frequency domain analysis of the RR interval**

For the frequency domain analysis, a spectral analysis of the RR interval was performed using fast Fourier transformation (FFT). The frequency domain analysis is widely used and is well recognised for analysing HRV. The low frequency (LF: 0.04 - 0.15 Hz) band and the high frequency (HF: 0.15 - 0.50 Hz) band were calculated.

### **2.3.2.3 Poincare plot of the RR interval**

Poincare plots were calculated for the geometric representations of the RR intervals. The non-linear Poincare plot presents a figure with each RR interval plotted against the previous RR interval (RRn+1, RRn) and provides a detail of beat-to-beat information for the total data set (Brennan et al., 2002).

Fig. 4. An example Poincare plot: A standard Poincare plot of the R-R intervals of a healthy neonate born with 35 weeks of gestational age (GA). SD1 and SD2 represent the dispersion along the minor and major axes of the fitted eclipse.

The SD1, the width of the Poincare plot (the dispersion of points perpendicular to the line of identity), expresses the level of short-term heart rate variability, and SD2, the length (the dispersion along the line of identity), expresses the level of long-term heart rate variability (Fig. 4) (Brennan et al., 2001). The equations for SD1 and SD2 are as follows:

The Emergence and Development of

p <0.05, AGA vs. SGA

rather than its existence.

AGA infants showed in Fig 7(Begum et al., 2010).

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 199

and compared with infants born with average weight for gestational age (AGA: > -1.5 to < 1.5SDs). In the cases for which significant circadian rhythms were observed, significant differences were not observed between SGA and AGA infants, however, the amplitudes of the heart rate circadian rhythms were significantly smaller in SGA infants compared to

Fig. 6. The relationship between rhythmicity and PCA in P1: r = 0.38, P < 0.0001

Fig. 7. The existence and amplitude of circadian rhythms in AGA and SGA neonates. A: Distribution of the infants with significant circadian rhythm and significant differences was not observed between AGA and SGA. B: Distribution of circadian amplitudes for AGA and SGA. The amplitudes were significantly smaller in SGA infants compared to AGA infants. \*

A decreased amplitudes in the circadian rhythm of SGA infants as shown above, intrauterine growth retardation might have influences on the quality of circadian rhythm

Width of RRn vs. RR (n+1) distribution (SD1)2 <sup>1</sup> <sup>2</sup> <sup>2</sup> *SDSD*

Length of RRn vs. RR(n+1) distribution (SD2) 2 2 <sup>1</sup> <sup>2</sup> <sup>2</sup>*SDSD SDSD*

### **3. Evidence of circadian rhythmicity during the early neonatal period with clinical relevance**

### **3.1 The development of a circadian rhythm in the neonatal period**

Infants admitted in the NICU are vulnerable because of their physiological instability, and hypothermia, apnoea, respiratory distress, bradycardia and hypotension are common features for them. Frequent medical examination and therapy are common for them. Within this fragile environment, how do they adapt a physiological homeostasis? In our study, we have observed the circadian rhythmicity of preterm and full-term infants at four postnatal periods to observe their adaptation and developmental processes. The circadian rhythms of the heart rate were analysed in 187 neonates. The median GA was 34 weeks (range: 23–42 weeks). (Begum et al, 2006).

By analysing the distribution of circadian rhythmicity (Fig. 5), circadian rhythms were observed to be dominant among preterm infants (40%, < 28 wks of GA) compared to fullterm infants, and a similar tendency was observed in other periods. These results partially support the previous studies (Dimitriou et al., 1999; Mirmiran & Kok, 1991).

In examining the relationship between circadian rhythmicity and post-conceptional age (PCA), an analysis of the correlation coefficient was performed using the amplitudes of each period. The amplitudes of the heart rate circadian rhythms were positively correlated with PCA in all four periods (Fig. 6).

The higher percentages for the existence of circadian rhythms in preterm infants compared to full-term infants suggest that the maternal influence on circadian rhythm may persist in preterm infants more strongly than in full-term infants during the neonatal period. However, the observed increase in circadian amplitude with PCA implies that the magnitude of circadian rhythmicity parallels the maturation of neonates.

Further, we analysed the heart rate circadian rhythmicity of infants born with small weight for gestational age (SGA: birth weight and height bellow < -2SDs) during 0-2 postnatal days

<sup>2</sup> *SDSD*

**3. Evidence of circadian rhythmicity during the early neonatal period with** 

Infants admitted in the NICU are vulnerable because of their physiological instability, and hypothermia, apnoea, respiratory distress, bradycardia and hypotension are common features for them. Frequent medical examination and therapy are common for them. Within this fragile environment, how do they adapt a physiological homeostasis? In our study, we have observed the circadian rhythmicity of preterm and full-term infants at four postnatal periods to observe their adaptation and developmental processes. The circadian rhythms of the heart rate were analysed in 187 neonates. The median GA was 34 weeks (range: 23–42

By analysing the distribution of circadian rhythmicity (Fig. 5), circadian rhythms were observed to be dominant among preterm infants (40%, < 28 wks of GA) compared to fullterm infants, and a similar tendency was observed in other periods. These results partially

<sup>2</sup> <sup>2</sup>*SDSD SDSD*

Width of RRn vs. RR (n+1) distribution (SD1)2 <sup>1</sup> <sup>2</sup>

**clinical relevance** 

weeks). (Begum et al, 2006).

PCA in all four periods (Fig. 6).

Length of RRn vs. RR(n+1) distribution (SD2) 2 2 <sup>1</sup>

**3.1 The development of a circadian rhythm in the neonatal period** 

support the previous studies (Dimitriou et al., 1999; Mirmiran & Kok, 1991).

Fig. 5. The distribution of circadian cycles according to gestational age groups

magnitude of circadian rhythmicity parallels the maturation of neonates.

In examining the relationship between circadian rhythmicity and post-conceptional age (PCA), an analysis of the correlation coefficient was performed using the amplitudes of each period. The amplitudes of the heart rate circadian rhythms were positively correlated with

The higher percentages for the existence of circadian rhythms in preterm infants compared to full-term infants suggest that the maternal influence on circadian rhythm may persist in preterm infants more strongly than in full-term infants during the neonatal period. However, the observed increase in circadian amplitude with PCA implies that the

Further, we analysed the heart rate circadian rhythmicity of infants born with small weight for gestational age (SGA: birth weight and height bellow < -2SDs) during 0-2 postnatal days and compared with infants born with average weight for gestational age (AGA: > -1.5 to < 1.5SDs). In the cases for which significant circadian rhythms were observed, significant differences were not observed between SGA and AGA infants, however, the amplitudes of the heart rate circadian rhythms were significantly smaller in SGA infants compared to AGA infants showed in Fig 7(Begum et al., 2010).

Fig. 6. The relationship between rhythmicity and PCA in P1: r = 0.38, P < 0.0001

Fig. 7. The existence and amplitude of circadian rhythms in AGA and SGA neonates. A: Distribution of the infants with significant circadian rhythm and significant differences was not observed between AGA and SGA. B: Distribution of circadian amplitudes for AGA and SGA. The amplitudes were significantly smaller in SGA infants compared to AGA infants. \* p <0.05, AGA vs. SGA

A decreased amplitudes in the circadian rhythm of SGA infants as shown above, intrauterine growth retardation might have influences on the quality of circadian rhythm rather than its existence.

The Emergence and Development of

mean birth body weight = 875g).

adaptation capacity and as a sign of maturity.

**Three case analyses for three different gestational ages** 

time after birth (Fig.9).

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 201

The power of LF region and HF region from the frequency domain analysis and SD1 and SD1 from Poincare plot analysis was calculated for each session. The power of both the LF and the HF was significantly lower at 3 hours or 6 hours after birth compared to the highest value at 7-days and the increase of LF and HF with the time after birth may indicate the maturation of the ANS. In the Poincare plot analysis, SD1 and SD2 also increased with the

Fig. 9. Alterations in the parameters from the frequency domain analyses and Poincare plot analysis of HRV through the early neonatal period (n=13, mean gestational age = 27weeks,

The decreased HRV till 12 hours of birth may indicate the loss of variability immediate birth and transition of adaptation period for the short - term regulatory system and the increase of HRV from 24-hour to 7-days after birth might be indicated the increase in the ability of the

The adverse consequences of pain response during the early neonatal period have been documented by numerous study (Grunau et al., 2006), and an emphasis has been placed on pain responses in the NICU (Brown, 2009; Johnston et al., 2010; Stevens et al., 2006, 2010). The most common pain event in NICU is blood procurement including heel lance and venepuncture (Carbajal et al., 2008). Changes in facial expression, sleep states, heart rate and SpO2 are used as indicator of pain response in term infants, not in preterm infants (Stevens et al., 1996). The measures calculated from HRV provide sensitive indices to investigate the response of stressful stimuli. These indices are very helpful to investigate the pain response in preterm infants (Lindh et al., 1999; Morison et al., 2001; Oberlander & Saul, 2002; Padhye et al., 2009). The intensity of pain response in infants has been thought to be related with ANS activity maturation (Grunau et al., 2006). We have experimentally observed three cases with different GA (26, 31, 35 weeks) to analyse the differences in the intensity of the response to the pain stimuli during blood sampling. A heel lance episode was divided into 5 sessions as follows: A: 0-30-second (prior to needle puncture); B: 30-60 second (start of

**4.2 Experimental observation of pain response during the early neonatal period:** 

### **4. Evidence of HRV in the early neonatal period and the clinical relevance**

### **4.1 Experimental observation of alterations in HRV during the hourly neonatal period in extremely premature infants**

The ANS plays an important role in the control of the physiological regulation. However, an immature ANS has been reported in the preterm infants after birth which prolongs to the later life also (De Rogalski Landrot et al., 2007). In preterm infants, particularly in extremely premature infants, the early neonatal period is a life-threatening period when cardiac failure, shock, loss of variability is frequently occurs to them. Numerous studies on ANS have been reported the GA based information with longitudinal evidences (Chatow et al., 1995; Longin et al., 2005, 2006) in preterm and full-term infants. However, the capacity and the recovery of ANS activity of extremely preterm infants immediate birth remain unclear. Thus, experimentally we observed the ANS activity of preterm infants born less than 28 weeks of GA from 3 hours after birth to 7 days to understand the changes into the capacity and recovery of ANS. For a micro-observation, ANS activity was observed at 3-hours, 6 hours, 12-hours, 24-hours, 72-hours and 7-days after birth using a dynamic analytic approach of HRV and observed the alterations in ANS through this critical period.

For analysis of HRV, a 30-minutes ECG signal data set was recorded for the calculation of RR interval from 13 extremely premature infants during a rest time at 3-hours, 6-hours, 12 hours, 24-hours, 72-hours, and 7-days after birth. From time domain analysis of HRV, mean RR, Mean HR, SDNN and RMSSD was calculated shown in Fig. 8. With the increase of time after birth, heart rates were decreased while RR intervals increased and this relation indicates the increase of the ability of the heart to function economically. Further, the increases in SDNN and RMSSD with the increase of the time after birth, might be indicated an increase in the amount of HRV. The increase in variability, which in turn indicates the significant interplay between SNS and PNS of the ANS.

Fig. 8. Alterations in the parameters for the time domain analysis of the RR interval for HRV through the early neonatal period in extremely preterm infants (n=13). The RR interval was obtained for 30 minutes at each time point.

**4. Evidence of HRV in the early neonatal period and the clinical relevance**

approach of HRV and observed the alterations in ANS through this critical period.

significant interplay between SNS and PNS of the ANS.

obtained for 30 minutes at each time point.

For analysis of HRV, a 30-minutes ECG signal data set was recorded for the calculation of RR interval from 13 extremely premature infants during a rest time at 3-hours, 6-hours, 12 hours, 24-hours, 72-hours, and 7-days after birth. From time domain analysis of HRV, mean RR, Mean HR, SDNN and RMSSD was calculated shown in Fig. 8. With the increase of time after birth, heart rates were decreased while RR intervals increased and this relation indicates the increase of the ability of the heart to function economically. Further, the increases in SDNN and RMSSD with the increase of the time after birth, might be indicated an increase in the amount of HRV. The increase in variability, which in turn indicates the

Fig. 8. Alterations in the parameters for the time domain analysis of the RR interval for HRV through the early neonatal period in extremely preterm infants (n=13). The RR interval was

**in extremely premature infants** 

**4.1 Experimental observation of alterations in HRV during the hourly neonatal period** 

The ANS plays an important role in the control of the physiological regulation. However, an immature ANS has been reported in the preterm infants after birth which prolongs to the later life also (De Rogalski Landrot et al., 2007). In preterm infants, particularly in extremely premature infants, the early neonatal period is a life-threatening period when cardiac failure, shock, loss of variability is frequently occurs to them. Numerous studies on ANS have been reported the GA based information with longitudinal evidences (Chatow et al., 1995; Longin et al., 2005, 2006) in preterm and full-term infants. However, the capacity and the recovery of ANS activity of extremely preterm infants immediate birth remain unclear. Thus, experimentally we observed the ANS activity of preterm infants born less than 28 weeks of GA from 3 hours after birth to 7 days to understand the changes into the capacity and recovery of ANS. For a micro-observation, ANS activity was observed at 3-hours, 6 hours, 12-hours, 24-hours, 72-hours and 7-days after birth using a dynamic analytic The power of LF region and HF region from the frequency domain analysis and SD1 and SD1 from Poincare plot analysis was calculated for each session. The power of both the LF and the HF was significantly lower at 3 hours or 6 hours after birth compared to the highest value at 7-days and the increase of LF and HF with the time after birth may indicate the maturation of the ANS. In the Poincare plot analysis, SD1 and SD2 also increased with the time after birth (Fig.9).

Fig. 9. Alterations in the parameters from the frequency domain analyses and Poincare plot analysis of HRV through the early neonatal period (n=13, mean gestational age = 27weeks, mean birth body weight = 875g).

The decreased HRV till 12 hours of birth may indicate the loss of variability immediate birth and transition of adaptation period for the short - term regulatory system and the increase of HRV from 24-hour to 7-days after birth might be indicated the increase in the ability of the adaptation capacity and as a sign of maturity.

### **4.2 Experimental observation of pain response during the early neonatal period: Three case analyses for three different gestational ages**

The adverse consequences of pain response during the early neonatal period have been documented by numerous study (Grunau et al., 2006), and an emphasis has been placed on pain responses in the NICU (Brown, 2009; Johnston et al., 2010; Stevens et al., 2006, 2010). The most common pain event in NICU is blood procurement including heel lance and venepuncture (Carbajal et al., 2008). Changes in facial expression, sleep states, heart rate and SpO2 are used as indicator of pain response in term infants, not in preterm infants (Stevens et al., 1996). The measures calculated from HRV provide sensitive indices to investigate the response of stressful stimuli. These indices are very helpful to investigate the pain response in preterm infants (Lindh et al., 1999; Morison et al., 2001; Oberlander & Saul, 2002; Padhye et al., 2009). The intensity of pain response in infants has been thought to be related with ANS activity maturation (Grunau et al., 2006). We have experimentally observed three cases with different GA (26, 31, 35 weeks) to analyse the differences in the intensity of the response to the pain stimuli during blood sampling. A heel lance episode was divided into 5 sessions as follows: A: 0-30-second (prior to needle puncture); B: 30-60 second (start of

The Emergence and Development of

other infants.

KC (Fig. 13).

ANOVA)

shift was observed for the procedure in a 26-week GA infant.

Feldman & Eidelman, 2003; Messmer et al., 1997).

**4.3 Kangaroo care for preterm infants as physiological relaxation** 

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 203

By Poincare analysis of the RR interval, the response of the painful procedure was well recognized (Fig. 11). The Poincare plot was strongly shifted to the upper right quadrant in the 35-week GA infant, and the shift was small in the 31-week GA infant, while almost no

Based on these findings, we suggest that an increased physiological response to painful events is related to the maturity of the ANS, and late preterm infants may be affected more strongly than extremely premature infants by painful events. Because of the immaturity of the ANS in infants <26 weeks, the response to pain was comparatively lower than that of

Kangaroo care (KC) has been widely practiced for preterm and low birth weight (LBW) infants in the neonatal intensive care unit (NICU). KC is now practiced not only in developing countries, but also in developed countries as developmental care because improved outcome in mortality and morbidity have been reported in infants who undergo KC. There have been many studies performed to evaluate the psychological and physiological responses during KC in preterm infants. Although still there is some contradiction on the influences of KC in infants, positive influences also reported by many literature (Bauer et al., 1997; Fohe et al., 2000; Acolet et al., 1989; Cattaneo et al., 1998;

In our study on the physiological responses during kangaroo care, significant differences was observed in the spectral power of heart rate within the different sessions (the %LF was significantly increased during KC, while %HF was decreased, without significant changes in the ratio of LF/HF, Fig. 12). Behavioural states were observed during kangaroo care using the Brazelton Neonatal Behaviour Assessment Scale (Brazelton, 1984). The percentage of infants with quiet sleep states remarkably increased during KC compared to those before KC. This percentage increased further at the end of KC and decreased again 30 minutes after

Fig. 12. Changes in heart rate variability during KC. A: mean difference in heart rate; B: power spectral density of LF and HF; C: ratio of LF/HF. \*P < 0.05 (Repeated measures

needle puncture); C: 60-90 second (squeezing for blood collection); D: 90-120 second (30 second after the end of blood collection); E: 120-150 second (after 60 second of blood collection). The RR interval data were analysed separately for each session.

Frequency domain analyses and Poincare plot analyses were applied to interpret the intensity of the response to pain. The distribution of the Fourier plot and Poincare plot is shown in Fig 10 and Fig. 11. The total spectral power increased after the pain procedure in all cases; however, it was low in the case of a 26-week GA neonate. The spectral power of the LF quickly increased following the needle puncture and gradually decreased with the cessation of the procedure in a 35-week GA infant. In contrast, the power of the HF was delayed in a 31-week GA infant and was low in a 26-week GA infant (Fig10).

Fig. 10. A Spectral plot on the intensity of pain response during blood sampling for three different GA during early neonatal period. A: 0-30-seconds (prior to needle puncture); B: 30- 60 second (start of needle puncture); C: 60-90 seconds (squeezing for blood sampling); D: 90- 120 seconds (30 seconds after the end of blood sampling); E: 120-150 second (after 60 seconds of blood sampling).

Fig. 11. Poincare plot analysis of the RR interval during blood sampling. A: 0-30 second (prior to needle puncture); B: 30-60 second (start of needle puncture); C: 60- 90 second (squeezing for blood sampling); D: 90-120 second (30 seconds after the end of blood sampling); E: 120-150 second (after 60 seconds of blood sampling).

needle puncture); C: 60-90 second (squeezing for blood collection); D: 90-120 second (30 second after the end of blood collection); E: 120-150 second (after 60 second of blood

Frequency domain analyses and Poincare plot analyses were applied to interpret the intensity of the response to pain. The distribution of the Fourier plot and Poincare plot is shown in Fig 10 and Fig. 11. The total spectral power increased after the pain procedure in all cases; however, it was low in the case of a 26-week GA neonate. The spectral power of the LF quickly increased following the needle puncture and gradually decreased with the cessation of the procedure in a 35-week GA infant. In contrast, the power of the HF was

Fig. 10. A Spectral plot on the intensity of pain response during blood sampling for three different GA during early neonatal period. A: 0-30-seconds (prior to needle puncture); B: 30- 60 second (start of needle puncture); C: 60-90 seconds (squeezing for blood sampling); D: 90- 120 seconds (30 seconds after the end of blood sampling); E: 120-150 second (after 60

Fig. 11. Poincare plot analysis of the RR interval during blood sampling. A: 0-30 second (prior to needle puncture); B: 30-60 second (start of needle puncture); C: 60- 90 second (squeezing for blood sampling); D: 90-120 second (30 seconds after the end of blood

sampling); E: 120-150 second (after 60 seconds of blood sampling).

seconds of blood sampling).

collection). The RR interval data were analysed separately for each session.

delayed in a 31-week GA infant and was low in a 26-week GA infant (Fig10).

By Poincare analysis of the RR interval, the response of the painful procedure was well recognized (Fig. 11). The Poincare plot was strongly shifted to the upper right quadrant in the 35-week GA infant, and the shift was small in the 31-week GA infant, while almost no shift was observed for the procedure in a 26-week GA infant.

Based on these findings, we suggest that an increased physiological response to painful events is related to the maturity of the ANS, and late preterm infants may be affected more strongly than extremely premature infants by painful events. Because of the immaturity of the ANS in infants <26 weeks, the response to pain was comparatively lower than that of other infants.

### **4.3 Kangaroo care for preterm infants as physiological relaxation**

Kangaroo care (KC) has been widely practiced for preterm and low birth weight (LBW) infants in the neonatal intensive care unit (NICU). KC is now practiced not only in developing countries, but also in developed countries as developmental care because improved outcome in mortality and morbidity have been reported in infants who undergo KC. There have been many studies performed to evaluate the psychological and physiological responses during KC in preterm infants. Although still there is some contradiction on the influences of KC in infants, positive influences also reported by many literature (Bauer et al., 1997; Fohe et al., 2000; Acolet et al., 1989; Cattaneo et al., 1998; Feldman & Eidelman, 2003; Messmer et al., 1997).

In our study on the physiological responses during kangaroo care, significant differences was observed in the spectral power of heart rate within the different sessions (the %LF was significantly increased during KC, while %HF was decreased, without significant changes in the ratio of LF/HF, Fig. 12). Behavioural states were observed during kangaroo care using the Brazelton Neonatal Behaviour Assessment Scale (Brazelton, 1984). The percentage of infants with quiet sleep states remarkably increased during KC compared to those before KC. This percentage increased further at the end of KC and decreased again 30 minutes after KC (Fig. 13).

Fig. 12. Changes in heart rate variability during KC. A: mean difference in heart rate; B: power spectral density of LF and HF; C: ratio of LF/HF. \*P < 0.05 (Repeated measures ANOVA)

The Emergence and Development of

Nakano for secretarial assistance.

**6. Acknowledgment** 

**7. References** 

8075

77-82.

*Heinemann*.

April, 2010

(2003/09), ISSN: 0301-2115,

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 205

We thank the staff of the Department of Nursing and the Department of Paediatrics and Neonatology for their assistance in the data collection in the NICU; we also thank Ms Taeko

Ardura J, Andres J, Revilla M, Aragon M (1997): Heart rate biorhythm changes during the

Acolet D, Sleath K, Whitelaw A (1989): Oxygenation, heart rate and temperature in very low

Akselrod S, Gordon D, Ubel FA, Shannon DC, Berger AC, Cohen RJ (1981): Power

Bueno C, Diambra L, Menna-Barreto L (2001): Sleep-Wake and Temperature Rhythms in

Bocking AD (2003): Assessment of fetal heart rate and fetal movements in detecting oxygen

Brazelton TB (1984): Neonatal Behavioral Assessment Scale 2nd Edition. *London:* 

Brennan M, Palaniswami M, Kamen P (2002): Poincare plot interpretation using a

Brennan M, Palaniswami M, Kamen P (2001): Do existing measures of Poincare plot

Bennet L, Gunn AJ (2009): The fetal heart rate response to hypoxia: insights from animal models. *Clin Perinatol*, 36(3). PP. 655-672., (2009/09), ISSN: 1557-9840 Begum E, Bonno M, Obata M, Yamamoto H, Kawai M, Komada Y (2006): Emergence of

Begum E, Bonno M, Omori Y, Sugino N, Sasaki N, Yamamoto H (2010): Blunted Circadian

Begum EA, Bonno M, Ohtani N, Yamashita S, Tanaka S, Yamamoto H, Kawai M, Komada Y

Bauer K, Uhrig C, Sperling P, Pasel K, Wieland C, Versmold HT (1997): Body temperatures

*Circ Physiol*, 283(5). PP. H1873-1886., (2002/11), ISSN: 0363-6135

care unit. *J Circadian Rhythms*, 4:11. (2006/09), ISSN: 1740-3391

infants. *BMC Pediatr* , 8:51, (2008/11), ISSN: 1471-2431

48(11). PP. 1342-1347. (2001/11), ISSN: 0018-9294

birthweight infants during skin-to-skin contact with their mothers. *Acta Paediatr* 

spectrum analysis of heart rate fluctuation: a quantitative probe of beat-to-beat cardiovascular control. *Science*, 213(4504). PP. 220-222., (1981/07), ISSN: 0036-

Preterm Babies Maintained in a Neonatal Care Unit. *Sleep Research Online*, 4(3). PP.

deprivation in-utero. *Eur J Obstet Gynecol Reprod Biol*, 110 Suppl 1. PP. S108-112.

physiological model of HRV based on a network of oscillators. *Am J Physiol Heart* 

geometry reflect nonlinear features of heart rate variability? *IEEE Trans Biomed Eng*,

physiological rhythmicity in term and preterm neonates in a neonatal intensive

Rhythms in Heart Rate of Small for Gestational Age Infants in Early Neonatal Period, Conference: Asian Society for Prdiatric Research (ASPR). *Taipei, Taiwan*,

(2008): Cerebral oxygenation responses during kangaroo care in low birth weight

and oxygen consumption during skin-to-skin (kangaroo) care in stable preterm

first three months of life. *Biol Neonate*, 72, PP. 94-101.

*Scand*, 78(2). PP. 189-193. (1989/03), ISSN: 0001-656X

Fig. 13. Changes in the behavioural states of infants during KC. Five different behavioural states were observed: 1) Quiet sleep; 2) Active sleep; 3) Drowsiness; 4) Alert inactivity; and 5) Active awake.

As shown above, an increase in %LF during KC was observed without fluctuation of the LF/HF ratio, and infants tended to sleep deeply during KC. These findings suggest that parasympathetic nerve activity may be suppressed during KC and that decreased parasympathetic nerve activity may induce deep sleep. Improved outcome and health status with KC may be due to relaxation of vagal tone.

### **5. Conclusion**

In this chapter, we explained the emergence and development of the ANS as physiological regulatory systems of new- born infants. Circadian rhythm, we defined it as long-term regulatory system, exist immidiate birth and persist through the neonatal period in preterm infants, while it does not persist after birth in full-term infants. On the contrary, the heart rate variability of extremely premature infants is suppressed on the day of birth and increases with time in the early neonatal period. The maternal influence on circadian rhythms may be important for extremely premature infants because the autonomic nervous system is not fully developed in these infants, as has been shown in the session on physiological response to pain.

In infants with later gestational ages, however, the physiological response to pain is stronger. During hospitalisation, infants with later gestational ages may be more sensitive to stressful stimuli. Developmental support such as Kangaroo care may offer relief from the stressful stimuli of the NICU, and relaxation during their stay in the NICU may be an important aspect for improved outcomes and later health status.

Finally, understanding the development of auto-regulation in newborn infants during the early neonatal period provides new information on the factors that influence the vulnerability of newborn infants in the NICU. Bedside monitoring of physiological hemodynamics and the implications of these monitored variables are key tools for neonatologists, not only for improving survival, but also for improving the quality of life for these vulnerable infants.

### **6. Acknowledgment**

We thank the staff of the Department of Nursing and the Department of Paediatrics and Neonatology for their assistance in the data collection in the NICU; we also thank Ms Taeko Nakano for secretarial assistance.

### **7. References**

204 Advances in Electrocardiograms – Clinical Applications

Fig. 13. Changes in the behavioural states of infants during KC. Five different behavioural states were observed: 1) Quiet sleep; 2) Active sleep; 3) Drowsiness; 4) Alert inactivity; and

As shown above, an increase in %LF during KC was observed without fluctuation of the LF/HF ratio, and infants tended to sleep deeply during KC. These findings suggest that parasympathetic nerve activity may be suppressed during KC and that decreased parasympathetic nerve activity may induce deep sleep. Improved outcome and health status

In this chapter, we explained the emergence and development of the ANS as physiological regulatory systems of new- born infants. Circadian rhythm, we defined it as long-term regulatory system, exist immidiate birth and persist through the neonatal period in preterm infants, while it does not persist after birth in full-term infants. On the contrary, the heart rate variability of extremely premature infants is suppressed on the day of birth and increases with time in the early neonatal period. The maternal influence on circadian rhythms may be important for extremely premature infants because the autonomic nervous system is not fully developed in these infants, as has been shown in the session on

In infants with later gestational ages, however, the physiological response to pain is stronger. During hospitalisation, infants with later gestational ages may be more sensitive to stressful stimuli. Developmental support such as Kangaroo care may offer relief from the stressful stimuli of the NICU, and relaxation during their stay in the NICU may be an

Finally, understanding the development of auto-regulation in newborn infants during the early neonatal period provides new information on the factors that influence the vulnerability of newborn infants in the NICU. Bedside monitoring of physiological hemodynamics and the implications of these monitored variables are key tools for neonatologists, not only for improving survival, but also for improving the quality of life for

important aspect for improved outcomes and later health status.

5) Active awake.

**5. Conclusion** 

physiological response to pain.

these vulnerable infants.

with KC may be due to relaxation of vagal tone.


The Emergence and Development of

9840

0147-8389

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 207

Fairchild KD, O'Shea TM (2010): Heart rate characteristics: physiomarkers for detection of

Faber TS, Staunton A, Hnatkova K, Camm AJ, Malik M (1996): Stepwise strategy of using

Fohe K, Kropf S, Avenarius S (2000): Skin-to-skin contact improves gas exchange in premature infants. *J Perinatol*, 20(5). PP. 311-315. (2000/08), ISSN: 0743-8346 Grunau RE, Whitfield MF, Fay T, Holsti L, Oberlander T, Rogers ML (2006): Biobehavioural

Johnston CC, Fernandes AM, Campbell-Yeo M (2010): Pain in neonates is different. *Pain*,

Korpelainen JT, Sotaniemi KA, Myllyla VV(1999): Autonomic nervous system disorders in

Kamen P (1996): Heart rate variability. *Aust Fam Physician*, 25(7). PP. 1087-1089, 1091-1085.

Karin J, Hirsch M, Akselrod S (1993): An estimate of fetal autonomic state by spectral

Korpelainen JT, Sotaniemi KA, Makikallio A, Huikuri HV, Myllyla VV (1999): Dynamic

Korte J, Wulff K, Oppe C, Siegmund R (2001): Ultradian and circadian activity-rest rhythms

Logier R, De Jonckheere J, Jeanne M, Matis R (2008): Fetal distress diagnosis using heart rate

Li X, Zheng D, Zhou S, Tang D, Wang C, Wu G (2005): Approximate entropy of fetal heart

Lindh V, Wiklund U, Hakansson S (1999): Heel lancing in term new-born infants: an

Longin E, Schaible T, Lenz T, Konig S (2005): Short term heart rate variability in healthy

Longin E, Gerstner T, Schaible T, Lenz T, Konig S (2006): Maturation of the autonomic

*Eng Med Biol Soc*. PP. 4728-4731. (2009/01, EPUB), ISSN: 1557-170X

*Obstet Gynecol Scand*, 84(9). PP. 837-843. (2005/09), ISSN: 0001-6349

*Chronobiol Int*, 18(4). PP. 697-708. (2001/7), ISSN: 0742-0528

80(1-2). PP. 143-148. (1999/03), ISSN: 0304-3959

*Perinat Med*, 34(4). PP. 303-308. (2006/07), ISSN: 0300-5577

663-671. (2005/08), ISSN: 0378-3782

analysis of fetal heart rate fluctuations. *Pediatr Res*, 34(2). PP. 134-138. (1993/08).

behavior of heart rate in ischemic stroke. *Stroke*, 30(5). PP. 1008-1013. (1999/05),

of preterm neonates compared to full-term neonates using actigraphic monitoring.

variability analysis: design of a high frequency variability index. *Conf Proc IEEE* 

rate variability as a predictor of fetal distress in women at term pregnancy. *Acta* 

evaluation of pain by frequency domain analysis of heart rate variability. *Pain*,

neonates: normative data and physiological observations. *Early Hum Dev*, 81(8). PP.

nervous system: differences in heart rate variability in premature vs. term infants. *J* 

*Med Child Neurol*, 48(6). PP. 471-476. (2006/06), ISSN: 0012-1622

stroke. *Clin Auton Res*, 9(6). PP. 325-333. (1999/12), ISSN: 0959-9851

152(3 Suppl). PP. S65-73. (2010/03), ISSN: 1872-6623

(1996/07), ISSN: 0300-8495

ISSN: 0031-3998

ISSN: 0039-2499

late-onset neonatal sepsis. *Clin Perinatol*, 37(3). PP. 581-598. (2010/09), ISSN: 1557-

short- and long-term heart rate variability for risk stratification after myocardial infarction. *Pacing Clin Electrophysiol*, 19(11 Pt 2). PP. 1845-1851. (1996/11), ISSN:

reactivity to pain in preterm infants: a marker of neuromotor development. *Dev* 

infants weighing less than 1500 grams. *J Pediatr*, 130(2). PP. 240-244. (1997/02), ISSN: 0022-3476


Bhatt DR, White R, Martin G, Van Marter LJ, Finer N, Goldsmith JP, Ramos C, Kukreja S,

Baldzer K, Dykes FD, Jones SA, Brogan M, Carrigan TA, Giddens DP (1989): Heart rate

Brown G (2009): NICU noise and the preterm infant. *Neonatal Netw*, 28(3).PP. 165-173.

Carbajal R, Rousset A, Danan C, Coquery S, Nolent P, Ducrocq S, Saizou C, Lapillonne A,

Cattaneo A, Davanzo R, Bergman N, Charpak N (1998): Kangaroo mother care in low-

Chatow U, Davidson S, Reichman BL, Akselrod S (1995): Development and maturation of

Dransfield DA, Spitzer AR, Fox WW (1983): Episodic airway obstruction in premature infants. *Am J Dis Child*, 137(5). PP. 441-443. (1983/05), ISSN: 0002-922X Dimitriou G, Greenough A, Kavvadia V, Mantagos S (1999) : Blood pressure rhythms during

De Rogalski Landrot I, Roche F, Pichot V, Teyssier G, Gaspoz JM, Barthelemy JC, Patural H

Di Fiore JM, Arko MK, Miller MJ, Krauss A, Betkerur A, Zadell A, Kenney SR, Martin RJ

Finley JP, Nugent ST, Hellenbrand W (1987): Heart-rate variability in children. Spectral

Frasch MG, Muller T, Weiss C, Schwab K, Schubert H, Schwab M (2009): Heart rate

Feldman R, Eidelman AI (2003): Skin-to-skin contact (Kangaroo Care) accelerates autonomic

studies. *Pediatrics*, 108(6). PP. 1304-1308. (2001/12), ISSN: 1098-4275

*Care*, 10(5 Suppl). PP. S15-17. (2010/10) ISSN: 1536-0911

44(5). PP. 279-282. (1998/10), ISSN: 0142-6338

*Hum Dev*, 56(1). PP. 49-56. (1999/9), ISSN: 0378-3782

65(10). PP. 2048-2052. (1987/10), ISSN: 0008-4212

PP. 274-281. (2003/04), ISSN: 0012-1622 (Print)

PP. 509-517. (2009/05), ISSN: 1933-7205

ISSN: 0022-3476

(2009/05), ISSN: 1539-2880

ISSN: 1538-3598

0031-3998

1566-0702

3998

infants weighing less than 1500 grams. *J Pediatr*, 130(2). PP. 240-244. (1997/02),

Ramanathan R (2010): Transitional hypothermia in preterm newborns. *Adv Neonatal* 

variability analysis in full-term infants: spectral indices for study of neonatal cardiorespiratory control. *Pediatr Res*, 26(3). PP. 188-195. (1989/09), ISSN: 0031-

Granier M, Durand P *et al* (2008): Epidemiology and treatment of painful procedures in neonates in intensive care units. *JAMA*, 300(1). PP. 60-70. (2008/07),

income countries. International Network in Kangaroo Mother Care. *J Trop Pediatr*,

the autonomic nervous system in premature and full-term infants using spectral analysis of heart rate fluctuations. *Pediatr Res*, 37(3). PP. 294-302. (1995/03), ISSN:

the perinatal period in very immature, extremely low birthweight neonates. *Early* 

(2007): Autonomic nervous system activity in premature and full-term infants from theoretical term to 7 years. *Auton Neurosci*, 136(1-2). PP. 105-109. (2007/10), ISSN:

(2001): Cardiorespiratory events in preterm infants referred for apnea monitoring

analysis of developmental changes between 5 and 24 years. *Can J Physiol Pharmacol*,

variability analysis allows early asphyxia detection in ovine fetus. *Reprod Sci* , 16(5).

and neurobehavioural maturation in preterm infants. *Dev Med Child Neurol,* 45(4).


The Emergence and Development of

ISSN: 1471-2431

*Res*, 33. PP. 83-112.

258. ( 2002/12), ISSN: 0167-5273

(1980/10), ISSN: 0002-9378

(1986/07), ISSN: 0029-7844

*London . PP.* 49-96.

0528

3782

reply 717. (2005/11), ISSN: 0022-3476

0749-8047

1619-1560

Physiological Regulatory Systems of Newborn Infants in a Neonatal Intensive Care Unit 209

pain in neonates. *Eur J Pain*, 14(7). PP. 735-741. (2010/08), ISSN: 1532-2149 Stevens B, McGrath P, Yamada J, Gibbins S, Beyene J, Breau L, Camfield C, Finley A, Franck

Stevens B, Johnston C, Petryshen P, Taddio A (1996): Premature Infant Pain Profile:

Schimmel M, Waterhouse J, Marques M, Weunert D (2002): Circadian and Ultradian

Sosnowski M, MacFarlane PW, Czyz Z, Skrzypek-Wanha J, Boczkowska-Gaik E, Tendera M

Trevisanuto D, Doglioni N, Ferrarese P, Zanardo V (2005): Thermal management of the

Tirosh E, Ariov-Antebi N, Cohen A (2010): Autonomic function, gastroesophageal reflux in

Updike PA, Accurso FJ, Jones RH (1985): Physiologic circadian rhythmicity in preterm

Upton CJ, Milner AD, Stokes GM (1992): Episodic bradycardia in preterm infants. *Arch Dis* 

Van Leeuwen P, Geue D, Lange S, Hatzmann W, Gronemeyer D (2003): Changes in the

van Geijn HP, Jongsma HW, de Haan J, Eskes TK (1980): Analysis of heart rate and beat-to-

Wennergren M, Krantz M, Hjalmarson O, Karlsson K (1986): Fetal heart rate pattern and risk

Warner RM (1998): Spectral Analysis of Time - Series Data. *The Guilford Press, New York,* 

Weinert D, Sitka U, Minors DS, Waterhouse JM (2005): The development of circadian

Weinert D, Sitka U, Minors DS, Waterhouse JM(1994): The development of circadian

infants. *Nurs Res* , 34(3). PP. 160-163. (1985/05), ISSN: 0029-6562

*Child* 1992, 67(7 Spec No). PP. 831-834., (1992/07), ISSN: 1468-2044

*Diagn*, 23(11). PP. 909-916. (2003/11) ISSN: 0197-3851

and procedure invasiveness on health professionals' management of procedural

L, Howlett A *et al (2006)*: Identification of pain indicators for infants at risk for neurological impairment: a Delphi consensus study. *BMC Pediatr*, 6(1). (2006/02),

development and initial validation. *Clin J Pain*, 12(1). PP. 13-22. (1996/03), ISSN:

Rhythmicity in Very Premature Neonates Maintained in Incubators. *Biol Rhythm* 

(2002): Age-adjustment of HRV measures and its prognostic value for risk assessment in patients late after myocardial infarction. *Int J Cardiol*, 86(2-3). PP. 249-

extremely low birth weight infants at birth. *J Pediatr*, 147(5). PP. 716-717; author

apparent life threatening event. *Clin Auton Res*, 20(3). PP. 161-166. (2010/06), ISSN:

frequency power spectrum of fetal heart rate in the course of pregnancy. *Prenat* 

beat variability: Interval difference index. *Am J Obstet Gynecol*, 138(3). PP. 246-252.

for respiratory disturbance in full-term newborns. *Obstet Gynecol*, 68(1). PP. 49-53.

rhythmicity in neonates. *Early Hum Dev*, 36(2). PP. 117-126. (2005/07), ISSN: 0742-

rhythmicity in neonates. *Early Hum Dev*, 36(2). PP. 117-126. (1994/02), ISSN: 0378-


Morison SJ, Grunau RE, Oberlander TF, Whitfield MF (2001): Relations between behavioral

Mirmiran M, Kok JH (1991): Circadian rhythms in early human development. *Early Hum* 

Malik M (1996): Heart rate variability. Standards of measurement, physiological

Mazurak N, Enck P, Muth E, Teufel M, Zipfel S (2010): Heart rate variability as a measure of

Messmer PR, Rodriguez S, Adams J, Wells-Gentry J, Washburn K, Zabaleta I, Abreu S

Nelson W, Tong YL, Lee JK, Halberg F (1979): Methods for cosinor rhythmometry.

Oberlander T, Saul JP (2002): Methodological considerations for the use of heart rate

Ozdemir OM, Ergin H, Sahiner T (2009): Electrophysiological assessment of the brain function in term SGA infants. *Brain Res*, 1270:33-38. (2009/05), ISSN: 1872-6240 Pontet J, Contreras P, Curbelo A, Medina J, Noveri S, Bentancourt S, Migliaro ER (2003):

septic patients. *J Crit Care* , 18(3). PP. 156-163. (2003/9), ISSN: 0883-9441 Padhye NS, Williams AL, Khattak AZ, Lasky RE (2009): Heart rate variability in response to

Panda S, Hogenesch JB, Kay SA (2002): Circadian rhythms from flies to human. *Nature*,

Patrick J, Campbell K, Carmichael L, Probert C (1982): Influence of maternal heart rate and

Patural H, Pichot V, Jaziri F, Teyssier G, Gaspoz JM, Roche F, Barthelemy JC (2008):

Russell R (1985): Significance table for the result of first Fourier transformations. *British J of* 

Reppert SM, Weaver DR (2002): Coordination of circadian timing in mammals. *Nature*,

Stevens B, McGrath P, Ballantyne M, Yamada J, Dupuis A, Gibbins S, Franck L, Allen Finley

use. *Eur Heart J*, 17(3). PP. 354-381. (1996/03), ISSN: 0195-668X

*Chronobiologia* , 6. PP. 305-323. (1979/10), ISSN: 0390-0037

*Psychobiol*, 51(8). PP. 638-649. (2009/12), ISSN: 1098-2302

*Obstet Gynecol*, 144(5). PP. 533-538. (1982/11), ISSN: 0002-9378

*Early Hum Dev*, 84(10). PP. 681-687. (2008/10), ISSN: 0378-3782

*Mathematical and statistical psychology*, 38. PP. 116-119.

418(6901). PP. 935-941. (2002/08), ISSN:0028-0836

417(6886). PP. 329-335. (2002/05), ISSN: 0028-0836

29(3). PP. 427-443. (2002/09), ISSN: 0095-5108

Niles R: http://www.robertniles.com/stats/stdev.shtml,Accessed 5/4/2003. 2003.

17(4). PP. 350-358. (2001/12), ISSN: 0749-8047

*Disord Rev* . (2010/12), ISSN: 1099-0968

408-414. (1997/07), ISSN: 0097-9805

*Dev*, 26(2). PP. 121-128. (1991/08), ISSN: 0378-3782

and cardiac autonomic reactivity to acute pain in preterm neonates. *Clin J Pain*,

interpretation, and clinical use. Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology Heart rate variability. Standards of measurement, physiological interpretation, and clinical

cardiac autonomic function in anorexia nervosa: A review of the literature. *Eur Eat* 

(1997): Effect of kangaroo care on sleep time for neonates. *Pediatr Nurs*, 23(4). PP.

variability as a measure of pain reactivity in vulnerable infants. *Clin Perinatol* ,

Heart rate variability as early marker of multiple organ dysfunction syndrome in

pain stimulus in VLBW infants followed longitudinally during NICU stay. *Dev* 

gross fetal body movements on the daily pattern of fetal heart rate near term. *Am J* 

Autonomic cardiac control of very preterm newborns: a prolonged dysfunction.

G, Howlett A, Johnston C *et al (2010)*: Influence of risk of neurological impairment

and procedure invasiveness on health professionals' management of procedural pain in neonates. *Eur J Pain*, 14(7). PP. 735-741. (2010/08), ISSN: 1532-2149


**12** 

*Turkey* 

**Automated Detection and Classification** 

**Electroencephalogram (EEG) Features** 

*2Baskent University, Electrical Electronics Engineering Department,* 

 *4GATA, Gulhane Military Medicine Academy, Biomedical Engineering Center,* 

Sleep, which is defined as a passive period in organic physiology until the mid-20th century, is accepted to be an indispensable period of life cycle with today's technological advances. While wakefulness is associated with the active excitation of Central Nervous System (CNS), sleep has been recognized as a passive period by the elimination of excitation. However, recent studies have shown that sleep is independent of wakefulness, generated by a sequence of changes in CNS, and a combination of five periods with clear boundaries. Sleep is not the disruption of daily life for a period of time or a waste of time. It is an active period which is important to renew our mental and physical health everyday and is covering one– third of our lives. Sleep activity is important for resting during the working period of basal

The advances in technology enabling the measurement and quantification of brain activity make possible the micro and macro analysis of brain during both sleep and wakefulness states. With the studies investigating the CNS, it is observed the existence of some centrals causing the sleep by inhibiting the other regions of brain. As a result, sleep, which is an active and other state of consciousness, is a brain state of high coordination (Erdamar,

Since breathing is established autonomously during sleep, it is affected by many anatomical and physiological parameters. Depending on this situation, various sleep disorders occur. There are more than eighty known sleeping diseases. Most of them cause person's health to deteriorate and a decrease in life quality. As a result of the research carried out for many years, a list of sleep disorders, which are generally occurring, can be seen as in Table 1. Sleep

disorders can be examined in two classes, parasomnia and dissomnia.

**1. Introduction** 

**1.1 Sleep and sleep disorders** 

metabolism of human body.

2007).

**of Sleep Apnea Types Using** 

**Electrocardiogram (ECG) and** 

Onur Kocak1, Tuncay Bayrak1, Aykut Erdamar1, Levent Ozparlak2, Ziya Telatar3 and Osman Erogul4 *1Baskent University, Biomedical Engineering Department,* 

*3Ankara University, Electronics Engineering Department,* 

Yiallourou SR, Sands SA, Walker AM, Horne RS (2010): Postnatal development of baroreflex sensitivity in infancy. *J Physiol*, 588(Pt 12). PP. 2193-2203. (2010/06), ISSN: 1469-7793

## **Automated Detection and Classification of Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features**

Onur Kocak1, Tuncay Bayrak1, Aykut Erdamar1, Levent Ozparlak2, Ziya Telatar3 and Osman Erogul4 *1Baskent University, Biomedical Engineering Department, 2Baskent University, Electrical Electronics Engineering Department, 3Ankara University, Electronics Engineering Department, 4GATA, Gulhane Military Medicine Academy, Biomedical Engineering Center, Turkey* 

### **1. Introduction**

210 Advances in Electrocardiograms – Clinical Applications

Yiallourou SR, Sands SA, Walker AM, Horne RS (2010): Postnatal development of

ISSN: 1469-7793

baroreflex sensitivity in infancy. *J Physiol*, 588(Pt 12). PP. 2193-2203. (2010/06),

### **1.1 Sleep and sleep disorders**

Sleep, which is defined as a passive period in organic physiology until the mid-20th century, is accepted to be an indispensable period of life cycle with today's technological advances. While wakefulness is associated with the active excitation of Central Nervous System (CNS), sleep has been recognized as a passive period by the elimination of excitation. However, recent studies have shown that sleep is independent of wakefulness, generated by a sequence of changes in CNS, and a combination of five periods with clear boundaries. Sleep is not the disruption of daily life for a period of time or a waste of time. It is an active period which is important to renew our mental and physical health everyday and is covering one– third of our lives. Sleep activity is important for resting during the working period of basal metabolism of human body.

The advances in technology enabling the measurement and quantification of brain activity make possible the micro and macro analysis of brain during both sleep and wakefulness states. With the studies investigating the CNS, it is observed the existence of some centrals causing the sleep by inhibiting the other regions of brain. As a result, sleep, which is an active and other state of consciousness, is a brain state of high coordination (Erdamar, 2007).

Since breathing is established autonomously during sleep, it is affected by many anatomical and physiological parameters. Depending on this situation, various sleep disorders occur. There are more than eighty known sleeping diseases. Most of them cause person's health to deteriorate and a decrease in life quality. As a result of the research carried out for many years, a list of sleep disorders, which are generally occurring, can be seen as in Table 1. Sleep disorders can be examined in two classes, parasomnia and dissomnia.

Automated Detection and Classification of

post-menopausal women (Firat, 2003).

**1.3 Sleep apnea and its types** 

parameters related to these types can be defined as follows.

intense activity in the chest and abdomen is observed.

mixed apneas are discussed as obstructive apneas.

**1.4 Sleep respiratory parameters** 

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 213

In 1997, ASDA (American Sleep Disorder Association) has defined obstructive sleep apnea syndrome as "A syndrome characterized by recurrent obstructions in upper respiratory tract (URT) during sleep and seen often with a decrease in oxygen saturation". The prevalence of the disease is 1-5%. Even generally knowing the risk factors during the beginning of this disease, which has no less than prevalence of Diabetes (Diabetes Mellitus) and Bronchial Asthma, the physiopathology of the obstructions are not totally explained. By a general evaluation, the stopping of breathing during sleep for at least 10 seconds is defined as '*sleep apnea*'. Due to stopping of breathing during sleep, sleep qualities of such patients are disturbed since they often wake up at night. Additionally, they become sleepy during most of the day and have promoted degrees of pulmonary artery pressure and arterial PCO2. Sleep apnea is mostly observed amongst premature infants, adult males and

The frequency of occurrence of apneas is high in obese and snoring individuals with narrow URT. Apneas can be observed in the stages of sleep other than rapid eye movement (REM) and non-REM (NREM) stages. The apnea types occurring during sleep and respiratory

In literature, there are three types of sleep apnea. These are listed as central, obstructive and mixed apnea. Obstructive sleep apnea (OSA) has the highest prevalence. Together with the absence of respiratory effort in the lungs, the absence of air flow inside the mouth and nose is defined as central sleep apnea. Despite the respiratory effort, the lack of air flow in the nose and mouth is obstructive sleep apnea. The situation starting with central sleep apnea and continuing as obstructive sleep apnea is defined as mixed sleep apnea. Mixed apnea patients can be treated by the methods applied to the patients with obstructive sleep apnea. Obstructive sleep apnea is the most common sleep apnea syndrome (Aydin et al., 2005). Obstructive sleep apnea is the state of absence of oral and nasal air flow despite the respiratory effort. Although the diaphragm and intercostal muscle activity continued, exchange of air through the nose and mouth stands (Aydin et al., 2005). In this case, it Is thought to be an obstruction at the URT of patient. In order to prevent the blockage, an

Central sleep apnea (CSA) is the state in the absence of both respiratory effort and air flow together. Central apneas grow by the corruption of the central regulation of respiration. Mixed sleep apnea is the state starting with central sleep apnea and continuing the absence of oral and nasal air flow when the respiratory effort begins. How the respiratory effort after the central sleep apnea starts is still a unresolved research topic. In the new terminology,

There are a few basic definitions of sleep respiratory parameters in literature. *Hypopnea* is the 50% reduction of air flow during sleep for at least 10 seconds, 3% decrease in blood oxygen saturation or the staging of arousal. *Arousal* is defined as sudden sleep state transition to lighter sleep stages or wakefulness. Arousal terminates the apnea or hypopnea. *Apnea Index (AI)* is defined as the number of apneas per hour during sleep *Apnea+Hypopnea Index (AHI)* is defined as the number of apneas and hypopneas per hour during sleep. It is also called as *Respiratory Distress Index (RDI)*. *Respiratory Arousal Index (RAI)* is defined as the number of arousals per hour during sleep. *Obstructive Sleep Apnea* is the situation AHI>5

Table 1. Sleep Disorders

### **1.2 Sleep respiration disorders**

A significant portion of the sleep disorders are the respiratory disorders during sleep. It is thought that sudden deaths during sleep, daytime sleepiness, fatigue and snoring at night are caused by respiratory disorders in sleep. Therefore, regular breathing during sleep has vital importance for human health (Aksahin, 2010). The most important one of the sleep breathing disorders is the *sleep apnea*. Identification and monitoring of apnea during sleep is of great importance. As a result, in order to help physicians during the process of diagnosis and treatment of apnea, there are many studies on the topics of the detection and quantitative characteristics of sleep apnea using analytical methods from sleep records in literature.

The most important work on sleep in the field of engineering is the measurement and recording of physiological signals during sleep.

The device used for measuring and recording physiological signals during sleep is called as polysomnograph and the signals retrieved from the device are called as polysomnography (PSG). By the use of PSG, it is possible to observe the physiological changes in humans during sleep.

Various physiological signals of the patients are recorded simultaneously by the PSG device, which has an embedded multi-channel data acquisition system. The recording process made as analog recordings in the 90s has left its place to digital recorders after the development of digital systems. Thus, the prevention of errors caused by the hardware chaos of analog systems is provided (Erogul, 2008). By the use of these devices, Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG), Electrooculogram (EOG), breathing, Pulseplethysmograph (PPG) and various desired or necessary signals of patients in sleep are recorded. In this way, the patients' statuses are determined during the night sleep and their diagnosis and treatment outcomes can be delineated. The classification of sleep apnea is also realized by the investigation of these physiological signals obtained from the PSG device.

The first time application of PSG by Gastaut in 1965 has increased the interest in research of breathing disorders in sleep. Sleep Apnea Syndrome defined as a separate disease by Guilleminault in 1973 is renamed as "Sleep Apnea-Hypopnea Syndrome" in 1988 by the identification of hypopneas with polysomnography (Erdamar, 2007).

In 1997, ASDA (American Sleep Disorder Association) has defined obstructive sleep apnea syndrome as "A syndrome characterized by recurrent obstructions in upper respiratory tract (URT) during sleep and seen often with a decrease in oxygen saturation". The prevalence of the disease is 1-5%. Even generally knowing the risk factors during the beginning of this disease, which has no less than prevalence of Diabetes (Diabetes Mellitus) and Bronchial Asthma, the physiopathology of the obstructions are not totally explained.

By a general evaluation, the stopping of breathing during sleep for at least 10 seconds is defined as '*sleep apnea*'. Due to stopping of breathing during sleep, sleep qualities of such patients are disturbed since they often wake up at night. Additionally, they become sleepy during most of the day and have promoted degrees of pulmonary artery pressure and arterial PCO2. Sleep apnea is mostly observed amongst premature infants, adult males and post-menopausal women (Firat, 2003).

The frequency of occurrence of apneas is high in obese and snoring individuals with narrow URT. Apneas can be observed in the stages of sleep other than rapid eye movement (REM) and non-REM (NREM) stages. The apnea types occurring during sleep and respiratory parameters related to these types can be defined as follows.

### **1.3 Sleep apnea and its types**

212 Advances in Electrocardiograms – Clinical Applications

SLEEP DISORDERS

DISSOMNIAS PARASONMINAS

Insomnias Dream Disorders Hypersomnias Somnambulism

A significant portion of the sleep disorders are the respiratory disorders during sleep. It is thought that sudden deaths during sleep, daytime sleepiness, fatigue and snoring at night are caused by respiratory disorders in sleep. Therefore, regular breathing during sleep has vital importance for human health (Aksahin, 2010). The most important one of the sleep breathing disorders is the *sleep apnea*. Identification and monitoring of apnea during sleep is of great importance. As a result, in order to help physicians during the process of diagnosis and treatment of apnea, there are many studies on the topics of the detection and quantitative

The most important work on sleep in the field of engineering is the measurement and

The device used for measuring and recording physiological signals during sleep is called as polysomnograph and the signals retrieved from the device are called as polysomnography (PSG). By the use of PSG, it is possible to observe the physiological changes in humans

Various physiological signals of the patients are recorded simultaneously by the PSG device, which has an embedded multi-channel data acquisition system. The recording process made as analog recordings in the 90s has left its place to digital recorders after the development of digital systems. Thus, the prevention of errors caused by the hardware chaos of analog systems is provided (Erogul, 2008). By the use of these devices, Electroencephalogram (EEG), Electrocardiogram (ECG), Electromyogram (EMG), Electrooculogram (EOG), breathing, Pulseplethysmograph (PPG) and various desired or necessary signals of patients in sleep are recorded. In this way, the patients' statuses are determined during the night sleep and their diagnosis and treatment outcomes can be delineated. The classification of sleep apnea is also realized by the investigation of these physiological signals obtained from

The first time application of PSG by Gastaut in 1965 has increased the interest in research of breathing disorders in sleep. Sleep Apnea Syndrome defined as a separate disease by Guilleminault in 1973 is renamed as "Sleep Apnea-Hypopnea Syndrome" in 1988 by the

identification of hypopneas with polysomnography (Erdamar, 2007).

characteristics of sleep apnea using analytical methods from sleep records in literature.

Sleep Disorders in Sleep - Awake Cycle Sleep Disorders in Respiration

recording of physiological signals during sleep.

Table 1. Sleep Disorders

during sleep.

the PSG device.

**1.2 Sleep respiration disorders** 

In literature, there are three types of sleep apnea. These are listed as central, obstructive and mixed apnea. Obstructive sleep apnea (OSA) has the highest prevalence. Together with the absence of respiratory effort in the lungs, the absence of air flow inside the mouth and nose is defined as central sleep apnea. Despite the respiratory effort, the lack of air flow in the nose and mouth is obstructive sleep apnea. The situation starting with central sleep apnea and continuing as obstructive sleep apnea is defined as mixed sleep apnea. Mixed apnea patients can be treated by the methods applied to the patients with obstructive sleep apnea. Obstructive sleep apnea is the most common sleep apnea syndrome (Aydin et al., 2005).

Obstructive sleep apnea is the state of absence of oral and nasal air flow despite the respiratory effort. Although the diaphragm and intercostal muscle activity continued, exchange of air through the nose and mouth stands (Aydin et al., 2005). In this case, it Is thought to be an obstruction at the URT of patient. In order to prevent the blockage, an intense activity in the chest and abdomen is observed.

Central sleep apnea (CSA) is the state in the absence of both respiratory effort and air flow together. Central apneas grow by the corruption of the central regulation of respiration.

Mixed sleep apnea is the state starting with central sleep apnea and continuing the absence of oral and nasal air flow when the respiratory effort begins. How the respiratory effort after the central sleep apnea starts is still a unresolved research topic. In the new terminology, mixed apneas are discussed as obstructive apneas.

### **1.4 Sleep respiratory parameters**

There are a few basic definitions of sleep respiratory parameters in literature. *Hypopnea* is the 50% reduction of air flow during sleep for at least 10 seconds, 3% decrease in blood oxygen saturation or the staging of arousal. *Arousal* is defined as sudden sleep state transition to lighter sleep stages or wakefulness. Arousal terminates the apnea or hypopnea. *Apnea Index (AI)* is defined as the number of apneas per hour during sleep *Apnea+Hypopnea Index (AHI)* is defined as the number of apneas and hypopneas per hour during sleep. It is also called as *Respiratory Distress Index (RDI)*. *Respiratory Arousal Index (RAI)* is defined as the number of arousals per hour during sleep. *Obstructive Sleep Apnea* is the situation AHI>5

Automated Detection and Classification of

color image processing.

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 215

This shows it is a new area in biomedical engineering and it needs some good research to reveal the relationships between bio-signals and sleep apnea. In order to detect apnea, this chapter presents a methodology over ECG recordings, which contains time and frequency domain features. Moreover, EEG recordings are used to verify the detection of sleep apnea. The basic frequency domain features of ECG are very low frequency (VLF), low frequency (LF), high frequency (HF), the ratio of LF and HF (LF/HF) over HRV information. To obtain true HRV information, we will use Teager's energy operator to detect R-R durations. The Teager energy operator (Hamila et al., 1999) was first defined over real-valued signals and then defined for multidimensional continuous-domain signals and used for image demodulation. From the first appearance of the Teager energy operator, it has been used in several applications such as one dimensional (1-D) signal processing, image processing, and

There are previously very few works about Teager energy operator for detection of RR intervals. We will also use wavelet information to find the exact positions for P, Q-R-S and T complexes. In addition, TEO is used for other physiological signals (speech, EEG etc.). For EEG signals, short time period correlations and power spectral densities will reveal the apnea regions and help us to find the type of apnea. Power spectral densities are calculated by a few methods for example Pyulear, Welch, Burg (Stoica, 1997). These features will have valuable information for the classification of apnea types. In extent, the change in frequency features of HRV information will help us to detect the apnea regions. The consideration of this change can be determined by using statistical methods such as Annova-Mannova, and types of t-test (Gula et al., 2003). These analyze methods can be used over the considerable measurements. Their logical validation is determined by using the reliability consistency test which is a statistical analysis method and measures how much accurate is a measurement set. For example, if a scale can give the same result with the measurement process repeated n times and statistical reliability factor is close to the value α= 1, then we conclude that the *measurements of the scale are reliable*. The presence of the nonparametric variables in the physiological signals obtained from the patient shows the idea that the

Sleep records began in the 1960s for the first time, took its final form with the addition of respiratory recordings and their final definition, which is still valid, is issued by Holland

Holland used the term polysomnography to indicate the simultaneous recording, analysis and interpretation of various physiological parameters during the sleep through the night.

Polysomnography is mainly investigating the structure and physiological changes of sleep during sleep. This investigation provides reviewing sleep structure, psychological, biological and pathological changes in sleep by relating them with the sleep stages. Furthermore, it enables to investigate the sleep changes in human physiology under the conditions of sleep itself. During these investigations, whether a physiological event can be addressed alone, or multiple event and their implications can be explicated (Aydn et al., 2005).By examining data collected during sleep, the turn-up form, characteristics, the process and the response to the treatment of various diseases can be examined. In Fig. 1,

reliability test cannot be applied to these types of measurements.

**2. Sleep recording and scoring** 

**2.1 Polysomnograph and recording system** 

output records for a sample polysomnogram are shown.

and his colleagues in 1974.

during sleep. *Obstructive Sleep Apnea Syndrome* is the clinical situation that AHI>5 and major symptoms (snoring, witnessed apnea, excessive daytime sleepiness or drowsiness) met together.

*Central Sleep Apnea Syndrome* is the situation where more than 50% of detected apneas are central type. The physician examining the patients with sleep respiratory parameters suggest the diagnosis of sleep apnea syndrome.

Therefore, the scoring process performed over the PSG data obtained after sleep recording operations is extremely critical, specialized work, time consuming and inferential process (Roche et al. 1999; Firat, 2003; Aydin et al., 2005).

### **1.5 Previous studies for the detection and classification of sleep apnea**

In sleep analyzing studies, there are a lot of approaches which are about detection and classification of sleep apnea. In recent years, the popular studies contain automated detection and classification of sleep apnea. The newly developed techniques have given an increased speed to the classification and detection of sleep apnea. However, these new techniques are usually built over the detection over a data scored by a health authority. Few works have tried to construct an automated system which will detect or classify the measured data and score by itself. Most of the automated systems developed in this manner have used only one bio-signal to detect or classify the apnea. However, a fusion of features extracted from the multiple bio-signals (ECG and EEG synchronization) can be expected to give better success rates for the detection and classification.

Another interesting issue for the previously studied works is the classifier selection. Most of the works are designed to work on neural networks. Thoroughly, sleep EEG series recorded from patients are classified by using Feed Forward Neural Network (FFNN). The NN architecture has several numbers of neurons and hidden layers. This method shows that the degree of central EEG synchronization during night sleep is closely related to CSA and OSA (Aksahin et al., 2010). However, neural networks are very unstable structures when the design constraints are not selected carefully. In contrasting, very few of the works use wellknown stable classifier such as support vector machines (SVM). Additionally, it is possible to use a feature selection scheme to observe the effects of the features to the success rates.

Khandoker et al. (Khandoker, 2009a) proposed an SVM based algorithm to identify Obstructive Sleep Apnea Syndrome (OSAS) over 125 patients by using their 8-hour-long ECG recordings. They achieved high success rates to classify OSAS and non-OSAS patients. Their method covers determining the class of a patient. However, they do not mention where the apnea occurs. In the same year, Khandoker et al. (Khandoker, 2009b) tried to differentiate between OSAS and hypopnea patient in a similar methodology. Mendez et al. (Mendez, 2008) have developed an autoregressive model to screen sleep apnea from a single ECG lead. They used RR intervals and the area of QRS complex as their features for the analysis. According to Xu et al. (Xu, 2009), it is possible classify the sleep apnea by monitoring the variations on heart rate variability (HRV) signal.

Another approach to detect the OSAS patient is to use EEG signal as a source to generate features from Liu et al. (Liu, 2008) have proposed a neural network based detection method for OSAS and narcolepsy patients. They used the energy of EEG signal, theta and beta wave activities as features to detect patients correctly.

In literature, there are no previous works combining the detection and classification procedures. In this context, an automated system can be constructed to detect apnea regions and classify them as obstructive, central or mixed apnea as a new approach.

This shows it is a new area in biomedical engineering and it needs some good research to reveal the relationships between bio-signals and sleep apnea. In order to detect apnea, this chapter presents a methodology over ECG recordings, which contains time and frequency domain features. Moreover, EEG recordings are used to verify the detection of sleep apnea. The basic frequency domain features of ECG are very low frequency (VLF), low frequency (LF), high frequency (HF), the ratio of LF and HF (LF/HF) over HRV information. To obtain true HRV information, we will use Teager's energy operator to detect R-R durations. The Teager energy operator (Hamila et al., 1999) was first defined over real-valued signals and then defined for multidimensional continuous-domain signals and used for image demodulation. From the first appearance of the Teager energy operator, it has been used in several applications such as one dimensional (1-D) signal processing, image processing, and color image processing.

There are previously very few works about Teager energy operator for detection of RR intervals. We will also use wavelet information to find the exact positions for P, Q-R-S and T complexes. In addition, TEO is used for other physiological signals (speech, EEG etc.). For EEG signals, short time period correlations and power spectral densities will reveal the apnea regions and help us to find the type of apnea. Power spectral densities are calculated by a few methods for example Pyulear, Welch, Burg (Stoica, 1997). These features will have valuable information for the classification of apnea types. In extent, the change in frequency features of HRV information will help us to detect the apnea regions. The consideration of this change can be determined by using statistical methods such as Annova-Mannova, and types of t-test (Gula et al., 2003). These analyze methods can be used over the considerable measurements. Their logical validation is determined by using the reliability consistency test which is a statistical analysis method and measures how much accurate is a measurement set. For example, if a scale can give the same result with the measurement process repeated n times and statistical reliability factor is close to the value α= 1, then we conclude that the *measurements of the scale are reliable*. The presence of the nonparametric variables in the physiological signals obtained from the patient shows the idea that the reliability test cannot be applied to these types of measurements.

### **2. Sleep recording and scoring**

214 Advances in Electrocardiograms – Clinical Applications

during sleep. *Obstructive Sleep Apnea Syndrome* is the clinical situation that AHI>5 and major symptoms (snoring, witnessed apnea, excessive daytime sleepiness or drowsiness) met

*Central Sleep Apnea Syndrome* is the situation where more than 50% of detected apneas are central type. The physician examining the patients with sleep respiratory parameters

Therefore, the scoring process performed over the PSG data obtained after sleep recording operations is extremely critical, specialized work, time consuming and inferential process

In sleep analyzing studies, there are a lot of approaches which are about detection and classification of sleep apnea. In recent years, the popular studies contain automated detection and classification of sleep apnea. The newly developed techniques have given an increased speed to the classification and detection of sleep apnea. However, these new techniques are usually built over the detection over a data scored by a health authority. Few works have tried to construct an automated system which will detect or classify the measured data and score by itself. Most of the automated systems developed in this manner have used only one bio-signal to detect or classify the apnea. However, a fusion of features extracted from the multiple bio-signals (ECG and EEG synchronization) can be expected to

Another interesting issue for the previously studied works is the classifier selection. Most of the works are designed to work on neural networks. Thoroughly, sleep EEG series recorded from patients are classified by using Feed Forward Neural Network (FFNN). The NN architecture has several numbers of neurons and hidden layers. This method shows that the degree of central EEG synchronization during night sleep is closely related to CSA and OSA (Aksahin et al., 2010). However, neural networks are very unstable structures when the design constraints are not selected carefully. In contrasting, very few of the works use wellknown stable classifier such as support vector machines (SVM). Additionally, it is possible to use a feature selection scheme to observe the effects of the features to the success rates. Khandoker et al. (Khandoker, 2009a) proposed an SVM based algorithm to identify Obstructive Sleep Apnea Syndrome (OSAS) over 125 patients by using their 8-hour-long ECG recordings. They achieved high success rates to classify OSAS and non-OSAS patients. Their method covers determining the class of a patient. However, they do not mention where the apnea occurs. In the same year, Khandoker et al. (Khandoker, 2009b) tried to differentiate between OSAS and hypopnea patient in a similar methodology. Mendez et al. (Mendez, 2008) have developed an autoregressive model to screen sleep apnea from a single ECG lead. They used RR intervals and the area of QRS complex as their features for the analysis. According to Xu et al. (Xu, 2009), it is possible classify the sleep apnea by

Another approach to detect the OSAS patient is to use EEG signal as a source to generate features from Liu et al. (Liu, 2008) have proposed a neural network based detection method for OSAS and narcolepsy patients. They used the energy of EEG signal, theta and beta wave

In literature, there are no previous works combining the detection and classification procedures. In this context, an automated system can be constructed to detect apnea regions

and classify them as obstructive, central or mixed apnea as a new approach.

**1.5 Previous studies for the detection and classification of sleep apnea** 

together.

suggest the diagnosis of sleep apnea syndrome.

(Roche et al. 1999; Firat, 2003; Aydin et al., 2005).

give better success rates for the detection and classification.

monitoring the variations on heart rate variability (HRV) signal.

activities as features to detect patients correctly.

Sleep records began in the 1960s for the first time, took its final form with the addition of respiratory recordings and their final definition, which is still valid, is issued by Holland and his colleagues in 1974.

Holland used the term polysomnography to indicate the simultaneous recording, analysis and interpretation of various physiological parameters during the sleep through the night.

### **2.1 Polysomnograph and recording system**

Polysomnography is mainly investigating the structure and physiological changes of sleep during sleep. This investigation provides reviewing sleep structure, psychological, biological and pathological changes in sleep by relating them with the sleep stages. Furthermore, it enables to investigate the sleep changes in human physiology under the conditions of sleep itself. During these investigations, whether a physiological event can be addressed alone, or multiple event and their implications can be explicated (Aydn et al., 2005).By examining data collected during sleep, the turn-up form, characteristics, the process and the response to the treatment of various diseases can be examined. In Fig. 1, output records for a sample polysomnogram are shown.

Automated Detection and Classification of

Fig. 2. QRS complex of a ECG signal (Köhler et al., 2002)

activity of the brain waves (Pehlivan, 2004).

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 217

Beta waves are observed at the frequencies higher than 13 Hz and their amplitudes change in the range of 1-5 μV. They are observed at focused attention, mental working states, and rapid eye movement stages of sleep. These waves correspond to the level of the highest

By using EEG signals, sleep stages during the sleep period can be distinguished. These stages are divided into 5 classes. In general, as the sleep gets deeper, a frequency decrease in EEG signals. In sleep, there are two change stages following each other periodically. These are REM (Rapid Eye Movement – Paradoxical Sleep) and NREM (Non-Rapid Eye Movement) stages. The period between eyes shut to sleep and full sleep stage is called as the

NREM sleep is occasionally divided into 5 phases based on the electroencephalographic changes occurring during the course of sleep. Phase 0 is wakefulness stage, Phase 1 and 2 is the shallow sleep period, Phase 3 and 4 are the periods of deep sleep. It is possible to observe high amplitude low frequency waves and spindles in EEG. In this stage, there is no eye movements, muscle tones are decreased, pulse and respiratory are slowed. In Table 2,

REM sleep is the dream stage of sleep or the dreams seen in this stage can be remembered in wakefulness state. This stage is interspersed between the other phases of the sleep. It is

Phase I The state one individual who is diving into sleep has faced. If

Phase II In this phase of sleep, consciousness is in a state that he will be

the individual is forced to wake up in this phase of sleep, he will say he was usually awake even he was not aware of the

aware of he was in sleep when he was forced to wake up. EEG patterns can be seen. K- complex and sleep spindles are

latent period of diving into sleep. After latent period, the exchange periods start.

the phases of NREM sleep and their characteristics are given (Park, 2000).

**NREM Sleep Phases Characteristics**  Phase 0 (Wakefulness) The stage before diving into sleep.

events around him.

expected to be seen.

connected with a large number of different features.

Phase III and IV Low frequency deep sleep. Table 2. NREM Sleep Phases and their characteristics.


Fig. 1. Output records for a polysomnogram

Simultaneous recording of the sleep parameters should be able to give sufficient information about both the sleep and breathing pattern in sleep. In order to do staging of sleep, at least 2-channel EEG, EOG, chin EMG, oro-nasal air flow, arterial oxygen saturation, respiratory effort and ECG or pulse recordings must be done. In many cases, the anterior tibial EMG recording can be useful to detect periodic motion disorder.

Polysomnogram is used, in general, for the investigation, diagnosis and following up the treatment of sleep related breathing disorders.

### **2.2 Electrocardiogram (ECG) recordings**

The technology interpreting the changes in electrical potential during heart activity is called as electrocardiography and the recorded by the device are called as electrocardiogram (ECG). Fig. 2 shows a normal ECG recording.

The duration of P wave and QRS complex is 0.1 seconds while the average wave amplitude (peak R) is around 1 mV. This peak value rarely rises up to 5 mV. These nominal values can change from one person to another by the patient's heart and body size and body conductivity.

### **2.3 Electroencephalography (EEG) recordings**

EEG is the system used to measure and record the electrical activity of the brain. The brain waves obtained from EEG system have delta, theta, alpha, and beta wave bands.

Delta waves have 0.5-4 Hz frequency band with 20-400 μV amplitudes and are encountered in the situations of very low activity of brain, such as deep sleep and general anesthetic state.

Theta waves have 4-8 Hz frequency band with 100-500 μV amplitudes and are encountered in the situations of low activity of brain, such as dreaming sleep and medium anesthetic state.

Alpha waves have 8-13 Hz frequency band with 2-10 μV amplitudes and they are closest ones to the sinusoidal form between EEG waves. They are observed at awake individuals with closed eyes in a physically and mentally full resting condition.

Fig. 2. QRS complex of a ECG signal (Köhler et al., 2002)

216 Advances in Electrocardiograms – Clinical Applications

Simultaneous recording of the sleep parameters should be able to give sufficient information about both the sleep and breathing pattern in sleep. In order to do staging of sleep, at least 2-channel EEG, EOG, chin EMG, oro-nasal air flow, arterial oxygen saturation, respiratory effort and ECG or pulse recordings must be done. In many cases, the anterior tibial EMG

Polysomnogram is used, in general, for the investigation, diagnosis and following up the

The technology interpreting the changes in electrical potential during heart activity is called as electrocardiography and the recorded by the device are called as electrocardiogram

The duration of P wave and QRS complex is 0.1 seconds while the average wave amplitude (peak R) is around 1 mV. This peak value rarely rises up to 5 mV. These nominal values can change from one person to another by the patient's heart and body size and body

EEG is the system used to measure and record the electrical activity of the brain. The brain

Delta waves have 0.5-4 Hz frequency band with 20-400 μV amplitudes and are encountered in the situations of very low activity of brain, such as deep sleep and general anesthetic

Theta waves have 4-8 Hz frequency band with 100-500 μV amplitudes and are encountered in the situations of low activity of brain, such as dreaming sleep and medium anesthetic

Alpha waves have 8-13 Hz frequency band with 2-10 μV amplitudes and they are closest ones to the sinusoidal form between EEG waves. They are observed at awake individuals

waves obtained from EEG system have delta, theta, alpha, and beta wave bands.

with closed eyes in a physically and mentally full resting condition.

Fig. 1. Output records for a polysomnogram

treatment of sleep related breathing disorders.

**2.2 Electrocardiogram (ECG) recordings** 

(ECG). Fig. 2 shows a normal ECG recording.

**2.3 Electroencephalography (EEG) recordings** 

conductivity.

state.

state.

recording can be useful to detect periodic motion disorder.

Beta waves are observed at the frequencies higher than 13 Hz and their amplitudes change in the range of 1-5 μV. They are observed at focused attention, mental working states, and rapid eye movement stages of sleep. These waves correspond to the level of the highest activity of the brain waves (Pehlivan, 2004).

By using EEG signals, sleep stages during the sleep period can be distinguished. These stages are divided into 5 classes. In general, as the sleep gets deeper, a frequency decrease in EEG signals. In sleep, there are two change stages following each other periodically. These are REM (Rapid Eye Movement – Paradoxical Sleep) and NREM (Non-Rapid Eye Movement) stages. The period between eyes shut to sleep and full sleep stage is called as the latent period of diving into sleep. After latent period, the exchange periods start.

NREM sleep is occasionally divided into 5 phases based on the electroencephalographic changes occurring during the course of sleep. Phase 0 is wakefulness stage, Phase 1 and 2 is the shallow sleep period, Phase 3 and 4 are the periods of deep sleep. It is possible to observe high amplitude low frequency waves and spindles in EEG. In this stage, there is no eye movements, muscle tones are decreased, pulse and respiratory are slowed. In Table 2, the phases of NREM sleep and their characteristics are given (Park, 2000).

REM sleep is the dream stage of sleep or the dreams seen in this stage can be remembered in wakefulness state. This stage is interspersed between the other phases of the sleep. It is connected with a large number of different features.


Table 2. NREM Sleep Phases and their characteristics.

Automated Detection and Classification of

Table 3. R-wave Detection Methods

Teager energy operator is defined as:

the Teager energy operator given below:

detect R-wave peak points.

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 219

���

���

<sup>M</sup> � x[n − �]

From Table 3, it can be said that the most commonly used method is the Teager energy operator based on derivative operation. Teager energy operator is a very convenient method for studying over a single component in both continuous and discrete domains. It carries a lot of attributes together and is derived from the signal energy. In continuous domain,

dt �

Teager energy operator is applied off-line over the signal on the basis of discrete time domain. Sampled at a particular frequency, ECG signal is applied to the discrete version of

Teager energy operator covering the three samples of the signal side by side, shows a very local feature of the signal (Kaiser, 1993). In Fig. 3, an exemplary sequence of the ECG signal

Before applying the Teager energy operator, ECG signal must be freed from noise. These noises are usually seen as baseline drift and motion artifact. In both time and frequency domains, there are algorithms used to eliminate these types of noises. For example, if there is a low frequency noise on ECG signal, a "moving average filter" in time domain

After the TEO algorithm is applied to ECG signal, a threshold value comparison should be made to determine the R-wave peak points. 60% of the maximum value of the output signal from TEO block can be chosen as the threshold value. If the output of TEO exceeds the threshold at time t0 and no greater value in the next 0.25 seconds is observed, t0 is marked as R-wave peak point (Erdamar, 2007). TEO output of a sample ECG signal and detected R

Heart rate variability is defined as the time differences between the two consecutive R-wave peak points and are expressed in units of milliseconds or seconds. Heart rate variability data given in Fig. 5 is examined in both time and frequency domains. In case of missing R wave or finding extra R waves, we can observe upside or downside peaks on the HRV data. In order to eliminate this situation, an adaptive thresholding algorithm must be applied to

�

− x(t)

d�x(t)

<sup>Ψ</sup>[x(n)] = x�(n) − x(n+1)x(n−1) (3)

dt� (2)

(1)

y� <sup>=</sup> <sup>1</sup>

Teager Energy Operator (TEO)

Wavelet Transform Support Vector Machines Pan- Tompkins Algorithm Filtering in Frequency Domain

Zero Crossing Rate

Ψ[x(t)] � �dx(t)

and the corresponding output of the Teager energy operator is given.

(Rangayyan, 2002) or high-pass filters in frequency domain are applied.

wave peaks after the threshold comparison is shown in Fig. 4.

### **2.4 Other physiological measurement parameters**

*EMG* (Electromyography) is the method of observing and recording of skeletal electrical activity. EMG signals contain very high frequency components ranging between 10 and 5000 Hz. *EOG* is the method of observing and recording of eye movements resulting from electrical activity of muscles.

*Thermistor* is the sensor detecting the air flow from the nose. With the help of thermistor, it is determined whether there is an obstruction at the upper respiratory tract or not. Microphone records the sound from the mouth. With the help of microphone, the recorded sounds of snoring can be examined.

*Thoracic respiratory signal*, which is obtained from the expansion and narrowing of the chest, is recorded by wounding a special type of tape around the chest. The breathing can be examined through the recorded signal.

*SpO2* is the signal showing the blood oxygen saturation. With the signal recorded through PPG device, amount of change of oxygen in the blood after breathing is analyzed.

### **3. The sleep apnea features based on electrocardiogram (ECG) and electroencephalogram (EEG) biosignals**

Sleep apnea detection approach over ECG and EEG physiological signals taken from polysomnograph recently appears to be a popular study. The methods apart from the classical methods followed by the physician to detect sleep apnea is based on the examination of features correlated with the sleep apnea from ECG and EEG signals to construct an automatic decision support mechanism.

### **3.1 Sleep apnea features based on electrocardiogram biosignals**

Heart rate variability (HRV) information takes the first place amongst the researchers conducted on ECG physiological signals. Heart rate variability is related to the nature and the presence of sleep apnea. While many commercial medical equipments measure the automatically, they cannot bring a fully automated approach on detection and diagnosis of disorders such as sleep apnea.

Heart rate variability includes many features in both time and frequency domains and needs different methods in calculation of these features. ECG raw data, which is analyzed statistically in both time and frequency domains, is separated as short term and long term by the length of data (Camm et al., 1996).

### **3.1.1 Calculation of heart rate variability**

In order to perform the calculation of heart rate variability from the ECG signal, first of all, R-wave detection needs to be done as much as possible. There are many approaches in detection of R-wave in the literature. Some of these methods are summarized in Table 3.

For example, there are methods based on wavelet transform and digital filtering for R-wave detection. In order to remove noise from ECG signal, it is passed through a wavelet transform block and QRS complex is emphasized in contrast to P and T waves by the moving average based low pass filter given in Eq. 1. In Eq. 1, y1 is the output signal, x[n] is the input signal and M is regarded as the length of the filter. Later, the processed ECG signal, which is passed through a nonlinear amplifier, is applied to an adaptive thresholding block (Chen et al., 2006).

$$\mathbf{y}\_1 = \frac{1}{\mathbf{M}} \sum\_{\mathbf{m}=\mathbf{0}}^{\mathbf{M}-1} \mathbf{x}[\mathbf{n} - \mathbf{m}] \tag{1}$$


Table 3. R-wave Detection Methods

218 Advances in Electrocardiograms – Clinical Applications

*EMG* (Electromyography) is the method of observing and recording of skeletal electrical activity. EMG signals contain very high frequency components ranging between 10 and 5000 Hz. *EOG* is the method of observing and recording of eye movements resulting from

*Thermistor* is the sensor detecting the air flow from the nose. With the help of thermistor, it is determined whether there is an obstruction at the upper respiratory tract or not. Microphone records the sound from the mouth. With the help of microphone, the recorded

*Thoracic respiratory signal*, which is obtained from the expansion and narrowing of the chest, is recorded by wounding a special type of tape around the chest. The breathing can be

*SpO2* is the signal showing the blood oxygen saturation. With the signal recorded through

Sleep apnea detection approach over ECG and EEG physiological signals taken from polysomnograph recently appears to be a popular study. The methods apart from the classical methods followed by the physician to detect sleep apnea is based on the examination of features correlated with the sleep apnea from ECG and EEG signals to construct an automatic

Heart rate variability (HRV) information takes the first place amongst the researchers conducted on ECG physiological signals. Heart rate variability is related to the nature and the presence of sleep apnea. While many commercial medical equipments measure the automatically, they cannot bring a fully automated approach on detection and diagnosis of

Heart rate variability includes many features in both time and frequency domains and needs different methods in calculation of these features. ECG raw data, which is analyzed statistically in both time and frequency domains, is separated as short term and long term

In order to perform the calculation of heart rate variability from the ECG signal, first of all, R-wave detection needs to be done as much as possible. There are many approaches in detection of R-wave in the literature. Some of these methods are summarized in Table 3. For example, there are methods based on wavelet transform and digital filtering for R-wave detection. In order to remove noise from ECG signal, it is passed through a wavelet transform block and QRS complex is emphasized in contrast to P and T waves by the moving average based low pass filter given in Eq. 1. In Eq. 1, y1 is the output signal, x[n] is the input signal and M is regarded as the length of the filter. Later, the processed ECG signal, which is passed through a nonlinear amplifier, is applied to an adaptive thresholding

PPG device, amount of change of oxygen in the blood after breathing is analyzed.

**3. The sleep apnea features based on electrocardiogram (ECG) and** 

**3.1 Sleep apnea features based on electrocardiogram biosignals** 

**2.4 Other physiological measurement parameters** 

electrical activity of muscles.

decision support mechanism.

disorders such as sleep apnea.

block (Chen et al., 2006).

by the length of data (Camm et al., 1996).

**3.1.1 Calculation of heart rate variability** 

sounds of snoring can be examined.

examined through the recorded signal.

**electroencephalogram (EEG) biosignals** 

From Table 3, it can be said that the most commonly used method is the Teager energy operator based on derivative operation. Teager energy operator is a very convenient method for studying over a single component in both continuous and discrete domains. It carries a lot of attributes together and is derived from the signal energy. In continuous domain, Teager energy operator is defined as:

$$
\Psi[\mathbf{x(t)}] \triangleq \left(\frac{\mathrm{d}\mathbf{x(t)}}{\mathrm{d}\mathbf{t}}\right)^2 - \mathbf{x(t)}\frac{\mathrm{d}^2\mathbf{x(t)}}{\mathrm{d}\mathbf{t}^2} \tag{2}
$$

Teager energy operator is applied off-line over the signal on the basis of discrete time domain. Sampled at a particular frequency, ECG signal is applied to the discrete version of the Teager energy operator given below:

$$
\Psi[\mathbf{x(n)}] = \mathbf{x^2(n) - x(n+1)x(n-1)}\tag{3}
$$

Teager energy operator covering the three samples of the signal side by side, shows a very local feature of the signal (Kaiser, 1993). In Fig. 3, an exemplary sequence of the ECG signal and the corresponding output of the Teager energy operator is given.

Before applying the Teager energy operator, ECG signal must be freed from noise. These noises are usually seen as baseline drift and motion artifact. In both time and frequency domains, there are algorithms used to eliminate these types of noises. For example, if there is a low frequency noise on ECG signal, a "moving average filter" in time domain (Rangayyan, 2002) or high-pass filters in frequency domain are applied.

After the TEO algorithm is applied to ECG signal, a threshold value comparison should be made to determine the R-wave peak points. 60% of the maximum value of the output signal from TEO block can be chosen as the threshold value. If the output of TEO exceeds the threshold at time t0 and no greater value in the next 0.25 seconds is observed, t0 is marked as R-wave peak point (Erdamar, 2007). TEO output of a sample ECG signal and detected R wave peaks after the threshold comparison is shown in Fig. 4.

Heart rate variability is defined as the time differences between the two consecutive R-wave peak points and are expressed in units of milliseconds or seconds. Heart rate variability data given in Fig. 5 is examined in both time and frequency domains. In case of missing R wave or finding extra R waves, we can observe upside or downside peaks on the HRV data. In order to eliminate this situation, an adaptive thresholding algorithm must be applied to detect R-wave peak points.

Automated Detection and Classification of

**RR intervals** 

**(sec)** 

finding 24 hours noiseless raw data.

segments

intervals.

(Roche et al., 1999).

Fig. 5. Heart rate variability obtained from ECG signal

SDANN Standard deviation of the mean of all consecutive 5 minute segments of normal RR intervals

r-MSSD Root mean square of successive differences between

NN50 count Number of pairs of adjacent NN intervals differing by

pNN50 NN50 count divided by the total number of all NN

more than 50 ms in the entire recording. Three variants are possible counting all such NN intervals pairs or only pairs in which the first or the second interval is longer.

**Frequency Domain Features**  VLF Very low frequency 0.015- 0.04 Hz short term LF Low frequency 0.04-0.15 Hz short term HF High frequency 0.15- 0.4 Hz short term LF/HF Short term

While the time domain analyses are calculated through HRV itself, the frequency domain analyses are calculated through the power spectral density of HRV. In order to do a frequency

In most studies, it is observed that the power spectral densities obtained from the signals containing the sleep apnea syndrome have more intense low frequency (LF) components than that of the normal signals. After the treatment of sleep apnea it is usually seen that low frequency density has decreased and converged to the nominal values of normal individuals

Table 5 exhibits the time and frequency domain feature values of a normal individual (Camm et al., 1996). On the other hand, it is possible to obtain different feature values over

domain analysis, HRV must be freed from its DC component and resampled at 2 Hz.

adjacent normal RR intervals

Table 4. Time and Frequency Domain Features of HRV

SDNN index Mean of the standard deviation of consecutive 5 minute

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 221

HRV parameter can be used to obtain the time and frequency domain features given in Table 4. Time domain features are calculated over 24 hours (long term) recordings and frequency domain features are calculated over 2-5 minutes (short term) recordings (Camm et al., 1996; Roche et al., 2002). The difficulty of the long term analysis is the impossibility of

**RR beat number**

**Time Domain Features**  SDNN Standard deviation of all normal RR intervals long term

long term

long term

long term

long term

long term

Fig. 3. ECG signal (top) and TEO output (bottom)

Fig. 4. TEO output of ECG signal (top) and detected R-wave peak points(bottom)

Fig. 5. Heart rate variability obtained from ECG signal

Fig. 3. ECG signal (top) and TEO output (bottom)

**Amplitude Amplitude** 

**Am**

**plitude Amplitude** 

Fig. 4. TEO output of ECG signal (top) and detected R-wave peak points(bottom)

**Number of Samples** 

**Number of Samples**

**Number of Samples**

**Number of Samples**

HRV parameter can be used to obtain the time and frequency domain features given in Table 4. Time domain features are calculated over 24 hours (long term) recordings and frequency domain features are calculated over 2-5 minutes (short term) recordings (Camm et al., 1996; Roche et al., 2002). The difficulty of the long term analysis is the impossibility of finding 24 hours noiseless raw data.


Table 4. Time and Frequency Domain Features of HRV

While the time domain analyses are calculated through HRV itself, the frequency domain analyses are calculated through the power spectral density of HRV. In order to do a frequency domain analysis, HRV must be freed from its DC component and resampled at 2 Hz.

In most studies, it is observed that the power spectral densities obtained from the signals containing the sleep apnea syndrome have more intense low frequency (LF) components than that of the normal signals. After the treatment of sleep apnea it is usually seen that low frequency density has decreased and converged to the nominal values of normal individuals (Roche et al., 1999).

Table 5 exhibits the time and frequency domain feature values of a normal individual (Camm et al., 1996). On the other hand, it is possible to obtain different feature values over

Automated Detection and Classification of

Rangayyan, 2002).

In form factor calculation, ��

(Malinowska et al., 2006).

observed at the corresponding sub-band range.

Fig. 7. a) Thermistor signal b) EEG sub-bands

portions within the scope of micro structure (Erdamar, 2007).

Delta Theta Alpha Beta 1 Beta 2

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 223

Another statistical methods based parameter is Form Factor. This parameter is a statistical value depending on the exchange of activity amongst one-second-long signals (Hjorth, 1973;

> μ� � �σ�� � σ� � � ���

�� � ���� ��

EEG signals are another physiological signal type used in the detection and classification of sleep apnea. Due to the morphological properties separating from the ECG signals, they are applied to different analyses. Since sleep apnea is a ten second long event, they can usually be detected by an examination of epochs before and after sleep apnea. If the analyses cover 20-30 seconds part, they contain macro structure. The processes such as spindle scaling, slow wave activities, and arousal detection are included in macro structure analysis

In Fig. 7, a regular part of the respiratory system and the corresponding sub-band representation of the EEG signal in that epoch are shown. The horizontal axis represents the time and the vertical axis represents amplitude. In occurrence of sleep apnea, changes

In Fig. 8, respiratory signal with apnea and the corresponding sub-bands of EEG signal are shown. In this representation, change in the sub-bands of the EEG section corresponding to the respiratory signal with apnea is observed. The existence and amount of this change can be proved by calculating power spectral density. The analyses are made over 10 second

Some approaches are based on the EEG arousal detection of micro-structures. This is because the EEG signal of a patient diagnosed with OSA contains arousal structures during

formula for variability and in Eq. 5, form factor formula is given.

**3.2 The features based on electroencephalogram biosignals** 

� (activity-variance) and ��� (variability) are used. In Eq. 4, the

� (5)

(4)

data gathered from different patient groups and different polysomnography devices. The reasons for these differences are the resolution of the device and patients having different characteristics of physiological parameters affecting AHI values. We suggest creating simultaneously a control group of healthy subjects as a reference in HRV analysis procedures.

Fig. 6. Power Spectral Density of HRV

Choosing the created control group as a reference and analyzing the other patient groups will be useful to obtain statistically significant results.

Statistical methods, such as Anova-Manova, student's t test, multi variable regression and correlation, are used for scientific analysis of the features obtained from the spectral analysis of HRV. For instance, ANOVA test can be applied to show the statistical difference between groups of patients with sleep apnea. Within the scope of this test, it is clarified as a result that two groups are different when P<0.05 between the groups is achieved (Gula et al., 2001).


Table 5. Normal values of HRV features

Another statistical methods based parameter is Form Factor. This parameter is a statistical value depending on the exchange of activity amongst one-second-long signals (Hjorth, 1973; Rangayyan, 2002).

In form factor calculation, �� � (activity-variance) and ��� (variability) are used. In Eq. 4, the formula for variability and in Eq. 5, form factor formula is given.

$$
\mu\_{\mathbf{x}} = \left(\frac{\sigma\_{\mathbf{x}}^2}{\sigma\_{\mathbf{x}}^2}\right)^{1/2} \tag{4}
$$

$$FF = \left(\frac{\mu\_{\ge \prime}}{\mu\_{\ge}}\right) \tag{5}$$

### **3.2 The features based on electroencephalogram biosignals**

222 Advances in Electrocardiograms – Clinical Applications

data gathered from different patient groups and different polysomnography devices. The reasons for these differences are the resolution of the device and patients having different characteristics of physiological parameters affecting AHI values. We suggest creating simultaneously a control group of healthy subjects as a reference in HRV analysis procedures.

Choosing the created control group as a reference and analyzing the other patient groups

Statistical methods, such as Anova-Manova, student's t test, multi variable regression and correlation, are used for scientific analysis of the features obtained from the spectral analysis of HRV. For instance, ANOVA test can be applied to show the statistical difference between groups of patients with sleep apnea. Within the scope of this test, it is clarified as a result that two groups are different when P<0.05 between the groups is achieved (Gula et al.,

Variables Units Normal Values (mean ± SD) **Time Domain Analysis of 24 h**  SDNN ms 141±39 SDANN ms 127±35 RMSSD ms 27±12 **Spectral analysis of stationary supine 5-min recording**  Total power ms2 3466±1018 LF ms2 1170±416 HF ms2 975±203

Fig. 6. Power Spectral Density of HRV

2001).

will be useful to obtain statistically significant results.

LF/HF ratio nu

Table 5. Normal values of HRV features

EEG signals are another physiological signal type used in the detection and classification of sleep apnea. Due to the morphological properties separating from the ECG signals, they are applied to different analyses. Since sleep apnea is a ten second long event, they can usually be detected by an examination of epochs before and after sleep apnea. If the analyses cover 20-30 seconds part, they contain macro structure. The processes such as spindle scaling, slow wave activities, and arousal detection are included in macro structure analysis (Malinowska et al., 2006).

In Fig. 7, a regular part of the respiratory system and the corresponding sub-band representation of the EEG signal in that epoch are shown. The horizontal axis represents the time and the vertical axis represents amplitude. In occurrence of sleep apnea, changes observed at the corresponding sub-band range.

Fig. 7. a) Thermistor signal b) EEG sub-bands

In Fig. 8, respiratory signal with apnea and the corresponding sub-bands of EEG signal are shown. In this representation, change in the sub-bands of the EEG section corresponding to the respiratory signal with apnea is observed. The existence and amount of this change can be proved by calculating power spectral density. The analyses are made over 10 second portions within the scope of micro structure (Erdamar, 2007).

Some approaches are based on the EEG arousal detection of micro-structures. This is because the EEG signal of a patient diagnosed with OSA contains arousal structures during

Automated Detection and Classification of

\*Statistically significant

other (P = 0.006 <0.05).

apnea osa18 – non-apnea osa18

Table 8. Paired Samples Test Outputs

to the FF series of an ECG recording with apnea.

decrease in heart rate in case of apnea is found.

(long term) ECG recordings.

each other.

Pair 1

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 225

prove the difference. In this context, the statistical data of the LF / HF rates obtained from

In independent samples T-test, P value less than 0.05 means that analyzed two groups are statistically different from each other. The results show that OSA, CSA and normal diagnosed patients can be separated by a statistical analysis on change of LF / HF rates of

In order to observe the ability to separate between the epochs with and without apnea, paired sampled t-test analysis method is used. In Table 8, it is shown that 10 epochs with apnea and 10 epochs without apnea of an OSA patient are statistically different from each

This statistical result scientifically indicates that the parts with apnea have different morphological features than the ones without apnea even for the same patient's 24 hours

Paired Differences

On the other hand, form factor (FF) calculation can be shown as a different approach for the detection of sleep apnea and different approach from literature. FF calculations made for a segment from 160 msec before to 240 msec after of each detected R-wave of the ECG signal can be thought as a determining factor for predicting the sleep apnea. In experimental works, a threshold FF value was determined in ECG recordings to identify apnea. In this context, apnea is appeared for 4 and upper FF values in ECG recordings. Fig. 9 corresponds

In Fig. 10, the thermistor signal with 256 Hz sampling frequency of non-apnea patient and corresponding FF values calculated on ECG signal are given. When a number of FF values obtained from apnea and non-apnea patient recordings with equal length are examined, a

Std. Error Mean

Mean tailed)

3.89091 3.74696 1.12975 1.37367 6.40815 3.444 10 **.006** 

95% Confidence Interval of the Difference Lower Upper

t df

Sig. (2-

600 epochs (300 minutes) ECG recordings is gathered in Table 7.

Table 7. Independent samples T-test between the patients groups

Std. Deviation

Number of Epoch

Normal OSA 200 - 200 P=0.000 (<0.05)\* Normal CSA 200 - 200 P=0.004 (<0.05)\* OSA CSA 200 - 200 P=0.001 (<0.05)\*

Fig. 8. a) Thermistor signal b) EEG sub-bands

sleep apnea. In detection of arousal structures, data from EEG, pressure and temperature of thermistor, chin-EMG, and tibialis EMG are used.

In order to detect the status that arousal response is longer than 3 seconds, the PSG data is examined in 1.28-second segments with a 0.4 Hz frequency resolution. Chin-EMG and tibialis EMG records are important for the detection of pathological events. Since arousal events inside the EEG signal generates rapid changes in all frequency components, direct analysis methods give no results (Sugi, 2008).

In addition, the important alterations are observed on previous divisions of sleep apneas. In sleep apnea events, the most significant alternations are observed in alpha and beta waves. These alterations cannot be realized in whole EEG spectrum because of their sizes. Therefore, EEG signals can be analyzed by using short time fouirer transform of narrow windows (Erdamar, 2007).

### **4. Results**

The R detection algorithm based on TEO and adaptive thresholding (Fig.3 and Fig 4.) was applied over 600 epochs (200 epochs normal patients, 200 epochs OSA, 200 epochs CSA). The LF/HF rates calculated from HRV of these patients are shown in Table 6.


Table 6. LF/HF rates of three different patients

In Table 6, it is shown that the LF/HF rates of OSA and CSA are higher than normal patients' values. This difference is visible on Table 6 but, we need to do a statistical test to


prove the difference. In this context, the statistical data of the LF / HF rates obtained from 600 epochs (300 minutes) ECG recordings is gathered in Table 7.

\*Statistically significant

224 Advances in Electrocardiograms – Clinical Applications

sleep apnea. In detection of arousal structures, data from EEG, pressure and temperature of

In order to detect the status that arousal response is longer than 3 seconds, the PSG data is examined in 1.28-second segments with a 0.4 Hz frequency resolution. Chin-EMG and tibialis EMG records are important for the detection of pathological events. Since arousal events inside the EEG signal generates rapid changes in all frequency components, direct

In addition, the important alterations are observed on previous divisions of sleep apneas. In sleep apnea events, the most significant alternations are observed in alpha and beta waves. These alterations cannot be realized in whole EEG spectrum because of their sizes. Therefore, EEG signals can be analyzed by using short time fouirer transform of narrow

The R detection algorithm based on TEO and adaptive thresholding (Fig.3 and Fig 4.) was applied over 600 epochs (200 epochs normal patients, 200 epochs OSA, 200 epochs CSA).

**mean** 

**OSA 1** 0.87 0.97 1.45 1.15 1.46 1.28 1.42 1.81 1.03 2.6 1.4 **OSA 2** 1.1 3.4 1.1 5.3 1.58 3.5 1.1 1.87 5 2.5 2.7 **CSA 1** 2.03 5.21 0.520 1.21 0.59 0.74 2.149 0.35 0.71 2.747 1.63 **CSA 2** 3.09 1.13 0.73 1.06 1.62 1.15 3.12 0.49 1.22 1.33 1.5

In Table 6, it is shown that the LF/HF rates of OSA and CSA are higher than normal patients' values. This difference is visible on Table 6 but, we need to do a statistical test to

0.73 0.49 0.3 0.81 0.3 0.73 0.671 0.7 0.21 0.89 0.58

0.4 0.64 1.126 0.45 0.66 0.82 0.9 0.86 0.23 1.52 0.76

**LF/HF** 

The LF/HF rates calculated from HRV of these patients are shown in Table 6.

Fig. 8. a) Thermistor signal b) EEG sub-bands

analysis methods give no results (Sugi, 2008).

Table 6. LF/HF rates of three different patients

windows (Erdamar, 2007).

**4. Results** 

**Normal Patient 1** 

**Normal Patient 2** 

thermistor, chin-EMG, and tibialis EMG are used.

Delta Theta Alpha Beta 1 Beta 2

Table 7. Independent samples T-test between the patients groups

In independent samples T-test, P value less than 0.05 means that analyzed two groups are statistically different from each other. The results show that OSA, CSA and normal diagnosed patients can be separated by a statistical analysis on change of LF / HF rates of each other.

In order to observe the ability to separate between the epochs with and without apnea, paired sampled t-test analysis method is used. In Table 8, it is shown that 10 epochs with apnea and 10 epochs without apnea of an OSA patient are statistically different from each other (P = 0.006 <0.05).

This statistical result scientifically indicates that the parts with apnea have different morphological features than the ones without apnea even for the same patient's 24 hours (long term) ECG recordings.


Table 8. Paired Samples Test Outputs

On the other hand, form factor (FF) calculation can be shown as a different approach for the detection of sleep apnea and different approach from literature. FF calculations made for a segment from 160 msec before to 240 msec after of each detected R-wave of the ECG signal can be thought as a determining factor for predicting the sleep apnea. In experimental works, a threshold FF value was determined in ECG recordings to identify apnea. In this context, apnea is appeared for 4 and upper FF values in ECG recordings. Fig. 9 corresponds to the FF series of an ECG recording with apnea.

In Fig. 10, the thermistor signal with 256 Hz sampling frequency of non-apnea patient and corresponding FF values calculated on ECG signal are given. When a number of FF values obtained from apnea and non-apnea patient recordings with equal length are examined, a decrease in heart rate in case of apnea is found.

Automated Detection and Classification of

to find approaches in both time and frequency domains.

working strategy in contrast to neural networks.

patients with sleep apnea can be easily differentiated.

discrete time series.

temp= chosen template QRS complex ecgi = windowing of raw ECG (size of temp)

all the biological system is more efficient.

ecg=raw data

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 227

limited. Thus, the sections, that have negative effect on variance and cause the algorithm to work inaccurately, are removed from the ECG signal and then the HRV analysis is done. When an automated system is requested, a general algorithm that can remove these corrupted fields from the ECG signal can be written at most. For this purpose, it is possible

In detection and classification of sleep apneas, usage of support vector machines (SVM) instead of neural network based algorithms may be a better and more original approach. SVM, similar to neural networks, uses a part of the data for training purposes and the rest of data to testing purposes. On contrary, by its structure, it can present a more determined

It is possible to obtain a very good HRV curve with the help of an algorithm that does not miss R-peaks as much as possible. In order to investigate the frequency domain attributes of HRV, the frequency characteristics are obtained from the HRV after the fast fourier transform (FFT) of it calculated. Thus, the features obtained from healthy patients and

With a quite smooth algorithm, the obtained HRV curves from the healthy patients fluctuate around a constant value in the amplitude range. In the fourier domain, we get the results in

For the detection of QRS complex, finding many adaptive approaches are possible. From the output of Teager energy operator, 60% of the average of the magnitudes of previous detected consecutive n (<6) R-peaks can be used as the threshold value for the detection of current R-peak. Thus, it will be possible to detect signal with slowly changing magnitude threshold value will be adaptive to catch the magnitudes correctly. However, the signal loss due to the misplacement of electrodes on the surface of body cannot be recovered with this

Another approach in design of an automated system can be a template matching algorithm. The basic requirement for this algorithm is the selection of a good clean QRS complex. There are numerous methods to decide a clean signal. If the cross correlation between the signal and the selected QRS complex, we get normalized outputs peaking at the R-peaks. In order to improve the results, this operation can be applied before the TEO operator and adaptive thresholding. This approach will give a more precise and an automated result. This

adaptive structure. Actually, there is possibly no way to get the signal from nothing.

Normalized Cross Correlation Function = ∑ (temp� − temp �������) <sup>∗</sup> (ecg� − ecg ����) ���

���

��� ���

The critical point of this methodology is that the correct choice of the QRS template should be done automatically. Selections can lead to incorrect outputs with a wrong template, and

During a decrease in the respiratory frequency, physiological systems of human body also reduce heart rhythm in order to provide homeostasis. In this way, oxygenated blood to spill

�∑ (temp� − temp �������)� ∗ ∑ (ecg� − ecg ����) ��� �

���

(6)

operation is done by the normalized cross-correlation formulation given as:

this may lead other analyses converge to an erroneous point.

Fig. 9. Thermistor signal with apnea and corresponding FF values

Fig. 10. Thermistor signal without apnea and corresponding FF values

In the ECG signal without apnea, more R-waves is found and as a result of that phenomena, it is understood that power spectral density of HRV has shifted to HF area. Thus, FF values are closely related to apnea events over ECG recordings.

### **5. Conclusion**

Due to the fact that human physiology do not continuously respond in the same way to the same stimulus, both automated and adaptive analysis procedures over biosignals are limited. Thus, the sections, that have negative effect on variance and cause the algorithm to work inaccurately, are removed from the ECG signal and then the HRV analysis is done. When an automated system is requested, a general algorithm that can remove these corrupted fields from the ECG signal can be written at most. For this purpose, it is possible to find approaches in both time and frequency domains.

In detection and classification of sleep apneas, usage of support vector machines (SVM) instead of neural network based algorithms may be a better and more original approach. SVM, similar to neural networks, uses a part of the data for training purposes and the rest of data to testing purposes. On contrary, by its structure, it can present a more determined working strategy in contrast to neural networks.

It is possible to obtain a very good HRV curve with the help of an algorithm that does not miss R-peaks as much as possible. In order to investigate the frequency domain attributes of HRV, the frequency characteristics are obtained from the HRV after the fast fourier transform (FFT) of it calculated. Thus, the features obtained from healthy patients and patients with sleep apnea can be easily differentiated.

With a quite smooth algorithm, the obtained HRV curves from the healthy patients fluctuate around a constant value in the amplitude range. In the fourier domain, we get the results in discrete time series.

For the detection of QRS complex, finding many adaptive approaches are possible. From the output of Teager energy operator, 60% of the average of the magnitudes of previous detected consecutive n (<6) R-peaks can be used as the threshold value for the detection of current R-peak. Thus, it will be possible to detect signal with slowly changing magnitude threshold value will be adaptive to catch the magnitudes correctly. However, the signal loss due to the misplacement of electrodes on the surface of body cannot be recovered with this adaptive structure. Actually, there is possibly no way to get the signal from nothing.

Another approach in design of an automated system can be a template matching algorithm. The basic requirement for this algorithm is the selection of a good clean QRS complex. There are numerous methods to decide a clean signal. If the cross correlation between the signal and the selected QRS complex, we get normalized outputs peaking at the R-peaks. In order to improve the results, this operation can be applied before the TEO operator and adaptive thresholding. This approach will give a more precise and an automated result. This operation is done by the normalized cross-correlation formulation given as:

$$\text{Normalized CrossCorrelation Function} = \frac{\sum\_{l=0}^{\text{N}-1} (\text{temp}\_{l} - \overline{\text{temp}}) \* (\text{erg}\_{l} - \overline{\text{erg}})}{\sqrt{\sum\_{l=0}^{\text{N}-1} (\text{temp}\_{l} - \overline{\text{temp}})^{2} \* \sum\_{l=0}^{\text{N}-1} (\text{ecg}\_{l} - \overline{\text{ecg}})^{2}}} \tag{6}$$

temp= chosen template QRS complex

ecgi = windowing of raw ECG (size of temp)

ecg=raw data

226 Advances in Electrocardiograms – Clinical Applications

**Thermistor- 256 Hz Sampling**

0 50 100 150 200 250 300

**Form Factor**

**300 seconds (10 epochs)**

0 50 100 150 200 250 300

**300 seconds (10 epochs)**

**<sup>0</sup> <sup>50</sup> <sup>100</sup> <sup>150</sup> <sup>200</sup> <sup>250</sup> <sup>300</sup> -500**

**Form Factor**

**300 seconds (10 epoches)**

**Thermistor- 256 Hz Sampling**

Fig. 9. Thermistor signal with apnea and corresponding FF values

Fig. 10. Thermistor signal without apnea and corresponding FF values

are closely related to apnea events over ECG recordings.

**5. Conclusion** 


**Amplitude**

**FF values**

**0 500 1000**

**Amplitude**

**FF values**

In the ECG signal without apnea, more R-waves is found and as a result of that phenomena, it is understood that power spectral density of HRV has shifted to HF area. Thus, FF values

<sup>0</sup> <sup>100</sup> <sup>200</sup> <sup>300</sup> <sup>400</sup> <sup>500</sup> <sup>600</sup> <sup>0</sup>

**300 seconds (10 epoches)**

Due to the fact that human physiology do not continuously respond in the same way to the same stimulus, both automated and adaptive analysis procedures over biosignals are The critical point of this methodology is that the correct choice of the QRS template should be done automatically. Selections can lead to incorrect outputs with a wrong template, and this may lead other analyses converge to an erroneous point.

During a decrease in the respiratory frequency, physiological systems of human body also reduce heart rhythm in order to provide homeostasis. In this way, oxygenated blood to spill all the biological system is more efficient.

Automated Detection and Classification of

*Congress,* Aydin, Turkey, 2008

(January 1999), pp. 260-262, ISSN 1053–587X.

6, ( November 2009), pp. 1057-1067, ISSN 1089-7771.

Vol. 25, No. 4, August 2006, pp. 26-31, ISSN 0739-5175

Seoul National University PhD Thesis, Seoul, Korea Pehlivan, F. (2004). Biophysics, Hacettepe Tas Printing, ISBN 975-7731-45-5

*Biomed. Eng.,* Vol. 56, No. 12, ISSN 0018-9294.

2003), Lecture Notes

321-325, ISSN 0013-4694

pp.149-152, ISSN 1520-6149

2009), pp. 37-48, ISSN 1089-7771.

2002), pp.42-57, ISSN 0739-5175

0169-2607

Sleep Apnea Types Using Electrocardiogram (ECG) and Electroencephalogram (EEG) Features 229

Erdamar, A. (2007). *A model development for prediction of sleep apnea and stimulation of genoglossus muscle*, Hacettepe University PhD Thesis, Ankara, Turkey Erogul, O. (2008). Engineering Approaches in Sleep Studies, *Proceedings of 9th Sleep Medicine* 

Firat, H. (2003). Sleep apnea syndrome diagnosis (definition, physiopathology, clinic,

Gula,L.J.; Krahn, A.D.; Skanes, A.; Kathleen, A. F.; George, C.; Yee, R. and Klein G.J. (2003).

Hjorth, B. (1973). Physical significance of time domain descriptors in EEG analysis,

Kaiser, J. F., (1993). Some useful properties of teager's energy operators, *acoustics, speech and* 

Khandoker, A. H.; Palaniswami, M.; & Karmakar, C. K. (2009a), Automated scoring of

Khandoker, A. H.; Gubbi, J.; & Palaniswami, M. (2009b), Support vector machines for

Köhler, B-U.; Henning, C. & Orglmeister, R. (2002). The Principles of Software QRS

Liu, D.; Pang, Z.; Lloyd, S. R. (2008). Neural Network Method for Detection of Obstructive

Mendez, M. O.; Bianchi, A. M.; Matteucci, M.; Cerutti, S.; & Penzel, T. (2008) Sleep Apnea

Park, H. (2000). *Automated Sleep Stage Analysis Using Hybrid Rule-Based and Case-Based* 

*and Programs in Biomedicine,* Vol. 82, No. 3, (November 2005), pp.187-195, ISSN

diagnosis methods, scores), *Turkish Thoracic Society Second Winter School*, (January

Heart rate variability in obstructive sleep apnea: a prospective study and frequency domain analysis, *A.N.E, Vol. 8, No. 2,* (april 2003), pp. 144-149, ISSN 1542-474X Hamila, R.; Astola, J.; Alaya Cheikh, F.; Gabbouj, M.; & Renfors, M. (1999). Teager energy

and the ambiguity function*, IEEE Transactions on Signal Processing*, Vol. 47, No. 1,

*Electroencephalography and Clinical Neurophysiology*, Vol. 34, No. 3, (March 1973), pp.

*signal processing, ICASSP-93, IEEE International Conference*, Vol.3, Issue April 1993,

obstructive sleep apnea and hypopnea events using short-term electrocardiogram recordings, *IEEE Transactions On Information Technology In Biomedicine*, Vol. 13, No.

automated recognition of obstructive sleep apnea syndrome from ECG recordings, *IEEE Transactions On Information Technology In Biomedicine,* Vol. 13, No. 1, (January

Detection-Reviewing and Comparing Algorithms for Detecting this Important ECG Waveform, IEEE Engineering In Medicine And Biology, Vol. 21, No. 1, (January

Sleep Apnea and Narcolepsy Based on Pupil Size and EEG*, IEEE Transactions on Neural Networks*, Vol. 19, No. 2, (February 2008), pp. 308-318, ISSN 1045-9227. Malinowska, U.; Durka, P-J.; Blinowska K-J.; Szelenberger W. & Wakarow A. (2006). Micro

and Macrostructure of Sleep EEG, *IEEE Engineering Medicine and Biology Magazine,*

Screening by Autoregressive Models From a Single ECG Lead, *IEEE Trans. on* 

*Reasoning,* Interdisciplinary Program of Medical and Biological Engineering Major

A slowdown in heart rhythm causes observing the HRV power spectral density to shift to VLF-LF region. In the normal case, an opposite loop is provided, and the HRV power spectrum density increases towards HF band.

Within the scope of sleep apnea detection processes, it is needed to use statistical analysis softwares to make sense in the scientific manner from the results. For example, for two different groups of independent samples, it is possible to prove that they different from each other in statistical sense by t-test.

Pulse Transit Time (PTT), which will constitute the basis of the study, is thought to be a very strong microstructure for the detection of sleep apnea and obtained from ECG, EEG and PPG signals. PTT analyses can rarely be seen in the literature (Smith et al. 1999). PTT is the duration of arterial pulse pressure occurring during the instance of pumping of blood from aortic leaflet to peripherals (limbs). PTT can be detected by reference to R-peaks on ECG signals.

After the ventricle depolarization corresponding to a R-peaking, PTT can be taken as the time of transmission of blood pressure to the fingers. For this purpose, the PPG signal from the oxygen plethysmograph probe is needed. The PPG located inside the PSG system with the name PLETH provides this data. By looking at the average duration of PTT, we can make a correlation analysis on respiratory effort. As a result of this analysis, the features to detect sleep apnea can be obtained (Pitson et al., 1995).

The experimental studies show that HRV power spectral densities in patients with OSA and CSA shift towards the low frequency (VLF-LF) band while a dense nature in high frequency (HF) band is observed for normal patients. In this extent, the statistical analyses before and after uvulopalatofarengoplasti (UPPP) surgical operation, one of the methods used in treatment of sleep apnea, can be a good approach to measure the performance of the surgical operation. Another study conducted by our group is the evaluation of statistical analysis methods on the preoperative and postoperative sleep time and frequency domain parameters.

It is thought that a prediction to sleep apnea and an implementation of indirect diagnosis system to help the physician are possible by investigating the cases called microstructures, which are impossible to see on ECG signals and EEG sub-bands, in the 1-2 epochs before and after the parts that contain sleep apnea in the EEG and ECG signals. This approach can provide more specific and original results.

### **6. References**


A slowdown in heart rhythm causes observing the HRV power spectral density to shift to VLF-LF region. In the normal case, an opposite loop is provided, and the HRV power

Within the scope of sleep apnea detection processes, it is needed to use statistical analysis softwares to make sense in the scientific manner from the results. For example, for two different groups of independent samples, it is possible to prove that they different from each

Pulse Transit Time (PTT), which will constitute the basis of the study, is thought to be a very strong microstructure for the detection of sleep apnea and obtained from ECG, EEG and PPG signals. PTT analyses can rarely be seen in the literature (Smith et al. 1999). PTT is the duration of arterial pulse pressure occurring during the instance of pumping of blood from aortic leaflet to peripherals (limbs). PTT can be detected by reference to R-peaks on ECG

After the ventricle depolarization corresponding to a R-peaking, PTT can be taken as the time of transmission of blood pressure to the fingers. For this purpose, the PPG signal from the oxygen plethysmograph probe is needed. The PPG located inside the PSG system with the name PLETH provides this data. By looking at the average duration of PTT, we can make a correlation analysis on respiratory effort. As a result of this analysis, the features to

The experimental studies show that HRV power spectral densities in patients with OSA and CSA shift towards the low frequency (VLF-LF) band while a dense nature in high frequency (HF) band is observed for normal patients. In this extent, the statistical analyses before and after uvulopalatofarengoplasti (UPPP) surgical operation, one of the methods used in treatment of sleep apnea, can be a good approach to measure the performance of the surgical operation. Another study conducted by our group is the evaluation of statistical analysis methods on the preoperative and postoperative sleep time and frequency domain

It is thought that a prediction to sleep apnea and an implementation of indirect diagnosis system to help the physician are possible by investigating the cases called microstructures, which are impossible to see on ECG signals and EEG sub-bands, in the 1-2 epochs before and after the parts that contain sleep apnea in the EEG and ECG signals. This approach can

Aksahin, M.F. (2010). *Classification of Sleep Apnea Types*, Baskent University PhD Thesis,

Aydin, H.; Ozgen, F.; Yetkin, S. & Sütcügil, L. (2005). *Sleep and Respiratory Disorders in Sleep*,

Camm, A. J.; Malik, M.; Bigger, J. T.; Breithardt, G.; Cerutti, S.; Cohen, R. J.; Coumel, P.;

Chen, S-W.; Chen H-C. & Chen H-C. (2005). A real time QRS detection method based on

European Heart Journal, Vol. 17, No.3, (March 1996), pp.354-381

Fallen, E. L.; Kennedy, H. L.; Kleiger, R. E.; Lombardi, F.; Malliani, A.; Moss, A. J.; Rottman, J. N.; Schmidt, G.; Schwartz, P. J. & Singer, D. H. (1996*). Heart rate variability-Standards of measurement, physiological interpretation, and clinical use*,

moving- averaging incorporating with wavelet denosing, *Elsevier Computer Methods* 

spectrum density increases towards HF band.

detect sleep apnea can be obtained (Pitson et al., 1995).

provide more specific and original results.

GATA Printing House, Ankara, Turkey

Ankara, Turkey

other in statistical sense by t-test.

signals.

parameters.

**6. References** 

*and Programs in Biomedicine,* Vol. 82, No. 3, (November 2005), pp.187-195, ISSN 0169-2607


**13** 

*USA* 

**Low Heart Rate Variability in** 

*Department of Physiology & Biophysics, College of Medicine,* 

Richard M. Millis, Stanley P. Carlyle, Mark D. Hatcher and Vernon Bond

The autonomic nervous system coordinates a person's responsiveness to physiological and environmental stressors. Attenuated respiratory sinus arrhythmia (RSA) with small standard deviation of the normal-normal electrocardiogram RR intervals (SDNN) is found in subjects with low heart rate variability (HRV) by time domain analysis (Montano et al., 2009). Low HRV is also characterized by sympathovagal balance shifted toward sympathetic predominance observed as an increase in the low frequency/high frequency ratio by frequency domain (fast Fourier transform) spectral analysis (Montano et al., 2009). However, inter- and intra-individual spectral changes are highly variable (Taverner et al., 1996) and the physiological significance is not always clear. Paced breathing at 0.2 Hz normally shifts sympathovagal balance toward greater vagal and less sympathetic activity (Driscoll and Dicicco, 2000) because of increased tidal volume and/or minute ventilation (Pinna et al., 2006). Sympathovagal imbalance related to low HRV is a risk factor for hypertension and various other cardiovascular and non-cardiovascular diseases (Kuch et al., 2001; Montano et al., 2009). Epidemiological studies estimate that the prevalence of hypertension ranges from 8% in an urban adolescent population (Rabinowitz et al., 1993) to 43% in black physicians twenty-two years after medical school (Gillum, 1999). However, no studies have determined the incidence and behavioral significance of sympathovagal imbalance in healthy young adult African-Americans. Abnormal autonomic responsiveness to environmental stressors is thought to be an important factor in the evolution of essential hypertension (Fauvel et al., 1996; Mezzacappa et al., 2001; Lucini et al., 2002a; 2002b) and African-American males are a subpopulation at high risk for such hypertension (Gillum, 1979; Kaplan, 1994; Gillum, 1999). The present study was, therefore, designed to determine whether sympathovagal imbalance related to low HRV occurs in a population of healthy young adult African-American males and whether it is an indicator of abnormal responsiveness to environmental stressors.

The study protocol was approved by the Howard University Human Participants Institutional Review Board, and each subject provided informed consent. A study population of 52 healthy normotensive 18-26 year-old African-American male university

**1. Introduction** 

**2. Materials and methods** 

**2.1 Study participants and design** 

**Healthy Young Adult Males** 

*Howard University, Washington, DC,* 


### **Low Heart Rate Variability in Healthy Young Adult Males**

Richard M. Millis, Stanley P. Carlyle, Mark D. Hatcher and Vernon Bond *Department of Physiology & Biophysics, College of Medicine, Howard University, Washington, DC, USA* 

### **1. Introduction**

230 Advances in Electrocardiograms – Clinical Applications

Pitson, D.J., Sandell, A., Hout, R.V., Stradling, J.R., 1995. Use of Pulse Transit Time as a

Rangayyan, R. M. (2002). *Biomedical Signal Analysis: A Case-Study Approach*, IEEE Press, ISBN

Roche, F.; Gaspoz, J-M.; Fortune, I.; Minini, P.; Pichot, V.; Duverney, D.; Costes, F.; Lacour J-

Roche F.; Duverney D.; Gortune-Court I.; Pichot V.; Costes F.; Lacour J-R.; Antonladis A.;

*Electrophysilogy*, Vol. 25, No.8, (August 2002), pp.1192-1199, 10504-0418 Smith, P.R.; Argod, J.; Pepin, J.L. & Levy, P.A. (1999). Pulse Transit Time: An Appraisal of

Stoica, P. & Moses , R.L. (1997), *Introduction to spectral analysis*, Prentice Hall*,* ISBN: 0-13-

Sugi, T.; Kawana, F. & Nakamura M. (2009). Automatic EEG Arousals Detection for

Xu, W., & Liu, X. (2009), Sleep Apnea Assessment by ECG Pattern, *2nd International Congress* 

*Respiratory Journal*, Vol.8, No.10, (October 1995), pp.1669-1674.

*Circulation*,Vol. 100, (June 1999), pp.1411-1415, 1524-4539

0-471-20811-6, USA

258419-0, New Jersey, U.S.

China, October 2009.

No. 4, (October 2009), pp. 329-337

– 457,

Measure of Inspiratory Effort in Patients with Obstructive Sleep Apnoea, *European* 

R. & Barthelemy, J-C. (1999). Screening of Obstructive Sleep Apnea Syndrome by Heart Rate Variability Analysis, *Journal of the American Heart Assocation-*

Gaspoz J-M. & Barthelemy J-C. (2002). Cardiac Interbeat Interval Increment for the Identication of Obstructive Sleep Apnea, *Journal of Pacing and Clinical* 

Potential Cliniccal Applications, *Journal of Throax*, Vol.54, No.5, (May 1999), pp.452

Obstructive Sleep Apnea Syndrome, *Biomedical Signal Processing and Control,* Vol. 4,

*on Image and Signal Processing, 2009. CISP '09*, ISBN: 978-1-4244-4129-7, Tianjin,

The autonomic nervous system coordinates a person's responsiveness to physiological and environmental stressors. Attenuated respiratory sinus arrhythmia (RSA) with small standard deviation of the normal-normal electrocardiogram RR intervals (SDNN) is found in subjects with low heart rate variability (HRV) by time domain analysis (Montano et al., 2009). Low HRV is also characterized by sympathovagal balance shifted toward sympathetic predominance observed as an increase in the low frequency/high frequency ratio by frequency domain (fast Fourier transform) spectral analysis (Montano et al., 2009). However, inter- and intra-individual spectral changes are highly variable (Taverner et al., 1996) and the physiological significance is not always clear. Paced breathing at 0.2 Hz normally shifts sympathovagal balance toward greater vagal and less sympathetic activity (Driscoll and Dicicco, 2000) because of increased tidal volume and/or minute ventilation (Pinna et al., 2006). Sympathovagal imbalance related to low HRV is a risk factor for hypertension and various other cardiovascular and non-cardiovascular diseases (Kuch et al., 2001; Montano et al., 2009). Epidemiological studies estimate that the prevalence of hypertension ranges from 8% in an urban adolescent population (Rabinowitz et al., 1993) to 43% in black physicians twenty-two years after medical school (Gillum, 1999). However, no studies have determined the incidence and behavioral significance of sympathovagal imbalance in healthy young adult African-Americans. Abnormal autonomic responsiveness to environmental stressors is thought to be an important factor in the evolution of essential hypertension (Fauvel et al., 1996; Mezzacappa et al., 2001; Lucini et al., 2002a; 2002b) and African-American males are a subpopulation at high risk for such hypertension (Gillum, 1979; Kaplan, 1994; Gillum, 1999). The present study was, therefore, designed to determine whether sympathovagal imbalance related to low HRV occurs in a population of healthy young adult African-American males and whether it is an indicator of abnormal responsiveness to environmental stressors.

### **2. Materials and methods**

#### **2.1 Study participants and design**

The study protocol was approved by the Howard University Human Participants Institutional Review Board, and each subject provided informed consent. A study population of 52 healthy normotensive 18-26 year-old African-American male university

Low Heart Rate Variability in Healthy Young Adult Males 233

Fig. 1. **Electrocardiogram RR intervals for broadband and narrowband subjects.**

the ordinate and time (s) on the abscissa for a representative subject exhibiting

group (Bottom Panel).

Representative cardiotachogram tracings showing electrocardiogram RR intervals (ms) on

characteristics for assignment to the broadband group (Top Panel) and to the narrowband

students was included in the study. Criteria for inclusion in the study were: non-smoking status, absence of alcohol abuse (less than two standard alcohol drinks a day), absence of use of medication that could interfere with autonomic modulation, resting systolic/diastolic blood pressure <140/90 mm Hg and body mass index less than 28 kg m-1.

### **2.1.1 Paced breathing**

After entering the laboratory subjects were instrumented and instructed as to the experimental procedures. Subjects breathed normally while seated and at rest and 5 min of this resting data was recorded. Following the normal breathing protocol subjects were instructed to perform 5 min of paced breathing in such a manner that each respiratory cycle was 5 s in duration or 12 breaths per min (0.2 Hz). The subject observed a visual tracking image on a computer monitor for periodic durations of inspirations and expirations. Each subject practiced paced breathing for a period of 1 min and was then instructed to perform the paced respiration for 5 min during which time the electrocardiogram signal was recorded using a Biopac MP100 data acquisition system (Biopac Systems, Santa Barbara, CA). The electrocardiogram electrodes were placed on the subject's chest in a standard three-lead position with recordings obtained from lead II.

### **2.1.2 Group assignment**

Two groups of subjects were a priori classified as either "broadband" or "narrowband". The criterion for the "narrowband" group was that the maximum SDNN value, a timedomain marker of vagal modulation, was more than one standard deviation below the mean SDNN for the study population during the 5 min trial of paced breathing. The "broadband" group, thereby, consisted of the subjects exhibiting SDNN one standard deviation less than that of the mean SDNN of the study population. Figure 1 shows cardiotachogram tracings of representative broadband and narrowband subjects.

### **2.1.3 Heart rate variability analyses**

Heart rate was measured in beats · min-1 and vagal modulation of HRV in the time domain was measured as the standard deviation of all normal-to-normal standard electrocardiogram inter-beat intervals (SDNN). Time domain HRV, measured as standard deviation of the RR intervals, was expressed in ms and was computed using data acquisition and analysis software specifically designed to measure HRV in the time and frequency domain (Nevrokard, Version 6.3.1, Ljubljana, Slovenia). All time domain HRV data were reported herein as mean SDNN ± standard error. Fast Fourier transform analysis of the electrocardiogram RR intervals was used to spectrally decompose HRV in the frequency domain. For the frequency domain analysis, vagal modulation was represented by the area under the high-frequency power spectrum (HF: 0.14-0.4 Hz) expressed as the power in ms2. We also included the LF/HF ratio of heart rate variability as a measure of sympathetic modulation according to the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996). However, this concept has received great criticism (Eckberg, 1999). All time and frequency domain analyses were carried out in accordance with the guidelines put forth by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996).

students was included in the study. Criteria for inclusion in the study were: non-smoking status, absence of alcohol abuse (less than two standard alcohol drinks a day), absence of use of medication that could interfere with autonomic modulation, resting systolic/diastolic

After entering the laboratory subjects were instrumented and instructed as to the experimental procedures. Subjects breathed normally while seated and at rest and 5 min of this resting data was recorded. Following the normal breathing protocol subjects were instructed to perform 5 min of paced breathing in such a manner that each respiratory cycle was 5 s in duration or 12 breaths per min (0.2 Hz). The subject observed a visual tracking image on a computer monitor for periodic durations of inspirations and expirations. Each subject practiced paced breathing for a period of 1 min and was then instructed to perform the paced respiration for 5 min during which time the electrocardiogram signal was recorded using a Biopac MP100 data acquisition system (Biopac Systems, Santa Barbara, CA). The electrocardiogram electrodes were placed on the subject's chest in a standard

Two groups of subjects were a priori classified as either "broadband" or "narrowband". The criterion for the "narrowband" group was that the maximum SDNN value, a timedomain marker of vagal modulation, was more than one standard deviation below the mean SDNN for the study population during the 5 min trial of paced breathing. The "broadband" group, thereby, consisted of the subjects exhibiting SDNN one standard deviation less than that of the mean SDNN of the study population. Figure 1 shows cardiotachogram tracings of

Heart rate was measured in beats · min-1 and vagal modulation of HRV in the time domain was measured as the standard deviation of all normal-to-normal standard electrocardiogram inter-beat intervals (SDNN). Time domain HRV, measured as standard deviation of the RR intervals, was expressed in ms and was computed using data acquisition and analysis software specifically designed to measure HRV in the time and frequency domain (Nevrokard, Version 6.3.1, Ljubljana, Slovenia). All time domain HRV data were reported herein as mean SDNN ± standard error. Fast Fourier transform analysis of the electrocardiogram RR intervals was used to spectrally decompose HRV in the frequency domain. For the frequency domain analysis, vagal modulation was represented by the area under the high-frequency power spectrum (HF: 0.14-0.4 Hz) expressed as the power in ms2. We also included the LF/HF ratio of heart rate variability as a measure of sympathetic modulation according to the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology (1996). However, this concept has received great criticism (Eckberg, 1999). All time and frequency domain analyses were carried out in accordance with the guidelines put forth by the Task Force of the European Society of Cardiology and the North American Society of Pacing and Electrophysiology

blood pressure <140/90 mm Hg and body mass index less than 28 kg m-1.

three-lead position with recordings obtained from lead II.

representative broadband and narrowband subjects.

**2.1.3 Heart rate variability analyses** 

**2.1.1 Paced breathing** 

**2.1.2 Group assignment**

(1996).

### Fig. 1. **Electrocardiogram RR intervals for broadband and narrowband subjects.**

Representative cardiotachogram tracings showing electrocardiogram RR intervals (ms) on the ordinate and time (s) on the abscissa for a representative subject exhibiting characteristics for assignment to the broadband group (Top Panel) and to the narrowband group (Bottom Panel).

Low Heart Rate Variability in Healthy Young Adult Males 235

The heart rates of the study population measured during the normal spontaneous breathing condition were found to be lower than those measured during the paced breathing control condition (67.9 ± 1.8 vs. 70.1 ± 1.7, P=0.01); this difference was extremely small (3%) and, therefore, not of physiological significance. On the other hand, the heart rates measured during the Stroop word color conflict and cold pressor testing were 16% higher (81.3 ± 2.3 and 81.3 ± 2.3 vs. 70.1 ± 1.7, P<0.01, respectively) and the heart rates during exercise at 30%- 50% of peak oxygen consumption were 34%-77% higher (94.5 ± 2.1 and 124.6 ± 3.5 vs. 70.1 ± 1.7, P<0.01, respectively) than those measured during the paced breathing control condition;

The study group of 52 subjects was characterized by a maximum SDNN of 249.70 ms, a minimum SDNN of 39.15 ms and a mean ± standard deviation of 108.40 ± 40.95 ms during the paced breathing control condition. A subgroup of subjects met the criterion for low HRV (6/52, 12%) and, by experimental design, exhibited a very small SDNN, more than one standard deviation below the mean of the study population, during paced breathing at a fixed frequency (0.2 Hz). This subgroup was differentiated from the larger study group by a minimum SDNN of 39.15 ms, a maximum of 64.55 and a mean ± standard deviation of 49.40

Figure 1 demonstrates the range of SDNN values measured in this study and shows a significantly greater SDNN of the study population for the paced breathing control condition than for the two conditions of exercise at 30% and 50% of peak oxygen consumption (P<0.0001). SDNN measured during the periods of normal spontaneous breathing, at 0.18-0.48 Hz while at rest sitting in a chair and on a cycle ergometer, and Stroop word color conflict testing (mental stress) were smaller than the SDNN measured during the paced breathing control period and intermediate between the largest SDNN during paced breathing and the smallest SDNN measured measured during aerobic exercise (P<0.001). The SDNN measured during cold pressor testing (cold stress) was not

Figure 2 shows that the SDNN of the study group during the control paced breathing was significantly greater than that during the uncontrolled spontaneous breathing, both measured while sitting in a chair (P<0.05). Figure 3 demonstrates that the group of 6 subjects ("narrowband" group) found to have significantly smaller SDNN of the remaining group of 46 subjects ("broadband" group) during normal spontaneous breathing (0.18-0.4 Hz), both measured at rest while sitting in a chair (P<0.05). Figure 4 shows that, between the normal spontaneous breathing and paced breathing conditions, the SDNN of this "low HRV, narrow heart rate bandwidth" or "narrowband" group of 6 subjects increased by 1.58 ± 13.59%; whereas, that of the remaining group of 46 "normal HRV, broad heart rate bandwidth" subjects ("broadband" group) increased by 30.41 ± 2.29% (P=0.0006) during

Figure 5 presents a comparison of the SDNN of the broad and narrow heart rate bandwidth groups across all of the physiological states studied. By experimental design, the SDNN measured during the paced breathing control condition differentiated between the narrow and broad heart rate bandwidth groups which were also found to be differentiated by the SDNN measured during the normal spontaneous breathing state. These groups were not differentiated by the SDNN measured during the experimental conditions of Stroop word color conflict testing (mental stress), cold pressor testing (cold stress) and exercise at 30%

these differences were considered to be physiologically significant.

significantly different that that during the paced breathing control (P>0.1).

± 7.15 ms (P<0.0001).

paced breathing at 0.2 Hz.

and 50% of peak oxygen consumption.

### **2.1.4 Aerobic capacity (VO2peak)**

The functional work capacity measure of VO2peak was performed during a graded exercise treadmill test using the Bruce protocol with stage 1 beginning at an incline of 10% at a speed of 1.7 mph for 3 min. After stage 1, the treadmill incline was increased by 2% and the speed by 0.8 mph every 3 min until voluntary fatigue. Respiratory measures of ventilation and gas fractions of oxygen and carbon dioxide were performed using a Physio-Dyne Max I metabolic system (Physio-Dyne Inc., Quogue, NY). VO2peak was defined as the VO2 value achieved during the last min of the graded treadmill exercise test. Before the graded exercise test of VO2peak, the metabolic system was calibrated with known gas concentrations of oxygen and carbon dioxide. The participants performed the VO2peak test during the first laboratory visit at the beginning of the study.

### **2.1.5 Paced and uncontrolled spontaneous breathing conditions**

The subjects were trained to breathe at 0.2 Hz by following an analog signal generated by a computer for timing the inspiratory and expiratory phases of respiration. Measurements of heart rate variability made during paced and uncontrolled spontaneous breathing were performed while sitting in a chair.

### **2.1.6 Mental and nociceptive stress conditions**

Mental stress was produced by Stroop word-color conflict testing and physical stress by cold pressor testing using single foot immersion in an ice bath.

### **2.2 Statistical analysis**

The significance of differences in SDNN, the absolute HF power, the LF/HF power was determined by comparing the paced breathing control to the spontaneous breathing condition using Student's t-test for paired samples. Linear regression analysis was used to verify the correlation between SDNN and HF power. The significance of intergroup ("broadband" vs. "narrowband") differences was determined using Student's t-test for independent samples. Probability for all analyses was set at P<0.05. Significance of differences in SDNN across testing conditions was evaluated by analysis of variance (ANOVA), with significance at P<0.05.

### **3. Results**

Table 1 presents the physiological characteristics of the study population showing that it consisted of a healthy group of 18-26 year-old healthy male African-American university students measured during the paced breathing control condition.


\* Mean ± standard deviation

Table 1. Physiological characteristics of the study population\*

The functional work capacity measure of VO2peak was performed during a graded exercise treadmill test using the Bruce protocol with stage 1 beginning at an incline of 10% at a speed of 1.7 mph for 3 min. After stage 1, the treadmill incline was increased by 2% and the speed by 0.8 mph every 3 min until voluntary fatigue. Respiratory measures of ventilation and gas fractions of oxygen and carbon dioxide were performed using a Physio-Dyne Max I metabolic system (Physio-Dyne Inc., Quogue, NY). VO2peak was defined as the VO2 value achieved during the last min of the graded treadmill exercise test. Before the graded exercise test of VO2peak, the metabolic system was calibrated with known gas concentrations of oxygen and carbon dioxide. The participants performed the VO2peak test during the first

The subjects were trained to breathe at 0.2 Hz by following an analog signal generated by a computer for timing the inspiratory and expiratory phases of respiration. Measurements of heart rate variability made during paced and uncontrolled spontaneous breathing were

Mental stress was produced by Stroop word-color conflict testing and physical stress by

The significance of differences in SDNN, the absolute HF power, the LF/HF power was determined by comparing the paced breathing control to the spontaneous breathing condition using Student's t-test for paired samples. Linear regression analysis was used to verify the correlation between SDNN and HF power. The significance of intergroup ("broadband" vs. "narrowband") differences was determined using Student's t-test for independent samples. Probability for all analyses was set at P<0.05. Significance of differences in SDNN across testing conditions was evaluated by analysis of variance

Table 1 presents the physiological characteristics of the study population showing that it consisted of a healthy group of 18-26 year-old healthy male African-American university

**2.1.4 Aerobic capacity (VO2peak)**

laboratory visit at the beginning of the study.

**2.1.6 Mental and nociceptive stress conditions** 

cold pressor testing using single foot immersion in an ice bath.

students measured during the paced breathing control condition.

Table 1. Physiological characteristics of the study population\*

Age (years) 20.92 ± 2.48 Weight (kg) 76.03 ± 10.89 Height (cm) 179.44 ± 7.06 Resting Systolic Blood Pressure (mm Hg) 130.08 ± 8.64 Resting Diastolic Blood Pressure (mm Hg) 76.36 ± 9.33 Resting Heart Rate (beats· min-1) 75.48 ± 9.65 Peak Oxygen Consumption (mL· kg-1· min-1) 35.30 ± 7.28

performed while sitting in a chair.

(ANOVA), with significance at P<0.05.

**2.2 Statistical analysis** 

\* Mean ± standard deviation

**3. Results** 

**2.1.5 Paced and uncontrolled spontaneous breathing conditions** 

The heart rates of the study population measured during the normal spontaneous breathing condition were found to be lower than those measured during the paced breathing control condition (67.9 ± 1.8 vs. 70.1 ± 1.7, P=0.01); this difference was extremely small (3%) and, therefore, not of physiological significance. On the other hand, the heart rates measured during the Stroop word color conflict and cold pressor testing were 16% higher (81.3 ± 2.3 and 81.3 ± 2.3 vs. 70.1 ± 1.7, P<0.01, respectively) and the heart rates during exercise at 30%- 50% of peak oxygen consumption were 34%-77% higher (94.5 ± 2.1 and 124.6 ± 3.5 vs. 70.1 ± 1.7, P<0.01, respectively) than those measured during the paced breathing control condition; these differences were considered to be physiologically significant.

The study group of 52 subjects was characterized by a maximum SDNN of 249.70 ms, a minimum SDNN of 39.15 ms and a mean ± standard deviation of 108.40 ± 40.95 ms during the paced breathing control condition. A subgroup of subjects met the criterion for low HRV (6/52, 12%) and, by experimental design, exhibited a very small SDNN, more than one standard deviation below the mean of the study population, during paced breathing at a fixed frequency (0.2 Hz). This subgroup was differentiated from the larger study group by a minimum SDNN of 39.15 ms, a maximum of 64.55 and a mean ± standard deviation of 49.40 ± 7.15 ms (P<0.0001).

Figure 1 demonstrates the range of SDNN values measured in this study and shows a significantly greater SDNN of the study population for the paced breathing control condition than for the two conditions of exercise at 30% and 50% of peak oxygen consumption (P<0.0001). SDNN measured during the periods of normal spontaneous breathing, at 0.18-0.48 Hz while at rest sitting in a chair and on a cycle ergometer, and Stroop word color conflict testing (mental stress) were smaller than the SDNN measured during the paced breathing control period and intermediate between the largest SDNN during paced breathing and the smallest SDNN measured measured during aerobic exercise (P<0.001). The SDNN measured during cold pressor testing (cold stress) was not significantly different that that during the paced breathing control (P>0.1).

Figure 2 shows that the SDNN of the study group during the control paced breathing was significantly greater than that during the uncontrolled spontaneous breathing, both measured while sitting in a chair (P<0.05). Figure 3 demonstrates that the group of 6 subjects ("narrowband" group) found to have significantly smaller SDNN of the remaining group of 46 subjects ("broadband" group) during normal spontaneous breathing (0.18-0.4 Hz), both measured at rest while sitting in a chair (P<0.05). Figure 4 shows that, between the normal spontaneous breathing and paced breathing conditions, the SDNN of this "low HRV, narrow heart rate bandwidth" or "narrowband" group of 6 subjects increased by 1.58 ± 13.59%; whereas, that of the remaining group of 46 "normal HRV, broad heart rate bandwidth" subjects ("broadband" group) increased by 30.41 ± 2.29% (P=0.0006) during paced breathing at 0.2 Hz.

Figure 5 presents a comparison of the SDNN of the broad and narrow heart rate bandwidth groups across all of the physiological states studied. By experimental design, the SDNN measured during the paced breathing control condition differentiated between the narrow and broad heart rate bandwidth groups which were also found to be differentiated by the SDNN measured during the normal spontaneous breathing state. These groups were not differentiated by the SDNN measured during the experimental conditions of Stroop word color conflict testing (mental stress), cold pressor testing (cold stress) and exercise at 30% and 50% of peak oxygen consumption.

Low Heart Rate Variability in Healthy Young Adult Males 237

Fig. 3. **Heart rate variability for paced and uncontrolled breathing conditions.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace) and normal uncontrolled spontaneous breathing at rest sitting in a chair (rest). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. The rest condition is different from paced breathing control (pace) at P<0.05. Figure 6 shows that the LF/HF ratio of HRV spectral power, a measure of cardiac sympathovagal balance, was significantly higher during paced breathing at 0.2 Hz than during uncontrolled spontaneous breathing for the study group of 52 subjects (P<0.01) and, during paced breathing at 0.2 Hz, was significantly higher for the narrowband band group

**PACE REST**

Humans vary in their responsiveness to environmental stressors and HRV measurements may serve as physiological markers for such variation (DeBecker et al., 1998). HRV in the time domain, measured as standard deviation or standard error, estimates the range of differences in the time intervals between heartbeats (Lucini et al., 2002c). HRV has been used as a measure of autonomic balance that emanates from endogenous sympathetic and parasympathetic rhythms which are partly modulated by respiratory sinus arrhythmia (Hayano et al., 1990; Hrushesky, 1991). Fast Fourier transform (FFT) analysis of HRV differentiates the frequency components of the heart's inter-beat intervals and yields more detailed information about autonomic tone than time domain analysis (Petretta et al., 1995). Such analysis makes it possible to differentiate conditions with physiological features in common. For example, common autonomic contributions to HRV have been evaluated by examining specific FFT frequency bands. This frequency domain analysis is useful under diverse physiological states such as dynamic exercise (Pichon et al., 2004), hypertension associated with sleep apnea (Vanninen et al., 1996; Narkiewicz et al., 1998; Salo et al., 2000) and optic neuropathy (Gutierrez et al., 2002). In this study, we found that healthy young adult African-Americans exhibit greater SDNN, a time domain measure of HRV, during conditions of paced breathing and cold stress than during conditions of normal breathing

of 6 subjects than for the broadband group of 46 subjects (P<0.05).

**0**

**25**

**50**

**75**

**SDNN (ms)**

**100**

**125**

**4. Discussion** 

**ALL SUBJECTS**

 **STATISTICAL SIGNIFICANCE TEST CONDITIONS**

Fig. 2. **Heart rate variability across various testing conditions.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace), normal uncontrolled spontaneous breathing at rest sitting in a chair (rest), Stroop word-color conflict testing of mental stress (mental), cold pressor testing of nociceptive stress (cold), normal uncontrolled spontaneous breathing at rest sitting on a cycle ergometer (rest cycle) and aerobic exercise stress at 30% and 50% of peak oxygen consumption (30% VO2peak, 50% VO2peak). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. ♦ SDNN different from paced breathing control (pace) at P<0.05.

The "broadband" group's total spectral power was significantly greater than the "narrowband" group's total HRV spectral power for paced but not for spontaneous breathing (16,600 ± 1,842 vs. 2,858 ± 176 ms2, P=0.01 and 7,807 ± 1,224 vs. 2,106 ± 424 ms2, P=0.1, respectively). The "broadband" versus "narrowband" intergroup difference in absolute LF power was not significant for spontaneous breathing. The "broadband" group's absolute LF power was significantly higher than the "narrowband" group's absolute LF power (12,830 ± 1,461 vs. 1,981 ± 296 ms2, P=0.01) for paced breathing. The "broadband" group's absolute HF power was not significantly different than the "narrowband" group's absolute HF power for both paced and spontaneous breathing (P=0.12). The "broadband" versus "narrowband" intergroup difference in the percentages of total LF and HF power was also not significant.

PACE

was also not significant.

0

10

20

30

40

50

60

**HRV(ms)**

70

80

90

100

110

120

REST

MENTAL

COLD

 **STATISTICAL SIGNIFICANCE TEST CONDITIONS**

The "broadband" group's total spectral power was significantly greater than the "narrowband" group's total HRV spectral power for paced but not for spontaneous breathing (16,600 ± 1,842 vs. 2,858 ± 176 ms2, P=0.01 and 7,807 ± 1,224 vs. 2,106 ± 424 ms2, P=0.1, respectively). The "broadband" versus "narrowband" intergroup difference in absolute LF power was not significant for spontaneous breathing. The "broadband" group's absolute LF power was significantly higher than the "narrowband" group's absolute LF power (12,830 ± 1,461 vs. 1,981 ± 296 ms2, P=0.01) for paced breathing. The "broadband" group's absolute HF power was not significantly different than the "narrowband" group's absolute HF power for both paced and spontaneous breathing (P=0.12). The "broadband" versus "narrowband" intergroup difference in the percentages of total LF and HF power

Fig. 2. **Heart rate variability across various testing conditions.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace), normal uncontrolled spontaneous breathing at rest sitting in a chair (rest), Stroop word-color conflict testing of mental stress (mental), cold pressor testing of nociceptive stress (cold), normal uncontrolled spontaneous breathing at rest sitting on a cycle ergometer (rest cycle) and aerobic exercise stress at 30% and 50% of peak oxygen consumption (30% VO2peak, 50% VO2peak). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. ♦ SDNN different from paced breathing control (pace) at P<0.05.

REST CYCLE

30%VO2PEAK

50%VO2PEAK

**ALL SUBJECTS**

Fig. 3. **Heart rate variability for paced and uncontrolled breathing conditions.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace) and normal uncontrolled spontaneous breathing at rest sitting in a chair (rest). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. The rest condition is different from paced breathing control (pace) at P<0.05.

Figure 6 shows that the LF/HF ratio of HRV spectral power, a measure of cardiac sympathovagal balance, was significantly higher during paced breathing at 0.2 Hz than during uncontrolled spontaneous breathing for the study group of 52 subjects (P<0.01) and, during paced breathing at 0.2 Hz, was significantly higher for the narrowband band group of 6 subjects than for the broadband group of 46 subjects (P<0.05).

### **4. Discussion**

Humans vary in their responsiveness to environmental stressors and HRV measurements may serve as physiological markers for such variation (DeBecker et al., 1998). HRV in the time domain, measured as standard deviation or standard error, estimates the range of differences in the time intervals between heartbeats (Lucini et al., 2002c). HRV has been used as a measure of autonomic balance that emanates from endogenous sympathetic and parasympathetic rhythms which are partly modulated by respiratory sinus arrhythmia (Hayano et al., 1990; Hrushesky, 1991). Fast Fourier transform (FFT) analysis of HRV differentiates the frequency components of the heart's inter-beat intervals and yields more detailed information about autonomic tone than time domain analysis (Petretta et al., 1995). Such analysis makes it possible to differentiate conditions with physiological features in common. For example, common autonomic contributions to HRV have been evaluated by examining specific FFT frequency bands. This frequency domain analysis is useful under diverse physiological states such as dynamic exercise (Pichon et al., 2004), hypertension associated with sleep apnea (Vanninen et al., 1996; Narkiewicz et al., 1998; Salo et al., 2000) and optic neuropathy (Gutierrez et al., 2002). In this study, we found that healthy young adult African-Americans exhibit greater SDNN, a time domain measure of HRV, during conditions of paced breathing and cold stress than during conditions of normal breathing

Low Heart Rate Variability in Healthy Young Adult Males 239

Fig. 5. **Percent change in heart rate variability for broadband and narrowband groups.** Bars represent percent increase in heart rate variability expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz. Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing less than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. The percent increase in SDNN of the BB group is different from that of the NB group at P<0.05.

**BB NB**

Respiratory sinus arrhythmia is thought to be the main source of HRV. However, there may be other non-autonomic contributions to respiratory sinus arrhythmia and to the high frequency (HF) components of HRV which may distort the signal-to-noise ratio and estimates of capacity for vagal modulation of heart rate (Pichon et al., 2004). HF HRV may mostly reflect noise if breathing shifts a substantial amount of HRV power to the low frequency (LF) range. The low time domain HRV occurring in subjects with apnea syndromes (Vanninen et al., 1996; Narkiewicz et al., 1998; Salo et al., 2000) could also be an effect of respiratory rate and/or tidal volume (Pinna et al., 2006). Low time domain HRV has been found in subjects exhibiting pre-hypertensive (Lucini et al., 2002a) and obesity (Salo et al., 2000) risk factors. Acetylcholine, when released from the vagus nerve, appears to act synergistically with vasoactive intestinal peptide to increase respiratory sinus arrhythmia (Markos and Snow, 2001). Higher HRV spectral frequencies and greater inter-beat intervals have been associated with a high capacity for vagal modulation of heart rate which occurs in normotensive healthy adults breathing at rest (Gutierrez et al., 2002). The peak HRV spectral frequency occurs in the range of HF in normotensive healthy adults breathing at rest and shifts to the range of LF during periods of exercise and stress, as well as, during disease

Because of the respiration-related variability of electrocardiogram inter-beat (RR) intervals, the necessity of controlling respiratory frequency during measurements of HRV has been demonstrated (De Meersman et al., 1995). We used the frequency of 0.2 Hz during the paced breathing trial because this frequency produced the most reproducible conditions across subjects. As expected, the time domain HRV during paced breathing was significantly greater than during spontaneous breathing in the same subjects. This breathing pattern was

states such as hypertension (Murakami et al., 1996)

**0**

**5**

**10**

**15**

**20**

**PERCENT CHANGE**

**25**

**30**

**35**

Fig. 4. **Heart rate variability for broadband and narrowband groups.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (Bottom Panel) and normal uncontrolled spontaneous breathing at rest sitting in a chair (Top Panel). Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. The rest condition is different from paced breathing control (pace) at P<0.05.

and mental stress. We also identified a small subpopulation (12%) of subjects exhibiting SDNN more than one SD below the mean SDNN of the larger study population of 52 males during the paced breathing condition. It was beyond the scope of this part of the study to measure HRV spectral power. Because of the high variability of frequency domain measurements during short time intervals (Taverner et al., 1996), we limited a part of this study to the time domain measurement of SDNN.

Fig. 4. **Heart rate variability for broadband and narrowband groups.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (Bottom Panel) and normal uncontrolled spontaneous breathing at rest sitting in a chair (Top Panel). Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. The

**BB NB**

**BB NB**

and mental stress. We also identified a small subpopulation (12%) of subjects exhibiting SDNN more than one SD below the mean SDNN of the larger study population of 52 males during the paced breathing condition. It was beyond the scope of this part of the study to measure HRV spectral power. Because of the high variability of frequency domain measurements during short time intervals (Taverner et al., 1996), we limited a part of this

rest condition is different from paced breathing control (pace) at P<0.05.

study to the time domain measurement of SDNN.

**0**

**25**

**50**

**75**

**SDNN (ms)**

**100**

**125**

**SDNN (m**

 **s)**

Fig. 5. **Percent change in heart rate variability for broadband and narrowband groups.** Bars represent percent increase in heart rate variability expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz. Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing less than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. The percent increase in SDNN of the BB group is different from that of the NB group at P<0.05.

Respiratory sinus arrhythmia is thought to be the main source of HRV. However, there may be other non-autonomic contributions to respiratory sinus arrhythmia and to the high frequency (HF) components of HRV which may distort the signal-to-noise ratio and estimates of capacity for vagal modulation of heart rate (Pichon et al., 2004). HF HRV may mostly reflect noise if breathing shifts a substantial amount of HRV power to the low frequency (LF) range. The low time domain HRV occurring in subjects with apnea syndromes (Vanninen et al., 1996; Narkiewicz et al., 1998; Salo et al., 2000) could also be an effect of respiratory rate and/or tidal volume (Pinna et al., 2006). Low time domain HRV has been found in subjects exhibiting pre-hypertensive (Lucini et al., 2002a) and obesity (Salo et al., 2000) risk factors. Acetylcholine, when released from the vagus nerve, appears to act synergistically with vasoactive intestinal peptide to increase respiratory sinus arrhythmia (Markos and Snow, 2001). Higher HRV spectral frequencies and greater inter-beat intervals have been associated with a high capacity for vagal modulation of heart rate which occurs in normotensive healthy adults breathing at rest (Gutierrez et al., 2002). The peak HRV spectral frequency occurs in the range of HF in normotensive healthy adults breathing at rest and shifts to the range of LF during periods of exercise and stress, as well as, during disease states such as hypertension (Murakami et al., 1996)

Because of the respiration-related variability of electrocardiogram inter-beat (RR) intervals, the necessity of controlling respiratory frequency during measurements of HRV has been demonstrated (De Meersman et al., 1995). We used the frequency of 0.2 Hz during the paced breathing trial because this frequency produced the most reproducible conditions across subjects. As expected, the time domain HRV during paced breathing was significantly greater than during spontaneous breathing in the same subjects. This breathing pattern was

Low Heart Rate Variability in Healthy Young Adult Males 241

under technical supervision and in the low range of normal respiratory rate. Breathing in the range of 3-9 breaths per minute (0.05-0.15 Hz) may produce higher amplitude respiratory sinus arrhythmia because of more complete acetylcholine metabolism during exhalation (Song and Lehrer, 2003). Respiratory sinus arrhythmia might also be maximized if subjects breathe at frequencies controlled by other physiological processes. For example, when breathing between 4-7 breaths per minute (0.07-0.12 Hz), the baroreceptor reflex might be stimulated; thereby, causing a resonance effect and an increase in HRV (Vaschillo et al., 2002). In the present study, paced breathing was associated with significantly lower SDNN (0.2 Hz) than the spontaneous breathing trials (0.18-0.48 Hz) and we did not measure tidal volume which has been shown to positively modulate the high frequency (vagal) component of HRV spectral power (Pinna et al., 2006). At a high respiratory frequency paced breathing has been shown to produce predominance of the HF power of HRV in normal healthy adult subjects during meditation (Cysarz and Bussing, 2005). HRV studied during paced breathing in subgroups of pre-hypertensive and hypertensive middle-aged men and women has been shown to produce greater HF power in the pre-hypertensive than

HRV in a subgroup of healthy normotensive middle-aged persons whose natural spontaneous respiratory frequency ranged 9-27 breaths per minute (0.15-0.45 Hz) with predominance of the HF power of HRV was compared to another subgroup whose natural respiratory frequency was less than 9 breaths per min (0.15 Hz) (Pinna et al., 2006). In that study, paced breathing at 15 breaths per min (0.25 Hz) failed to alter either the time domain or the frequency domain HRV parameters. In a study population of healthy adults, data sets matching respiratory frequencies at rest with those during dynamic exercise demonstrated that the LF and HF powers of HRV were not changed by controlled breathing in the absence of dynamic exercise but were significantly decreased at the same respiratory frequencies during exercise (Bartels et al., 2004). Because of the expected transients and uncertainties about the role of respiration and significance of changes in frequency domain HRV, we limited this part of the study to use of a time domain measure of HRV that, based on the

This study has demonstrated a wide range of SDNN measurements performed under various experimental conditions from the largest SDNN measured during paced breathing at 0.2 Hz to the smallest during exercise at 30%-50% of peak oxygen consumption in a group of 52 healthy young adult African-American males. The SDNN measured during short periods of normal spontaneous breathing at 0.18-0.48 Hz while at rest sitting both in a chair and on a cycle ergometer, cold stress and mental stress were intermediate, between those of paced breathing and aerobic exercise. SDNN was used to differentiate a "narrowband" subgroup (6/52 subjects, 12%) on the basis of a much smaller observed difference in SDNN between paced and normal spontaneous breathing and, despite an increase in the heart rate, the absence of a decrement in SDNN during a state of mental stress. These findings suggest that mental stress, elicited by Stroop word conflict testing, seems to have a differentiating effect on SDNN, an easily performed and interpreted time domain measure of heart rate variability. In the frequency domain, higher LF/HF heart rate variability spectral power, a reliable measure of sympathetic modulation of the heart rate, differentiated the paced from the uncontrolled spontaneous breathing condition and the same "narrowband" subgroup,

in the hypertensive subgroups (Prakash et al., 2005).

current knowledge, would be more easily interpreted.

**5. Conclusions** 

as demonstrated in figures 7 and 8.

Fig. 6. **Heart rate variability across various testing conditions for broadband and narrowband subjects.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace), normal uncontrolled spontaneous breathing at rest sitting in a chair (rest), Stroop word-color conflict testing of mental stress (mental), cold pressor testing of nociceptive stress (cold), normal uncontrolled spontaneous breathing at rest sitting on a cycle ergometer (rest cycle) and aerobic exercise stress at 30% and 50% of peak oxygen consumption (30% VO2peak, 50% VO2peak). Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. ♦ NB different from BB at P<0.05.

Fig. 6. **Heart rate variability across various testing conditions for broadband and narrowband subjects.** Bars represent heart rate variability (HRV, ms) expressed as time domain measure of mean standard deviation of normal-normal electrocardiogram RR intervals (SDNN) during paced breathing at 0.2 Hz (pace), normal uncontrolled

46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. ♦ NB different from

BB at P<0.05.

spontaneous breathing at rest sitting in a chair (rest), Stroop word-color conflict testing of mental stress (mental), cold pressor testing of nociceptive stress (cold), normal uncontrolled spontaneous breathing at rest sitting on a cycle ergometer (rest cycle) and aerobic exercise stress at 30% and 50% of peak oxygen consumption (30% VO2peak, 50% VO2peak). Subjects are under technical supervision and in the low range of normal respiratory rate. Breathing in the range of 3-9 breaths per minute (0.05-0.15 Hz) may produce higher amplitude respiratory sinus arrhythmia because of more complete acetylcholine metabolism during exhalation (Song and Lehrer, 2003). Respiratory sinus arrhythmia might also be maximized if subjects breathe at frequencies controlled by other physiological processes. For example, when breathing between 4-7 breaths per minute (0.07-0.12 Hz), the baroreceptor reflex might be stimulated; thereby, causing a resonance effect and an increase in HRV (Vaschillo et al., 2002). In the present study, paced breathing was associated with significantly lower SDNN (0.2 Hz) than the spontaneous breathing trials (0.18-0.48 Hz) and we did not measure tidal volume which has been shown to positively modulate the high frequency (vagal) component of HRV spectral power (Pinna et al., 2006). At a high respiratory frequency paced breathing has been shown to produce predominance of the HF power of HRV in normal healthy adult subjects during meditation (Cysarz and Bussing, 2005). HRV studied during paced breathing in subgroups of pre-hypertensive and hypertensive middle-aged men and women has been shown to produce greater HF power in the pre-hypertensive than in the hypertensive subgroups (Prakash et al., 2005).

HRV in a subgroup of healthy normotensive middle-aged persons whose natural spontaneous respiratory frequency ranged 9-27 breaths per minute (0.15-0.45 Hz) with predominance of the HF power of HRV was compared to another subgroup whose natural respiratory frequency was less than 9 breaths per min (0.15 Hz) (Pinna et al., 2006). In that study, paced breathing at 15 breaths per min (0.25 Hz) failed to alter either the time domain or the frequency domain HRV parameters. In a study population of healthy adults, data sets matching respiratory frequencies at rest with those during dynamic exercise demonstrated that the LF and HF powers of HRV were not changed by controlled breathing in the absence of dynamic exercise but were significantly decreased at the same respiratory frequencies during exercise (Bartels et al., 2004). Because of the expected transients and uncertainties about the role of respiration and significance of changes in frequency domain HRV, we limited this part of the study to use of a time domain measure of HRV that, based on the current knowledge, would be more easily interpreted.

### **5. Conclusions**

This study has demonstrated a wide range of SDNN measurements performed under various experimental conditions from the largest SDNN measured during paced breathing at 0.2 Hz to the smallest during exercise at 30%-50% of peak oxygen consumption in a group of 52 healthy young adult African-American males. The SDNN measured during short periods of normal spontaneous breathing at 0.18-0.48 Hz while at rest sitting both in a chair and on a cycle ergometer, cold stress and mental stress were intermediate, between those of paced breathing and aerobic exercise. SDNN was used to differentiate a "narrowband" subgroup (6/52 subjects, 12%) on the basis of a much smaller observed difference in SDNN between paced and normal spontaneous breathing and, despite an increase in the heart rate, the absence of a decrement in SDNN during a state of mental stress. These findings suggest that mental stress, elicited by Stroop word conflict testing, seems to have a differentiating effect on SDNN, an easily performed and interpreted time domain measure of heart rate variability. In the frequency domain, higher LF/HF heart rate variability spectral power, a reliable measure of sympathetic modulation of the heart rate, differentiated the paced from the uncontrolled spontaneous breathing condition and the same "narrowband" subgroup, as demonstrated in figures 7 and 8.

Low Heart Rate Variability in Healthy Young Adult Males 243

Bartels, M.N., Jelic, S., Ngai, P., et al (2004). The effect of ventilation on spectral analysis of

Burutcu, I., Esen, A.M., Kaya, D., et al (2005). Cigarette smoking and HRV: dynamic

Cysarz, D., and Bussing, A (2005). Cardiorespiratory synchronization during Zen

De Becker, P., Dendale, P., De Meirleir, K., et al (1998). Autonomic testing in patients with

De Meersman, R.E., Reisman, S.S., Daum, M., et al (1995). Influence of respiration on

Delaney, J.P., and Brodie, D.A (2000). Effects of short-term psychological stress on the time and frequency domains of heart-rate variability. *Percept Mot Skills*. 91, 515-524. Driscoll, D., and Dicicco, G (2000). The effects of metronome breathing on the variability of autonomic activity measurements. *J Manipulative Physiol Ther*. 23, 610-614. Fauvel, J.P., Bernard, N., Laville, M., et al (1996). Reproducibility of the cardiovascular

Gillum, R.F (1979). Pathophysiology of hypertension in blacks and whites. A review of the

Gillum, R.F (1999). Risk factors for stroke in blacks: a critical review. *Am J Epidemiol*. 150,

Gutierrez, J., Santiesteban, R., Garcia, H., et al (2002). High blood pressure and decreased HRV in the Cuban epidemic neuropathy. *J Neurol Neurosurg Psychiatry*. 73, 71-72. Hayano, J., Sakakibara, Y., Yamada, M., et al (1990). Decreased magnitude of heart rate

Hrushesky, W.J (1991). Quantitative respiratory sinus arrhythmia analysis. A simple

Kuch, B., Hense, H.W., Sinnreich, R., et al (2001). Determinants of short-period HRV in the

Lucini, D., Mela, G.S., Malliani, A., et al (2002a). Impairment in cardiac autonomic

Lucini, D., Guzzetti, S., Casiraghi, S., et al (2002c). Correlation between baroreflex gain and

Markos, F. and Snow, H.M (2001). The potentiation of sinus arrhythmia by vasoactive intestinal polypeptide (VIP) in the anaesthetized dog. *Neuropeptides.* 35, 238-243.

to real life stress conditions in humans. *Hypertension.* 39, 184-188.

basis of racial blood pressure differences. *Hypertension.* 1, 468-475.

heart rate and blood pressure variability during exercise. *Respir Physiol Neurobiol*.

influence of parasympathetic and sympathetic maneuvers. *Ann Noninvasive* 

metabolic, hemodynamic, psychometric, and R-R interval power spectral

reactivity to a computerized version of the Stroop stress test in normotensive and

spectral components in coronary artery disease. Its relation to angiographic

noninvasive, reimbursable measure of cardiac wellness and dysfunction. *Ann NY* 

regulation preceding arterial hypertension in humans: insights from spectral analysis of beat-by-beat cardiovascular variability. *Circulation* 106, 2673-2679. Lucini, D., Norbiato, G., Clerici, M., et al (2002b). Hemodynamic and autonomic adjustments

**6. References** 

144, 91-98.

1266-1274.

*Electrocardiol*. 10, 324-329.

meditation. *Eur J Appl Physiol*. 95, 88-95.

chronic fatigue syndrome. *Am J Med*. 105, :22S-26S.

parameters. *Am J Physiol*. 269(4 Pt 2), H1437-H1440.

hypertensive subjects. *Clin Auton Res*. 6, 219-224.

Kaplan, N.M (1994). Ethnic aspects of hypertension. *Lancet.* 344, 450-452.

general population. *Cardiology*. 95, 131-138.

24-h indices of HRV. *J Hypertens*. 20, 1625-1631.

severity. *Circulation.* 81, 1217-1224.

*Acad Sci*. 618, 67-101.

Fig. 7. **Heart rate variability spectral power measure of sympathovagal balance for paced and uncontrolled breathing conditions.** Bars represent low frequency/high frequency ratio (LF/HF) of heart rate variability spectral power measured by fast Fourier transform analysis of electrocardiogram RR intervals during paced breathing at 0.2 Hz (pace) and normal uncontrolled spontaneous breathing at rest sitting in a chair (rest). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. The rest condition is different from paced breathing control (pace) at P<0.01.

Fig. 8. **Heart rate variability spectral power measure of sympathovagal balance for broadband and narrowband groups.** Bars represent low frequency/high frequency ratio (LF/HF) of heart rate variability spectral power measured by fast Fourier transform analysis of electrocardiogram RR intervals during paced breathing at 0.2 Hz. Subjects are 46 broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. LF/HF of the BB group is different from that of the NB group at P<0.05.

### **6. References**

242 Advances in Electrocardiograms – Clinical Applications

Fig. 7. **Heart rate variability spectral power measure of sympathovagal balance for paced and uncontrolled breathing conditions.** Bars represent low frequency/high frequency ratio (LF/HF) of heart rate variability spectral power measured by fast Fourier transform analysis of electrocardiogram RR intervals during paced breathing at 0.2 Hz (pace) and normal uncontrolled spontaneous breathing at rest sitting in a chair (rest). Subjects are 52 healthy young adult African-American males. Data in mean ± standard error. The rest condition is

**PACE REST**

Fig. 8. **Heart rate variability spectral power measure of sympathovagal balance for broadband and narrowband groups.** Bars represent low frequency/high frequency ratio (LF/HF) of heart rate variability spectral power measured by fast Fourier transform analysis

of electrocardiogram RR intervals during paced breathing at 0.2 Hz. Subjects are 46

broadband (BB) healthy young adult African-American males with normal HRV compared to a similar group of 6 narrowband (NB) subjects with low HRV defined as exhibiting SDNN during paced breathing more than one standard deviation from the mean SDNN of the study group of 52 subjects. Data in mean ± standard error. LF/HF of the BB group is

**BB NB**

different from paced breathing control (pace) at P<0.01.

**0.0**

**2.5**

**5.0**

**7.5**

**LF/HF**

**10.0**

**12.5**

different from that of the NB group at P<0.05.

**0**

**1**

**2**

**3**

**LF/HF**

**4**

**5**

**6**


**14** 

*Latvia* 

Iveta Mintale et al.\*

**The Role of Exercise Test After** 

**Percutaneous Coronary Intervention** 

*Paul Stradins Clinical University Hospital, Latvian Centre of Cardiology* 

Cardiovascular diseases is the most common cause of death worldwide and coronary artery disease (CAD) still remains the leading cause of death in Latvia. The incidence of CAD in Europe is 20-40 thousands per one million people (Fox et al., 2006). According to the data of Health Statistics and Medical Technology Agency in Latvia 16 079 individuals died due to cardiovascular disease in 2009 (54% of total mortality rate). Mortality rate due to CAD is three times higher in Latvia than average in European Society in individuals group younger than 64 years. The loss of productive years of life has negative influence neither to private,

The common electrocardiographic finding on exercise stress test, which can be evident for CAD, is recurrent, load-induced ST-segment depression. It points to myocardial ischemia in patient with significantly narrowed coronary arteries in status of progressive oxygen demand, while at rest blood flow is not limited. The sensitivity of the stress test increases along with severity of disease. It is possible correctly to identify the patients with proximal several arteries disease or left main artery stenosis performing a standard exercise stress test, nevertheless it gives unsufficient prognostic information in patient with less severe

Progression of coronary artery restenosis after percutaneous coronary intervention (PCI) is a clinical "final" result, which reflects complex pathophysiological process, including different combinations of residual coronary stenosis and neointimal proliferation. Unfortunately, the set of clinical symptoms is uncertain criteria for detection of coronary artery restenosis, patient's complaints could be similar to non-coronary pain post revascularization ('falsepositive' symptoms), at the same time – "silent" ischemia may be present in many patients ('pseudo-negative' symptoms). Restenosis is observed in 25% of cases (data of balloon angioplasty) in asymptomatic patients with documented ischemic changes on exercise test

Discussions are still extended, whether early performed strategy of exercise test is with prognostic value for clinical events in patients after PCI. There are no conclusive results about association of complaints' limited exercise tests and long-term prognosis early after

\*Milana Zabunova, Dace Lurina, Inga Narbute, Sanda Jegere, Ilja Zakke, Vilnis Taluts Dzerve and

**1. Introduction** 

nor public economical sector.

obstructive disease.

(Bengston et al., 1990).

PCI. 

Andrejs Erglis


### **The Role of Exercise Test After Percutaneous Coronary Intervention**

Iveta Mintale et al.\* *Paul Stradins Clinical University Hospital, Latvian Centre of Cardiology Latvia* 

### **1. Introduction**

244 Advances in Electrocardiograms – Clinical Applications

Mezzacappa E.S., Kelsey, R.M., Katkin, E.S., et al (2001). Vagal rebound and recovery from

Montano, N., Porta, A., Cogliati, C., et al (2009). Heart rate variability explored in the

Murakami, E., Matsuzaki, K., Sumimoto, T., et al (1996). Clinical significance of pressor responses to laboratory stressor testing in hypertension. *Hypertens Res*. 19, 133-137.

Narkiewicz, K., Montano, N., Cogliati, C., et al (1998). Altered cardiovascular variability in

Petretta, M., Marciano, F., Bianchi, V., et al (1995). Power spectral analysis of heart period

Pichon, A.P., De Bisschop, C., Roulaud, M., et al (2004). Spectral analysis of HRV during

Pinna, G.D., Maestri, R., La Rovere, M.T., et al (2006). Effect of paced breathing on

Pinna, G.D., Maestri, R., La Rovere, M.T., et al (2006). Effect of paced breathing on

Prakash, E.S., Madanmohan-Sethuraman, K.R., Narayan, S.K (2005). Cardiovascular

Salo, T.M., Jula, A.M., Piha, J.S., et al (2000). Comparison of autonomic withdrawal in men

Song, H-S. and Lehrer, P.M (2003). The effects of specific respiratory rates on heart rate and

Taverner, D., Nunan, T.A., Tonkin, A.L et al (1996). Reproducibility of conventional and

Vanninen, E., Tuunainen, A., Kansanen, M., et al (1996). Cardiac sympathovagal balance

Vaschillo, E., Lehrer, P., Rishe, N., et al (2002). HRV biofeedback as a method for assessing

exercise in trained subjects. *Med Sci Sports Exerc*. 36, 1702-1708.

among urban adolescents. *J Adolesc Health*. 14, 314-318.

condition. *Am J Cardiol*. 85, 232-238.

*Appl Psychophysiol Biofeedback*. 27, 1-27.

HRV. *Appl Psychophysiol Biofeedback*. 28, 13-23.

subjects. *Clin Exp Pharmacol Physiol*. 23, 804-806.

during sleep apnea episodes. *Clin Physiol*. 16, 209-216.

frequency domain: a tool to investigate the link between heart and behavior.

variability in hypertensive patients with left ventricular hypertrophy. *Am J* 

ventilatory and cardiovascular variability parameters during short-term investigations of autonomic function. *Am J Physiol Heart Circ Physiol*. 290, H424-

ventilatory and cardiovascular variability parameters during short-term investigations of autonomic function. *Am J Physiol Heart Circ Physiol*. 290, H424-

autonomic regulation in subjects with normal blood pressure, high-normal blood pressure and recent-onset hypertension. *Clin Exp Pharmacol Physiol*. 32, 488-494. Rabinowitz, A., Kushner, H. and Falkner, B (1993). Racial differences in blood pressure

with obstructive sleep apnea syndrome, systemic hypertension, and neither

power spectral measurements of cardiovascular sympathetic activation in normal

baroreflex function: a preliminary study of resonance in the cardiovascular system.

psychological stress. Psychosom Med. 63, 650-657.

obstructive sleep apnea. *Circulation.* 98, 1071-1077.

*Neurosci Biobehav Rev*. 33, 71-80.

*Hypertens*. 8(12 Pt 1), 1206-1213.

H433.

H433.

Cardiovascular diseases is the most common cause of death worldwide and coronary artery disease (CAD) still remains the leading cause of death in Latvia. The incidence of CAD in Europe is 20-40 thousands per one million people (Fox et al., 2006). According to the data of Health Statistics and Medical Technology Agency in Latvia 16 079 individuals died due to cardiovascular disease in 2009 (54% of total mortality rate). Mortality rate due to CAD is three times higher in Latvia than average in European Society in individuals group younger than 64 years. The loss of productive years of life has negative influence neither to private, nor public economical sector.

The common electrocardiographic finding on exercise stress test, which can be evident for CAD, is recurrent, load-induced ST-segment depression. It points to myocardial ischemia in patient with significantly narrowed coronary arteries in status of progressive oxygen demand, while at rest blood flow is not limited. The sensitivity of the stress test increases along with severity of disease. It is possible correctly to identify the patients with proximal several arteries disease or left main artery stenosis performing a standard exercise stress test, nevertheless it gives unsufficient prognostic information in patient with less severe obstructive disease.

Progression of coronary artery restenosis after percutaneous coronary intervention (PCI) is a clinical "final" result, which reflects complex pathophysiological process, including different combinations of residual coronary stenosis and neointimal proliferation. Unfortunately, the set of clinical symptoms is uncertain criteria for detection of coronary artery restenosis, patient's complaints could be similar to non-coronary pain post revascularization ('falsepositive' symptoms), at the same time – "silent" ischemia may be present in many patients ('pseudo-negative' symptoms). Restenosis is observed in 25% of cases (data of balloon angioplasty) in asymptomatic patients with documented ischemic changes on exercise test (Bengston et al., 1990).

Discussions are still extended, whether early performed strategy of exercise test is with prognostic value for clinical events in patients after PCI. There are no conclusive results about association of complaints' limited exercise tests and long-term prognosis early after PCI.

<sup>\*</sup>Milana Zabunova, Dace Lurina, Inga Narbute, Sanda Jegere, Ilja Zakke, Vilnis Taluts Dzerve and Andrejs Erglis

The Role of Exercise Test After Percutaneous Coronary Intervention 247

1-3 months after invasive treatment – immediate risk evaluation and correction of

3-6 months after PCI – distant period – determination of possibility of restenosis and

12-24 months after PCI – late period – determination of possibility of restenosis and

Control coronary angiography was performed based on indications, which included

patients with new-onset angina proved by ischemic changes on ECG during exercise

patients without typical coronary complaints, but with registered ischemic changes on

 patients with non-sufficient information on exercise test in order to evaluate indications for coronary angiography, therefore exercise test with additional visualization method

 Patients without indications for control coronary angiography underwent the correction of medical treatment (if it was needed) and were included in the next scheduled

Large patients group was revealed during the follow-up process – patients without typical coronary chest pain but with registered ischemic changes on ECG during exercise test. These patients were defined like "silent" ischemia patients group which has the higher risk of

All the patients of this group underwent control coronary angiography and were divided

Patients with restenosis or new stenosis underwent PCI and were included in consequent follow-up plan and medical treatment correction. Patients without restenosis or new

The analysis of used medications groups was performed because of medical treatment correction during follow-up process. It was dependent on obtained data during exercise test (e.g., arterial blood pressure and/or heart rate, compliance of these parameters changes to physical load, etc.). Medications after performed PCI (recommended on discharge) and in

Very high risk patients were included into the separate group analysis (400 patients) with LM stenotic lesion. All patients were informed about possibility to be included in the 24 months follow-up plan. 133 patients have accomplished complete follow-up programme (1, 3, 6, 12 and 24 months after PCI). In order to define precise indications for control coronary angiography, myocardial perfusion scintigraphy with physical or pharmacological test was performed in this patients group (94 examinations). This examination was performed in 12-

Phone follow-up survey was also performed in this patients group in order to clarify possible coronary events – hospitalization, myocardial infarction or death. Patients groups

24 months period after PCI in patients without indications for control angiography.

with and without performed follow-up programme were compared.

patients with restenosis of coronary artery (in previously repaired segment);

patients without restenosis or new stenosis on angiographic finding.

medical treatment;

following patients groups:

exercise test ECG;

exercise test control.

cardiovascular events and worse prognosis.

one year follow-up period were compared.

into different groups according to angiographic finding:

patients with new stenosis of coronary artery;

stenosis continued previously defined follow-up plan.

stress test;

correction of medical treatment;

correction of medical treatment.

was performed – myocardial perfusion scintigraphy.

CAD patients with left main coronary artery (LM) stenosis is another one group of interest, and there is stable introduced invasive PCI treatment and stent implantation for this patients' group. It is important to evaluate the efficacy of new treatment, outcomes and following risk of cardiovascular events. Patients after LM PCI are at higher risk of thrombosis and restenosis, angiographic follow-up of these patients is expensive enough, with additional radiation dose and contrast amount for patients. Thus, high attention is focused on non-invasive investigation methods in patients with CAD.

The results of recent studies should be taken into account, that the higher risk of cardiovascular events is observed in patients with unstable atheromatous plaque, even in case if stenosis of coronary artery is not significant (in segment of this plaque location). Exercise test in this situation should be evaluated more precisely and seriously in order to improve the prognostic value of the method. The question is – whether exercise test (with electrocardiogramm (ECG)) can be used in evaluation of prognostic criteria for patient along with possibility to detect obstructive lesion of coronary arteries with a goal to decrease effectively the risk of sudden death and serious coronary events?

### **2. Objective of the study**

The aim of the study was to evaluate effectiveness of medical and invasive treatment in patients with CAD after PCI by performing physical test and proving test prognostic value of cardiovascular events - clinical (myocardial infarction, recurrent hospitalization, cardiac death) and angiographic (restenosis and/or new stenosis of coronary arteries) events. The main goals were defined:


### **3. Materials and methods**

The follow-up programme was developed in 1990 in the Latvian Centre of Cardiology and surveyed patients with CAD which underwent PCI treatment method.

The programme has been proceeded since the first patient was treated with the method of coronary angioplasty in Latvia.

The study was implemented from January, 2004 till January, 2009. The patients with established CAD (based on coronary angiography data) were included in the study. The patients underwent invasive treatment – PCI in the Latvian Centre of Cardiology. The number of performed PCIs during mentioned period of time – 16 109. All patients were informed about possibility to be included in the follow-up programme.

The data of 7 300 patients with CAD and complete 24-months follow-up after PCI were included in the study.

The patients were observed in defined follow-up periods of time after PCI by performing exercise stress test (veloergometry – bicycle ergometer exercise test):

CAD patients with left main coronary artery (LM) stenosis is another one group of interest, and there is stable introduced invasive PCI treatment and stent implantation for this patients' group. It is important to evaluate the efficacy of new treatment, outcomes and following risk of cardiovascular events. Patients after LM PCI are at higher risk of thrombosis and restenosis, angiographic follow-up of these patients is expensive enough, with additional radiation dose and contrast amount for patients. Thus, high attention is

The results of recent studies should be taken into account, that the higher risk of cardiovascular events is observed in patients with unstable atheromatous plaque, even in case if stenosis of coronary artery is not significant (in segment of this plaque location). Exercise test in this situation should be evaluated more precisely and seriously in order to improve the prognostic value of the method. The question is – whether exercise test (with electrocardiogramm (ECG)) can be used in evaluation of prognostic criteria for patient along with possibility to detect obstructive lesion of coronary arteries with a goal to decrease

The aim of the study was to evaluate effectiveness of medical and invasive treatment in patients with CAD after PCI by performing physical test and proving test prognostic value of cardiovascular events - clinical (myocardial infarction, recurrent hospitalization, cardiac death) and angiographic (restenosis and/or new stenosis of coronary arteries) events.

3. to create the prognostic model, which can help to evaluate and reveal the patients with

The follow-up programme was developed in 1990 in the Latvian Centre of Cardiology and

The programme has been proceeded since the first patient was treated with the method of

The study was implemented from January, 2004 till January, 2009. The patients with established CAD (based on coronary angiography data) were included in the study. The patients underwent invasive treatment – PCI in the Latvian Centre of Cardiology. The number of performed PCIs during mentioned period of time – 16 109. All patients were

The data of 7 300 patients with CAD and complete 24-months follow-up after PCI were

The patients were observed in defined follow-up periods of time after PCI by performing

1. to define specific criteria of exercise stress test for possible restenosis diagnostics; 2. to develop the algorhythm for the patients' functional status evaluation after PCI, as well as for evaluation the possibility of coronary artery stenosis correction

unfavorable long-term outcomes or unsatisfactory treatment results timely;

focused on non-invasive investigation methods in patients with CAD.

effectively the risk of sudden death and serious coronary events?

4. to integrate targeted follow-up patients' programme.

surveyed patients with CAD which underwent PCI treatment method.

informed about possibility to be included in the follow-up programme.

exercise stress test (veloergometry – bicycle ergometer exercise test):

**2. Objective of the study** 

The main goals were defined:

**3. Materials and methods** 

coronary angioplasty in Latvia.

included in the study.

(intervention);


Control coronary angiography was performed based on indications, which included following patients groups:


Large patients group was revealed during the follow-up process – patients without typical coronary chest pain but with registered ischemic changes on ECG during exercise test. These patients were defined like "silent" ischemia patients group which has the higher risk of cardiovascular events and worse prognosis.

All the patients of this group underwent control coronary angiography and were divided into different groups according to angiographic finding:


Patients with restenosis or new stenosis underwent PCI and were included in consequent follow-up plan and medical treatment correction. Patients without restenosis or new stenosis continued previously defined follow-up plan.

The analysis of used medications groups was performed because of medical treatment correction during follow-up process. It was dependent on obtained data during exercise test (e.g., arterial blood pressure and/or heart rate, compliance of these parameters changes to physical load, etc.). Medications after performed PCI (recommended on discharge) and in one year follow-up period were compared.

Very high risk patients were included into the separate group analysis (400 patients) with LM stenotic lesion. All patients were informed about possibility to be included in the 24 months follow-up plan. 133 patients have accomplished complete follow-up programme (1, 3, 6, 12 and 24 months after PCI). In order to define precise indications for control coronary angiography, myocardial perfusion scintigraphy with physical or pharmacological test was performed in this patients group (94 examinations). This examination was performed in 12- 24 months period after PCI in patients without indications for control angiography.

Phone follow-up survey was also performed in this patients group in order to clarify possible coronary events – hospitalization, myocardial infarction or death. Patients groups with and without performed follow-up programme were compared.

The Role of Exercise Test After Percutaneous Coronary Intervention 249

especially, if these changes were associated with chest pain, appeared at lower load (<





Following information was included in the protocol of veloergometry (accordingly the

The data mentioned above (collected from veloergometry test protocols) were included into

There were no any complications registered during exercise test perfoming – myocardial

multiply by maximal systolic arterial pressure, divided by 100)

75 W) and lasting for more than 3 minutes after load was finished.



Following parameters were analized: a. Electrocardiographic parameters:

upslopping);


b. Hemodynamic parameters: - maximal heart rate;

was started);

c. Symptomatic parameters:

methodological guidelines):

maximal heart rate;

 patient's complaints; test termination reasons;

load-limiting factors.

the total database of the study.

double rate-pressure product;

recovering period after load;

 maximal load; total time of the load; heart rate at the beginning;






initial systolic and diastolic blood pressure;

the intensity and quantity of ECG changes;

infarction, life-threating arrhythmias or death.

maximal systolic and diastolic blood pressure;

tolerance of the exercise (load) (high, satisfactory, lowered, low);

SPSS 12.0.1. version was used for statistical analyses. Baseline characteristics were summarized as frequencies and percentages for categorical variables and as means and SDs for continuous variables. Analyses were performed using χ2/Fisher exact test for categorical variables and Student *t* test/ *ANOVA* method for continuous variables. The correction with *Tukey* test was performed in *post hoc* analysis for multiple comparisons correction. Differences were considered statistically significant at *p*<0.05.

### **3.1 Exercise test – veloergometry**

Exercise stress test – veloergometry was performed in sitting or reclining position in patients of the study, depending on used bicycle ergometer (in sitting position – veloergometer "Ergometrs 900" and since April, 2008 – veloergometer "e-bike" for test performing in reclining position). Modified Bruce protocol was used during exercise test – standardized load grading programme according to the protocol accepted by Latvian Cardiology Society:


Maximal allowed load was defined as 250 W. In specific cases (if it was necessary) or according to patient condition (patients with chronic heart failure) the protocol was accommodated individually – the load was increased for 25 W every three minutes with the beginning of second degree of the test protocol.

ECG was monitored continuously during the test time, increasing the load till the goal heart rate was achieved – 85 % of maximal heart rate according (appropriate) to concrete age or the load was limited by symptoms of ischemia.

Following indications were defined for the test discontinuation:


Exercise test was defined as positive, if there were registered changes on ECG associated with myocardial ischemia:

 horizontal or downslopping ST-segment depression or elevation, greater than or equal with 1 mm 60 msec after QRS complex;

 especially, if these changes were associated with chest pain, appeared at lower load (< 75 W) and lasting for more than 3 minutes after load was finished.

Following parameters were analized:

248 Advances in Electrocardiograms – Clinical Applications

SPSS 12.0.1. version was used for statistical analyses. Baseline characteristics were summarized as frequencies and percentages for categorical variables and as means and SDs for continuous variables. Analyses were performed using χ2/Fisher exact test for categorical variables and Student *t* test/ *ANOVA* method for continuous variables. The correction with *Tukey* test was performed in *post hoc* analysis for multiple comparisons correction.

Exercise stress test – veloergometry was performed in sitting or reclining position in patients of the study, depending on used bicycle ergometer (in sitting position – veloergometer "Ergometrs 900" and since April, 2008 – veloergometer "e-bike" for test performing in reclining position). Modified Bruce protocol was used during exercise test – standardized load grading programme according to the protocol accepted by Latvian Cardiology Society:

Maximal allowed load was defined as 250 W. In specific cases (if it was necessary) or according to patient condition (patients with chronic heart failure) the protocol was accommodated individually – the load was increased for 25 W every three minutes with the

ECG was monitored continuously during the test time, increasing the load till the goal heart rate was achieved – 85 % of maximal heart rate according (appropriate) to concrete age or

a. typical increasing chest pain and/or ST-segment changes on ECG (ST-segment depression > 2 mm – as a relative indication for the test interruption, ST-segment

d. systolic blood pressure decrease (> 10 mmHg compared with onset blood pressure

Exercise test was defined as positive, if there were registered changes on ECG associated

horizontal or downslopping ST-segment depression or elevation, greater than or equal

depression > 3 mm – as a absolute indication for the test interruption);

b. ST-segment elevation > 1 mm (if pathologic Q-wave is not present on ECG at rest); c. arrhythmias (supraventricular tachycardia, multiple politope and pair premature beats

e. hypertension (> 250 mmHg systolic and > 115 mmHg diastolic blood pressure);

Differences were considered statistically significant at *p*<0.05.

 first degree – three minutes with 50 Watt (W) load; second degree – three minutes with 100 W load; third degree – three minutes with 150 W load; fourth degree – three minutes with 200 W load; fifth degree – three minutes with 250 W load.

beginning of second degree of the test protocol.

the load was limited by symptoms of ischemia.

level), despite of the load increase;

g. signs of decreased perfusion (cyanosis); h. fatigue, dyspnoe, leg cramps, claudication; i. complete left His bundle branch block; j. achieved submaximal heart rate; k. patient's request to terminate the test;

with 1 mm 60 msec after QRS complex;

(extrasystoles));

l. technical problems.

with myocardial ischemia:

Following indications were defined for the test discontinuation:

f. central nervous system symptoms (ataxy, weakness, syncope);

**3.1 Exercise test – veloergometry** 

	- maximal ST-segment depression;
	- maximal ST-segment elevation;
	- configuration (shape) of ST-segment depression (horizontal, downslopping, upslopping);
	- the number of leads ST-segment changes registered in;
	- duration of ST-segment changes at the test phase of rest (after load discontinued);
	- ST/pulse index;
	- ventricular arrhythmias induced by exercise;
	- the time to ST-segment changes appeared at.
	- maximal heart rate;
	- maximal systolic blood pressure;
	- maximal double rate-pressure product (Robinson index) (maximal heart rate multiply by maximal systolic arterial pressure, divided by 100)
	- total exercise (load) time;
	- hypotension (arterial pressure decrease under the blood pressure level before load was started);
	- chronotropic incompetence.
	- angina provoked by load;
	- load-limiting symptoms;
	- time to the onset of angina.

Following information was included in the protocol of veloergometry (accordingly the methodological guidelines):


The data mentioned above (collected from veloergometry test protocols) were included into the total database of the study.

There were no any complications registered during exercise test perfoming – myocardial infarction, life-threating arrhythmias or death.

The Role of Exercise Test After Percutaneous Coronary Intervention 251

Fig. 1. Results of the follow-up programme (the total number of the patients – 7300).

Fig. 2. Patients with "silent" ischemia.

**Exercise test is useful in detection of patients with "silent" ischemia (22%), which have the indications for control angiography; restenosis is found in 9% of this patients group.** 

Myocardial perfusion scintigraphy was perfomed in patients at the case of embarrassing interpretation of ECG, and in order to evaluate the indications for control coronary angiography.

Myocardial perfusion scintigraphy was also performed in patients after LM PCI.

### **3.2 Coronary angiography and PCI**

The information about atherosclerotic lesions degree of coronary arteries was collected and included into the total database of the study (based on coronary angiography finding). Follow-up period results were included – data of coronary angiography finding (restenosis and/or new stenosis of coronary artery(-ies)), acute coronary events (recurrent hospitalization, myocardial infarction, coronary artery bypass grafting, recurrent coronary angiography and/ or PCI).

The result of revascularization was defined, basing on the data of coronary angiography finding:


Restenosis of coronary artery at the follow-up period was defined as a narrowing of coronary artery(-ies) lumen greater than or equal with 50 % and progression of coronary artery disease, if new stenosis of coronary artery was diagnosed (Chalela et al., 2006).

### **4. Results**

### **4.1 Data registry of the patients with coronary artery disease treated with PCI**

In total 7300 patients with established CAD and performed treatment with PCI, with complete follow-up programme – 1, 3, 6 and 12 months follow-up visits, were included into the study. The patients underwent physical stress test, correction of used medications plan and doses of drugs, if it was necessary, and the control of risk factors. Control coronary angiography was performed in 2583 patients accordingly the indications.

Indications for performance of control coronary angiography were determined basing on exercise test (13% of cases), restenosis on angiography was detected in 6.4% of the total patients' group.

In the patients' group with left main artery stenosis (LM group) 400 patients were observed, 133 of them worked out complete follow-up program and underwent control coronary angiography. Myocardial perfusion scintigraphy was performed in 98 LM-patients in order to evaluate more precisely the indications for coronary angiography.

In total patients' group 1226 patients (17%) showed coronary complaints, and ST-segment changes on ECG, which could be associated with myocardial ischemia, were observed in 975 patients (13% of total number of patients). According to the indications these patients underwent coronary angiography. In general, angiographically confirmed coronary artery restenosis was diagnosed in 470 patients (6.4%) of all controlled patients' group (7300) (Figure 1.).

Myocardial perfusion scintigraphy was perfomed in patients at the case of embarrassing interpretation of ECG, and in order to evaluate the indications for control coronary

The information about atherosclerotic lesions degree of coronary arteries was collected and included into the total database of the study (based on coronary angiography finding). Follow-up period results were included – data of coronary angiography finding (restenosis and/or new stenosis of coronary artery(-ies)), acute coronary events (recurrent hospitalization, myocardial infarction, coronary artery bypass grafting, recurrent coronary

The result of revascularization was defined, basing on the data of coronary angiography



Restenosis of coronary artery at the follow-up period was defined as a narrowing of coronary artery(-ies) lumen greater than or equal with 50 % and progression of coronary

In total 7300 patients with established CAD and performed treatment with PCI, with complete follow-up programme – 1, 3, 6 and 12 months follow-up visits, were included into the study. The patients underwent physical stress test, correction of used medications plan and doses of drugs, if it was necessary, and the control of risk factors. Control coronary

Indications for performance of control coronary angiography were determined basing on exercise test (13% of cases), restenosis on angiography was detected in 6.4% of the total

In the patients' group with left main artery stenosis (LM group) 400 patients were observed, 133 of them worked out complete follow-up program and underwent control coronary angiography. Myocardial perfusion scintigraphy was performed in 98 LM-patients in order

In total patients' group 1226 patients (17%) showed coronary complaints, and ST-segment changes on ECG, which could be associated with myocardial ischemia, were observed in 975 patients (13% of total number of patients). According to the indications these patients underwent coronary angiography. In general, angiographically confirmed coronary artery restenosis was diagnosed in 470 patients (6.4%) of all controlled patients' group (7300)

artery disease, if new stenosis of coronary artery was diagnosed (Chalela et al., 2006).

**4.1 Data registry of the patients with coronary artery disease treated with PCI** 

angiography was performed in 2583 patients accordingly the indications.

to evaluate more precisely the indications for coronary angiography.

Myocardial perfusion scintigraphy was also performed in patients after LM PCI.

angiography.

finding:

**4. Results** 

patients' group.

(Figure 1.).

**3.2 Coronary angiography and PCI** 

angiography and/ or PCI).

(Wenaweser et al., 2008);

(Chalela et al., 2006).

Fig. 1. Results of the follow-up programme (the total number of the patients – 7300).

**Exercise test is useful in detection of patients with "silent" ischemia (22%), which have the indications for control angiography; restenosis is found in 9% of this patients group.** 

Fig. 2. Patients with "silent" ischemia.

The Role of Exercise Test After Percutaneous Coronary Intervention 253

physical load also showed lower parameters in the first group, which is characterized by double rate-pressure product compared in both groups – 212.8±42.9 and 252.62±36.9,

Fig. 4. Analysis of average parameters of veloergometry data (maximal heart rate, maximal

**Maximal systolic blood pressure**

121.4±33.2 118.8±14.6 181.5±28.3 212.8±42.9 1.3±0.5

157.7±38.5 130.8±14.9 187.6±30.7 239.6±46.9 1.0±0.3

160.0±25.8 135.0±40.1 189.2±23.3 252.62±36.9 0.9±0.2

Table 1. Analysis of average parameters of veloergometry data (achieved physical load, maximal heart rate, maximal systolic blood pressure, Robinson index, ST/pulse index).

**Robinson index (ratepressure product)**

**ST/pulse index**

systolic arterial pressure, maximal rate-pressure product – Robinson index).

**Maximal heart rate**  (when ECG changes appear)

**Maximal physical load, W**

**Patients' groups/ Parameter on veloergometry**

**1. Restenosis (n=656)** 

**2. New stenosis**

**(n=146)** 

**3. Without restenosis/ new stenosis (n=806)** 

accordingly, in the first and third patients' group (p**<**0.05) (Figure 4.).

It was observed during patients follow-up, that a separate group of the patients is forming, which is characterized by no complaints of chest pain, but significant ST-segment changes on exercise test ECG (so-called, silent ischemia, poor prognostic factor). These are 22% of the patients (1608 patients), all of them underwent coronary angiography. Restenosis of coronary artery was detected in 656 patients (9% of the total number of patients) and new hemodinamically significant stenosis in coronary artery (> 50% of artery lumen) – in 146 patients (2%) of this patients group (Figure 2.).

### **4.2 New specific criteria in restenosis diagnostics**

Several new specific criteria were found in restenosis diagnostics field additionally to STsegment ischemic changes on ECG during exercise stress test – reduced load tolerance, decreased maximal heart rate and maximal systolic blood pressure, maximal rate-pressure product (double product) – Robinson index, also increased ST/pulse index (ST/p index).

Differences in parameters derived from exercise tests in patients with restenosis, new stenosis and in patients without restenosis/new stenosis on coronary angiography are shown in Figure 3.

Fig. 3. Results of exercise test in "silent" ischemia patients' group.

Patients with restenosis on coronary angiography (first group) showed definitely ischemic changes on exercise ECG at lower physical load 121.4±33.2 W compared with those patients' group with neither restenosis, nor new stenosis on coronary angiography (third group) – 160.0±25.8 W (p**<**0.05). Achieved maximal heart rate and systolic blood pressure during

It was observed during patients follow-up, that a separate group of the patients is forming, which is characterized by no complaints of chest pain, but significant ST-segment changes on exercise test ECG (so-called, silent ischemia, poor prognostic factor). These are 22% of the patients (1608 patients), all of them underwent coronary angiography. Restenosis of coronary artery was detected in 656 patients (9% of the total number of patients) and new hemodinamically significant stenosis in coronary artery (> 50% of artery lumen) – in 146

Several new specific criteria were found in restenosis diagnostics field additionally to STsegment ischemic changes on ECG during exercise stress test – reduced load tolerance, decreased maximal heart rate and maximal systolic blood pressure, maximal rate-pressure product (double product) – Robinson index, also increased ST/pulse index (ST/p index). Differences in parameters derived from exercise tests in patients with restenosis, new stenosis and in patients without restenosis/new stenosis on coronary angiography are

patients (2%) of this patients group (Figure 2.).

shown in Figure 3.

**4.2 New specific criteria in restenosis diagnostics** 

Fig. 3. Results of exercise test in "silent" ischemia patients' group.

Patients with restenosis on coronary angiography (first group) showed definitely ischemic changes on exercise ECG at lower physical load 121.4±33.2 W compared with those patients' group with neither restenosis, nor new stenosis on coronary angiography (third group) – 160.0±25.8 W (p**<**0.05). Achieved maximal heart rate and systolic blood pressure during physical load also showed lower parameters in the first group, which is characterized by double rate-pressure product compared in both groups – 212.8±42.9 and 252.62±36.9, accordingly, in the first and third patients' group (p**<**0.05) (Figure 4.).

Fig. 4. Analysis of average parameters of veloergometry data (maximal heart rate, maximal systolic arterial pressure, maximal rate-pressure product – Robinson index).


Table 1. Analysis of average parameters of veloergometry data (achieved physical load, maximal heart rate, maximal systolic blood pressure, Robinson index, ST/pulse index).

The Role of Exercise Test After Percutaneous Coronary Intervention 255

24 (18.0) 41 (30.8) 51 (38.3)

51 (38.3) 54 (40.6) 28 (21.1)

47 (38.0) 47 (35.3)

Characteristics Prevalence Gender (males), n (%) 107 (80.5) Age, mea(±SD) 61.3 (± 9.8) Dyslipidemia, n (%) 96 (72.2) Arterial hypertension, n (%) 77 (57.9) Diabete mellitus, n (%) 11 (8.3)

Positive family history, n (%) 48 (36.1) Prior myocardial infarction, n (%) 60 (45.1) Prior PCI, n (%) 26 (19.5) Prior coronary artery bypass grafting, n (%) 7 (5.3)

**Characteristics Prevalence** 

Complete/incomplete revascularization, % 40/27

 *PCI LCx*, n (%) 26 (19.5)  *PCI RCA*, n (%) 14 (10.5)  *PCI RIM*, n (%) 4 (3.0)  *PCI D1*, n (%) 3 (2.3) *PCI M1*, *RPL*, *RPD*, n (%) 1 (0.8)

*RPL*–right posterolateral branch; *RPD*-right posterodiaphragmal branch.

Table 4. Caracteristics of angiographic and PCI data of the patients (n=133).

*LM*–left main coronary artery; *LAD*–left anterior descending artery; *LCx*–left circumflex artery; *RCA*– right coronary artery; *RIM*–ramus intermedius; *D1*–first diagonal branch; *M1*–first marginal branch;

Table 3. Clinical characteristics of the patients (demographic parameters and risk factors of

Smoking status:

 Active smoker, n (%) Ex-smoker, n (%) Non-smoker, n (%)

coronary artery disease) (n=133).

Coronary artery disease: One-vessel disease, n (%) Two- vessel disease, n (%) Three- vessel disease, n (%)

*PCI LM* only, n (%) *PCI LM* and other artery:  *PCI LAD*, n (%)

Parameters of neither maximal heart rate, nor maximal systolic arterial blood pressure, and so Robinson index, accordingly, are significantly lower in patients' group witht coronary artery restenosis. The most important parameter, which can reflect patient's risk of restenosis more precisely is ST/pulse index. In restenosis group this parameter is the highest one (1.3±0.5) (Table 1. and 2.).


Table 2. Statistical analysis in comparison of different patients' groups.

### **4.3 Patients with left main coronary artery (LM) stenosis**

### **The database of the patients with LM stenosis was created.**

The group of the patients (n=400) with LM stenosis and performed PCI with drug-eluting stents (DES) was analyzed. In a period of years 2005-2010, 133 patients accomplished complete follow-up programme. Veloergometry was regularly performed in 3, 6, 12 and 24 months after PCI, also control coronarography was performed in 6-12 months after PCI.

### *Clinical characteristics of the patients:*

LM patients group – 80% of the patients are males, with mean age 61.3 ± 9.8 years. The most part of the patients is characterized with most important risk factors of coronary artery disease: in 96% of cases – dyslipidemia, in 77% of cases – arterial hypertension, smoking habits – active smokers (18%) or ex-smokers (30.8%). Previous myocardial infarction – in 45.1% of patients, positive family history (first-line relatives at young age with CAD) in 36.1% of cases (Table 3.).

All the patients underwent LM PCI: 47 patients (38%) underwent solely LM PCI and 86 patients (65%) – also PCI additionally in other artery (Table 4.).

The changes of the parameters registered on exercise test (maximal heart rate, maximal systolic blood pressure and Robinson index) in every follow-up period (independently of the follow-up data on coronary angiography) are showed in Table 5.

### **4.3.1 Association of veloergometry results with clinical events**

Recurrent hospitalization was ascertained for 14 patients (8.3%) in analysis of clinical events in follow-up period. The reasons for recurrent admission to the hospital: in two cases – myocardial infarction, in four cases – chest pain (angina) (three of these patients underwent recurrent coronary angiography), in two cases – next step PCI, in two cases – heart rhythm disorders (in one case – implantation of permanent pacemaker), in one case – stroke, in one case – elevated arterial blood pressure, in two cases – non-coronary reasons for hospitalization.

Parameters of neither maximal heart rate, nor maximal systolic arterial blood pressure, and so Robinson index, accordingly, are significantly lower in patients' group witht coronary artery restenosis. The most important parameter, which can reflect patient's risk of restenosis more precisely is ST/pulse index. In restenosis group this parameter is the

> **Maximal systolic blood pressure**

**< 0.05 < 0.05 0.012 < 0.05 < 0.05** 

**< 0.05 < 0.05 < 0.05 < 0.05 < 0.05** 

The group of the patients (n=400) with LM stenosis and performed PCI with drug-eluting stents (DES) was analyzed. In a period of years 2005-2010, 133 patients accomplished complete follow-up programme. Veloergometry was regularly performed in 3, 6, 12 and 24 months after PCI, also control coronarography was performed in 6-12 months after PCI.

LM patients group – 80% of the patients are males, with mean age 61.3 ± 9.8 years. The most part of the patients is characterized with most important risk factors of coronary artery disease: in 96% of cases – dyslipidemia, in 77% of cases – arterial hypertension, smoking habits – active smokers (18%) or ex-smokers (30.8%). Previous myocardial infarction – in 45.1% of patients, positive family history (first-line relatives at young age with CAD) in

All the patients underwent LM PCI: 47 patients (38%) underwent solely LM PCI and 86

The changes of the parameters registered on exercise test (maximal heart rate, maximal systolic blood pressure and Robinson index) in every follow-up period (independently of

Recurrent hospitalization was ascertained for 14 patients (8.3%) in analysis of clinical events in follow-up period. The reasons for recurrent admission to the hospital: in two cases – myocardial infarction, in four cases – chest pain (angina) (three of these patients underwent recurrent coronary angiography), in two cases – next step PCI, in two cases – heart rhythm disorders (in one case – implantation of permanent pacemaker), in one case – stroke, in one case – elevated arterial blood pressure, in two cases – non-coronary reasons for

0.396 0.118 0.48 **< 0.05 0.004** 

**Robinson index** 

**ST/pulse index** 

**Maximal heart rate** 

Table 2. Statistical analysis in comparison of different patients' groups.

**4.3 Patients with left main coronary artery (LM) stenosis The database of the patients with LM stenosis was created.** 

patients (65%) – also PCI additionally in other artery (Table 4.).

the follow-up data on coronary angiography) are showed in Table 5.

**4.3.1 Association of veloergometry results with clinical events** 

highest one (1.3±0.5) (Table 1. and 2.).

*Clinical characteristics of the patients:* 

36.1% of cases (Table 3.).

hospitalization.

**Load, Watts** 

p value (between patients' groups)

**First and second** 

**Second and third** 

**First and third** 

**group** 

**group** 

**group** 


Table 3. Clinical characteristics of the patients (demographic parameters and risk factors of coronary artery disease) (n=133).


*LM*–left main coronary artery; *LAD*–left anterior descending artery; *LCx*–left circumflex artery; *RCA*– right coronary artery; *RIM*–ramus intermedius; *D1*–first diagonal branch; *M1*–first marginal branch; *RPL*–right posterolateral branch; *RPD*-right posterodiaphragmal branch.

Table 4. Caracteristics of angiographic and PCI data of the patients (n=133).

The Role of Exercise Test After Percutaneous Coronary Intervention 257

In maximal systolic blood pressure analysis, almost statistically significant difference was observed (trend) between 'restenosis' group and 'non-restenosis' patients group on exercise

> **Maximal systolic blood pressure (mean ± SD)**

> > **'Nonrestenosis' group**

**Number of the patients**

**p (between groups)** 

**p (between groups)** 

**Number of the patients**

1-3 months 161.70 ± 23.64 10 182.27 ± 32.32 49 0.06 3-6 months 170.33 ± 27.55 15 185.00 ± 28.91 58 0.08 6-12 months 176.44 ± 37.51 16 182.67 ± 26.22 54 0.45 12-24 months 177.14 ± 30.76 14 184.16 ± 30.60 38 0.47 > 24 months 186.92 ± 32.27 13 183.39 ± 32.90 38 0.74

Table 7. Maximal systolic blood pressure in different exercise test control periods.

Robinson index changes between patients groups was not ascertained.

parameter to ST-segment changes in restenosis diagnostics (Table 8., Figure 5.).

The difference in analysis of double product (maximal rate-pressure product) or Robinson index was expected, on basis of previously obtained results. The following trend was observed – patients with coronary artery restenosis at control coronary angiography 1-3 months after performed PCI showed lower parameters of Robinson index. In turn, independently of stenosis location – LM or other coronary artery, or patients with new detected coronary artery stenosis – statistically significant difference and association with

Variability of Robinson index changes on exercise test could become possible additional

**Robinson index (mean ± SD)** 

**'Restenosis' group 'Non-restenosis'** 

Patients with angiographically detected coronary artery stenosis have achieved lower physical load (in Watts), not achieving submaximal pulse, because of termination the veloergometry, if ischemic changes or typical chest pain is appearing. This explains the curve difference of Robinson index in patients with and without coronary artery restenosis, also, statistically significant difference between mentioned groups is not present in analysis of maximal achieved heart rate and maximal systolic blood pressure. However, oscillations of Robinson index parameters are obviously seen in patiens of 'restenosis' group in relative

1-3 months 185.70 ± 32.96 217.51 ± 54.09 0.08 3-6 months 208.12 ± 56.00 226.73 ± 53.81 0.24 6-12 months 210.41 ± 51.57 223.22 ± 45.16 0.34 12-24 months 197.89 ± 27.17 223.41 ± 50.73 0.08 > 24 months 210.23 ± 46.75 221.48 ± 50.63 0.49

Table 8. Robinson index analysis in different exercise test control periods.

**group** 

test 1-3 months and 3-6 months follow-ups (Table 7.).

**'Restenosis' group** 

Exercise test control (months after PCI)

Exercise test control (months after PCI)


Table 5. Mean values of parameters analyzed on exercise test follow-up (maximal heart rate, maximal systolic blood pressure, Robinson index).

Indications for control angiography were defined according to ST-segment ischemic changes on ECG and/or typical patients' complaints during exercise test.

Control angiography in follow-up period was performed in 33.8% of cases – 3-6 months after PCI, in 36.8% and in 10.5% of cases – 6-12 months and 12-24 months after PCI, respectively.

### **4.3.2 Analysis of the results of control exercise tests (1-3 and 3-6 months after PCI) and coronary angiography**

There was no significant difference between patients group with LM restenosis and/or other coronary artery restenosis on control coronary angiography and patients group without diagnosed restenosis, analyzing maximal achieved physical load in early follow-up period (control exercise test performed 1-3 and 3-6 months after PCI).

Statistically significant difference (in maximal achieved physical load analysis) was observed comparing with the patients group with new coronary artery stenosis on control coronary angiography: these patients achieved lower maximal load on 1-3 months control exercise test – 90 ± 17 W vs. 129 ± 26 W, accordingly (p=0.027).

There was no significant difference observed between patients with and without LM restenosis (or other artery restenosis) in analysis of achieved maximal heart rate during exercise test 1-3 and 3-6 months after PCI, also between patients groups with and without diagnosed new artery stenosis on coronary angiography (Table 6.).


Table 6. Maximal heart rate analysis in different exercise test control periods.

(bpm ± SD) 118 ± 16 122 ± 17 121 ± 16 119 ± 15 119 ± 15

pressure (mmHg ± SD) 180 ± 31 181 ± 29 182 ± 29 183 ± 31 184 ± 32

Table 5. Mean values of parameters analyzed on exercise test follow-up (maximal heart rate,

Indications for control angiography were defined according to ST-segment ischemic changes

Control angiography in follow-up period was performed in 33.8% of cases – 3-6 months after PCI, in 36.8% and in 10.5% of cases – 6-12 months and 12-24 months after PCI,

**4.3.2 Analysis of the results of control exercise tests (1-3 and 3-6 months after PCI)** 

There was no significant difference between patients group with LM restenosis and/or other coronary artery restenosis on control coronary angiography and patients group without diagnosed restenosis, analyzing maximal achieved physical load in early follow-up

Statistically significant difference (in maximal achieved physical load analysis) was observed comparing with the patients group with new coronary artery stenosis on control coronary angiography: these patients achieved lower maximal load on 1-3 months control

There was no significant difference observed between patients with and without LM restenosis (or other artery restenosis) in analysis of achieved maximal heart rate during exercise test 1-3 and 3-6 months after PCI, also between patients groups with and without

> **Maximal heart rate (mean ± SD)**

> > **'Nonrestenosis' group**

**Number of the patients**

**p (between groups)** 

**Number of the patients**

1-3 months 114.70 ± 8.87 10 118.49 ± 19.96 49 0.49 3-6 months 120.93 ± 17.43 15 122.07 ± 16.92 58 0.82 6-12 months 119.06 ± 14.63 16 120.69 ± 15.39 55 0.71 12-24 months 112.79 ± 10.03 14 120.58 ± 15.71 38 0.09 > 24 months 112.4 6± 13.48 13 120.74 ± 15.90 38 0.10

Table 6. Maximal heart rate analysis in different exercise test control periods.

**1-3 months 3-6 months 6-12 months 12-24 months >24 months** 

213.27±51.02 220.93±52.55 221.55±45.9 217.84±47.68 219.32±48.46

**Data of exercise test/ follow-up period** 

Maximal heart rate

respectively.

**and coronary angiography** 

Exercise test control (months after PCI)

Maximal systolic blood

Robinson index(max heart rate x max systolic blood pressure/100) ± SD

maximal systolic blood pressure, Robinson index).

on ECG and/or typical patients' complaints during exercise test.

period (control exercise test performed 1-3 and 3-6 months after PCI).

exercise test – 90 ± 17 W vs. 129 ± 26 W, accordingly (p=0.027).

diagnosed new artery stenosis on coronary angiography (Table 6.).

**'Restenosis' group** 

In maximal systolic blood pressure analysis, almost statistically significant difference was observed (trend) between 'restenosis' group and 'non-restenosis' patients group on exercise test 1-3 months and 3-6 months follow-ups (Table 7.).


Table 7. Maximal systolic blood pressure in different exercise test control periods.

The difference in analysis of double product (maximal rate-pressure product) or Robinson index was expected, on basis of previously obtained results. The following trend was observed – patients with coronary artery restenosis at control coronary angiography 1-3 months after performed PCI showed lower parameters of Robinson index. In turn, independently of stenosis location – LM or other coronary artery, or patients with new detected coronary artery stenosis – statistically significant difference and association with Robinson index changes between patients groups was not ascertained.

Variability of Robinson index changes on exercise test could become possible additional parameter to ST-segment changes in restenosis diagnostics (Table 8., Figure 5.).


Table 8. Robinson index analysis in different exercise test control periods.

Patients with angiographically detected coronary artery stenosis have achieved lower physical load (in Watts), not achieving submaximal pulse, because of termination the veloergometry, if ischemic changes or typical chest pain is appearing. This explains the curve difference of Robinson index in patients with and without coronary artery restenosis, also, statistically significant difference between mentioned groups is not present in analysis of maximal achieved heart rate and maximal systolic blood pressure. However, oscillations of Robinson index parameters are obviously seen in patiens of 'restenosis' group in relative

The Role of Exercise Test After Percutaneous Coronary Intervention 259

**ST/pulse index (mean ± SD)** 

Table 9. ST/pulse (ST/p) index in LM patients' group according to angiographic finding (restenosis, new stenosis of coronary artery or without detected restenosis or new stenosis of

Restenosis 50% 43% 7%

Table 10. Restenosis and ST-segment changes on exercise ECG prior the control coronary

The sensitivity of veloergometry test in 1-3 months – 29%, specificity – 100 %. Lower

Correlation of ST-segment changes with restenosis development is obvious also in later control of veloergometry (3-6 months after PCI). The sensitivity of veloergometry test – 50%, specificity – 86 %. All the patients with diagnosed new coronary artery stenosis (in other coronary artery) at control coronary angiography (3-6 months after PCI) showed STsegment depression on exercise ECG in early stress test control (1-3 months), but in patients group without restenosis ECG without any dynamical changes is observed in 92.9% of cases.

Basing on the previously obtained data and on the significance of exercise test in patients' follow-up after performed PCI, several algorithms were developed for evaluation of patients' functional status after PCI, also for evaluation of possible treatment strategy of

depression

parameters of Robinson index also were registered in this patients' group.

In LM patients' group analysis – 50% of cases of coronary artery restenosis are detected in early period (3-6 months after PCI), left 43% and 7% – 6-12 and 12-24 months after PCI, respectively. Angiographic results in early coronary angiography follow-up period (3-6 months after performed PCI) correlate with ECG changes on 1-3 months control of physical test: there is statistically significant difference between patients groups – ST-segment depression on exercise ECG is obvious in all patients with restenosis established at control

**'Without new stenosis' group** 

1.07 ± 0.98 6 0.57 ± 0.34 53 0.01

0.80 ± 0.67 7 0.61 ± 0.40 66 0.27

0.76 ± 0.54 11 0.61 ± 0.40 60 0.28

3-6 months 6-12 months 12-24 months

ST-segment depression

**Number of the patients**

**p (between groups)** 

ST-segment depression

**Number of the patients**

Exercise test control

1-3 months after PCI

3-6 months after PCI

6-12 months after PCI

coronary artery).

angiography.

**4.4 Algorithms** 

coronary artery stenosis.

coronary angiography (Table 10.).

ST-segment changes ST-segment

The sensitivity of the test – 33%, specificity – 93 %.

**'New stenosis' group** 

1 – without restenosis on control coronary angiography

2 – with restenosis on control coronary angiography

Fig. 5. Analysis of different exercise test parameters (Robinson index, maximal heart rate and maximal systolic blood pressure) during follow-ups in patients with and without coronary artery restsenosis (on control angiography).

risk period – 6-12 months follow-up. These curves repeatedly are establishing the role of targeted follow-up programme after performed PCI for more precise evaluation of risk of restenosis, taking into account not only patient's complaints and ischemic changes on ECG, but also the changes of Robinson index parameters in dynamics.

It is supposed, that the number of analyzed patients was not sufficient for statistical significance between groups in ST/p index analysis, in order to definitely conclude the role of this parameter in restenosis diagnostics (Table 9.)


Fig. 5. Analysis of different exercise test parameters (Robinson index, maximal heart rate and maximal systolic blood pressure) during follow-ups in patients with and without

risk period – 6-12 months follow-up. These curves repeatedly are establishing the role of targeted follow-up programme after performed PCI for more precise evaluation of risk of restenosis, taking into account not only patient's complaints and ischemic changes on ECG,

It is supposed, that the number of analyzed patients was not sufficient for statistical significance between groups in ST/p index analysis, in order to definitely conclude the role

> **ST/pulse index (mean ± SD)**

> > **'Non-restenosis' group**

0.76 ± 0.55 10 0.59 ± 0.44 49 0.28

0.72 ± 0.35 15 0.61 ± 0.45 58 0.36

0.76 ± 0.47 16 0.59 ± 0.41 55 0.16

**Number of the patients**

**p (between groups)** 

**Number of the patients**

1 – without restenosis on control coronary angiography 2 – with restenosis on control coronary angiography

coronary artery restsenosis (on control angiography).

of this parameter in restenosis diagnostics (Table 9.)

**'Restenosis' group** 

**Exercise test control** 

1-3 months after PCI

3-6 months after PCI

6-12 months after PCI

but also the changes of Robinson index parameters in dynamics.


Table 9. ST/pulse (ST/p) index in LM patients' group according to angiographic finding (restenosis, new stenosis of coronary artery or without detected restenosis or new stenosis of coronary artery).

In LM patients' group analysis – 50% of cases of coronary artery restenosis are detected in early period (3-6 months after PCI), left 43% and 7% – 6-12 and 12-24 months after PCI, respectively. Angiographic results in early coronary angiography follow-up period (3-6 months after performed PCI) correlate with ECG changes on 1-3 months control of physical test: there is statistically significant difference between patients groups – ST-segment depression on exercise ECG is obvious in all patients with restenosis established at control coronary angiography (Table 10.).


Table 10. Restenosis and ST-segment changes on exercise ECG prior the control coronary angiography.

The sensitivity of veloergometry test in 1-3 months – 29%, specificity – 100 %. Lower parameters of Robinson index also were registered in this patients' group.

Correlation of ST-segment changes with restenosis development is obvious also in later control of veloergometry (3-6 months after PCI). The sensitivity of veloergometry test – 50%, specificity – 86 %. All the patients with diagnosed new coronary artery stenosis (in other coronary artery) at control coronary angiography (3-6 months after PCI) showed STsegment depression on exercise ECG in early stress test control (1-3 months), but in patients group without restenosis ECG without any dynamical changes is observed in 92.9% of cases. The sensitivity of the test – 33%, specificity – 93 %.

### **4.4 Algorithms**

Basing on the previously obtained data and on the significance of exercise test in patients' follow-up after performed PCI, several algorithms were developed for evaluation of patients' functional status after PCI, also for evaluation of possible treatment strategy of coronary artery stenosis.

The Role of Exercise Test After Percutaneous Coronary Intervention 261

**Exercise stress test** 

**Incomplete revascularization**  (hemodinamically significant stenosis on coronary


**Next step PCI**  (invasive cardiologist visit)

exercise ECG

exercise ECG

Fig. 7. Exercise test algorithm in patients after PCI (incomplete revascularization).

Ischemic changes **Next step PCI** 

control coronary angiography.

No ischemic changes **- Medical treatment - Exercise test every 6** 

**months** 

**Patients with incomplete revascularization ( Figure 7.):** 


**Myocardial perfusion scintigraphy or stress echocardiography**

In case of angina and ischemic changes on exercise ECG patients should undergo

 Exercise test should be performed in 1-3 months after PCI in patients with incomplete revascularization (and hemodinamically sigfnificant stenosis, left on angiography); Myocardial perfusion scintigraphy investigation or stress echocardiography should be performed in case of no angina and no ischemic changes on exercise ECG. If ischemic

Fig. 6. Exercise test algorithm in patients after PCI (complete revascularization).

### **Patients with complete revascularization (Figure 6):**


**Complete revascularization** (without hemodinamically significant stenosis on coronary angiography)


**Myocardial perfusion scintigraphy or stress echocardiography (average 4 months after PCI)** 


> **Control coronary angiography**

Ischemic changes **Control coronary angiography** 

Fig. 6. Exercise test algorithm in patients after PCI (complete revascularization).

Exercise test should be performed in 1-3 months after PCI in patients with complete

Patients should continue medical treatment and exercise test control every 6 months, if

 Patients should undergo myocardial perfusion scintigraphy investigation or stress echocardiography in case of no angina, but with presentation of ischemic changes on exercise ECG. If ischemic changes are not present on additional examination tests, medical treatment shoul be continued and control of exercise test every 6 months;

**Patients with complete revascularization (Figure 6):** 

**months** 


**- Medical treatment - Exercise test every 6** 

exercise ECG

**months** 

revscularization (performed with PCI method);

No ischemic changes **- Medical treatment - Exercise test every 6** 

**Exercise stress test after 1-3 and 6 months** 

angina or ischemic changes on exercise ECG are not present;

Fig. 7. Exercise test algorithm in patients after PCI (incomplete revascularization).

 In case of angina and ischemic changes on exercise ECG patients should undergo control coronary angiography.

### **Patients with incomplete revascularization ( Figure 7.):**


The Role of Exercise Test After Percutaneous Coronary Intervention 263

 Patients should continue medical treatment and should perform exercise test every 6 months in case of no hemodinamically significant stenosis on coronary angiography; Myocardial perfusion scintigraphy investigation or stress echocardiography should be performed in case of ischemic changes on exercise ECG. If If ischemic changes are present (in additional investigations), coronary angiography should be performed. If ischemic changes are not present, patients should continue medical treatment and

 Patients with incomplete revascularization and angiographycally with hemodynamically significant stenosis are left, exercise test should be performed in 1-3 months after PCI. If ischemic changes are not present, patients continue medical treatment and perform exercise test every 6 months. In case of ischemic changes on exercise test, patient should

In this study we raised the question whether physical exercise test alone beside accurate diagnosis of obstructive lesions provides an opportunity to prognose cardiovascular risk. We claim that this is possible, which is demonstrated in our study by a significant number of patients with silent ischemia. The potential gain from reducing the risk of reverse cardiac events is actually greatest in those with silent ischemia at the time of follow-up presentation, because those patients are at especially high risk (Pepine and Deedwania, 1994). Nowadays it is clear that chest pain in history does not correlate tightly with ischemic ST-segment deviation following physical loading and with haemodynamically significant stenosis on

Published data as well as national and international guidelines state that physical exercise test following PCI is a useful method in diagnosis of restenosis in high risk or symptomatic patients and should be conducted six months after intervention (Gibbons et al., 2002). Our results showed that the exercise test should be done not later than three to six months after intervention. The highest rate of restenosis was observed in patients after interventional treatment of LM disease in 1-3 months after procedure. The importance of the physical test does not decrease later; more than third part of all documented restenosis was registered at six months after intervention (including the patients with newly diagnosed stenosis on coronary angiography). One year after PCI the specific weight of restenosis was 16.7% of all

The clinical value of exercise electrocardiograms following interventional therapy in the early period is high in evaluation of early post-interventional results (Roffi et al., 2003). This is supported with our results, which demonstrated high specificity of exercise test performed in the first six months after PCI. Exercise electrocardiograms 1–3 months after PCI did not show a shift of the ST-segment for all patients without restenosis. However, angiography depression of the ST-segment was registered on exercise electrocardiogram only for patients with restenosis. This demonstrates a clear correlation between ST-segment

Adequate implementation of drug therapies depending on patient clinical and functional status allow to reach goals set by national and international guidelines. Based on epidemiological studies, even small decreases in low density lipoprotein cholesterol and blood pressure levels translate into significant reductions in cardiovascular morbidity and mortality (LaRosa et al., 2005). Moreover, especially important is the patient compliance,

which can be achieved through frequently scheduled visits.

visit invasive cardiologist in order to make decision of next step PCI.

perform exercise test every 6 months;

**5. Discussion** 

angiography either.

diagnosed restenosis.

depression and restenosis.

changes are present (in additional investigations), next step PCI should be performed. If ischemic changes are not present, patients should continue medical treatment and perform exercise test every 6 months;

 In case of ischemic changes on exercise ECG, independently of patient's complaints, next step PCI should be performed.

Fig. 8. Exercise test algorithm in patients after LM PCI.

### **Patients after left main coronary artery (LM) revascularization (Figure 8.):**

 Exercise test should be performed in 1-3 months after LM PCI. Myocardial perfusion scintigraphy could be performed in 4-6 months in case of inconclusive ECG;


### **5. Discussion**

262 Advances in Electrocardiograms – Clinical Applications

In case of ischemic changes on exercise ECG, independently of patient's complaints,

**Patients after left main coronary artery (LM) PCI** 

> **Incomplete revascularization**  (hemodinamically significant stenosis on coronary angiography)

> > Ischemic changes **Next step PCI**  (invasive cardiologist visit)

No ischemic changes **- Medical treatment - Exercise test or myocardial perfusion scintigraphy every 6** 

**months** 

perform exercise test every 6 months;

next step PCI should be performed.

**Complete revascularization**  (without hemodinamically significant stenosis on coronary angiography)

No ischemic changes **- Medical treatment - Exercise test or myocardial perfusion scintigraphy every 6** 

**months** 

**Exercise stress test after 1-3 and 6 months or myocardial perfusion scintigraphy after 4-6 months** 

> Ischemic changes **Myocardial perfusion scintigraphy or stress echocardiography**

Fig. 8. Exercise test algorithm in patients after LM PCI.

Ischemic changes **Coronary angiography**

**Patients after left main coronary artery (LM) revascularization (Figure 8.):** 

**6 months** 

Exercise test should be performed in 1-3 months after LM PCI. Myocardial perfusion

No ischemic changes **- Medical treatment - Exercise test or every** 

scintigraphy could be performed in 4-6 months in case of inconclusive ECG;

changes are present (in additional investigations), next step PCI should be performed. If ischemic changes are not present, patients should continue medical treatment and

> In this study we raised the question whether physical exercise test alone beside accurate diagnosis of obstructive lesions provides an opportunity to prognose cardiovascular risk. We claim that this is possible, which is demonstrated in our study by a significant number of patients with silent ischemia. The potential gain from reducing the risk of reverse cardiac events is actually greatest in those with silent ischemia at the time of follow-up presentation, because those patients are at especially high risk (Pepine and Deedwania, 1994). Nowadays it is clear that chest pain in history does not correlate tightly with ischemic ST-segment deviation following physical loading and with haemodynamically significant stenosis on angiography either.

> Published data as well as national and international guidelines state that physical exercise test following PCI is a useful method in diagnosis of restenosis in high risk or symptomatic patients and should be conducted six months after intervention (Gibbons et al., 2002). Our results showed that the exercise test should be done not later than three to six months after intervention. The highest rate of restenosis was observed in patients after interventional treatment of LM disease in 1-3 months after procedure. The importance of the physical test does not decrease later; more than third part of all documented restenosis was registered at six months after intervention (including the patients with newly diagnosed stenosis on coronary angiography). One year after PCI the specific weight of restenosis was 16.7% of all diagnosed restenosis.

> The clinical value of exercise electrocardiograms following interventional therapy in the early period is high in evaluation of early post-interventional results (Roffi et al., 2003). This is supported with our results, which demonstrated high specificity of exercise test performed in the first six months after PCI. Exercise electrocardiograms 1–3 months after PCI did not show a shift of the ST-segment for all patients without restenosis. However, angiography depression of the ST-segment was registered on exercise electrocardiogram only for patients with restenosis. This demonstrates a clear correlation between ST-segment depression and restenosis.

> Adequate implementation of drug therapies depending on patient clinical and functional status allow to reach goals set by national and international guidelines. Based on epidemiological studies, even small decreases in low density lipoprotein cholesterol and blood pressure levels translate into significant reductions in cardiovascular morbidity and mortality (LaRosa et al., 2005). Moreover, especially important is the patient compliance, which can be achieved through frequently scheduled visits.

The Role of Exercise Test After Percutaneous Coronary Intervention 265

2. The database of the patients with left main coronary artery stenosis was analyzed, evaluating the efficacy of invasive treatment of left main stenosis, analyzing possible

3. New specific criteria were extracted for risk evaluation and detection with exercise

4. Algorithms were developed for CAD patients' clinical status evaluation after

5. Follow-up programme with targeted exercise test allows to personalize exact treatment

Arab D,Valeti V, Schunemann H, Lopez-Candales A (2000) Usefulness of the QTc Interval in

Babapulle M, Diodati J, Blankenship J, Huynh T, Cugno S, Puri R, Nguyen P, Eisenberg M

Bengtson J, Mark D, Honan M, et al (1990) Detection of Restenosis after Elective

Berntsen R, Gjestvang F, Rasmussen K (1995) QRS Prolongation as an Indicator of Risk of

Califf R, Mark D, Harrell F (1988) Importance of Clinical Measures of Ischemia in the

Cecchi F, Olivotto I, Gistri R, Lorenzoni R, Chiriatti G, Camici P (2003) Coronary

Chalela W, Kreling J, Falcao A, Hueb W, Moffa P, Pereyra P, Ramires J (2006) Exercise Stress

Angioplasty. Brazilian Journal of Medical and Biological Research 39:475-482. Cerqueira M D, Weissman N J, Dilsizian V, Jacobs A K, Kaul S, Lasney W K, Pennel D J,

Erglis A et al. The guidelines of the Latvian Society of Cardiology on prevention and

Predicting Myocardial Ischemia in Patients undergoing Exercise Stress Testing. Am

(2007) Utility of Routine Exercise Treadmill Testing early after Percutaneous

Percutaneous Transluminal Coronary Angioplasty using the Exercise Treadmill

Ischemia-Related Ventricular Tachycardia and Fibrillation Induced by Exercise. Am

Prognosis of Patients with Documented Coronary Artery Disease J Am Coll

Microvascular Dysfunction and Prognosis in Hypertrophic Cardiomyopathy. N

Testing before and after successsful Multivessel Percutaneous Transluminal Coronary

Rumberger J A, Ryan T, Verani M S (2002) Standardized Myocardial Segmentation and Nomenclature for Tomographic Imaging of the Heart. A Statement for Healthcare Professionals From the Cardiac Imaging Committee of the Council on Clinical Cardiology of the American Heart Association. Circulation 105:539-542. Eisenberg M, Schechter D, Lefkovits J, Goudreau E, Deligonul U, Mak K, Del Core M, Duerr

R, Garzon P, Huynh T, Smilovitch M, Sedlis S, Brown D, Brieger D, Pilote L (2001) ROSETTA Investigators: Use of Routine Functional Testing after Percutaneous Transluminal Coronary Angioplasty: Results from the ROSETTA Registry. Am

management of stable coronary heart disease, Riga, 2007.Fletcher G, Balady G, Froelicher V, Hartley L, Haskell W, Pollock M (1995) Exercise Standards. A

progression of restenosis and evaluating following treatment steps.

strategy and risk factor correction for each patient individually.

Coronary Intervention. BMC Cardiovascular Disorders 7.

balloon angioplasty, etc.)

percutaneous coronary intervention.

J Cardiol 85:764-766.

Heart J 129:542-548.

Cardiol 11:20-26.

Heart J 141:837-846

Test. Am J Cardiol 65:28-34.

Engl J Med 349:1027-1035.

stress test.

**7. References** 

with CAD, evolving different invasive treatment methods (with drug-eluting stents,

The clinical course and prognosis of patients with coronary artery disease can be modified favourably by successful translation of recommendations for secondary coronary prevention into effective clinical care. Crucial to the outcome of any preventive strategy is patient attitude to lifestyle modification and compliance with drug therapies over the long term. We tend to claim that an aggressive follow-up programme could help us to reach the goals in secondary prevention of coronary artery disease set by European Societies and Latvian Society of Cardiology.

The physical exercise test is a safe method with high specificity, but is limited by poor test sensitivity for the evaluation of efficacy of interventional and medical treatment for patients with coronary artery disease. A focussed exercise test performed on a regular basis indirectly influences clinical results and prognosis. Timely set diagnosis of restenosis provides necessary treatment measures, therefore, alienating adverse cardiac events such as unstable angina and myocardial infarction. Moreover, an exercise test performed on a regular basis provides essential corrections in drug therapy. After comparison of drug therapies recommended three months and one year after PCI, we can conclude that regular follow-ups provide the opportunity to sustain acceptable patient compliance and use of medications even twelve months after an intervention.

A focussed follow-up programme with an exercise test allows to evaluate clinical status of patients as well as to determine timely possible risk of restenosis, to adapt medication doses, to reduce risk factors and to influence positively patient compliance. The exercise test provides accurate estimation of possible restenosis in patients with complete revascularization. In patients with incomplete revascularization the exercise test specificity is reduced. Taking into account the results of the current study, we are sure, that focussed physical exercise test should be advised for all patients after interventional treatment. It is of high importance to achieve submaximal heart rate during the exercise test. In all cases when this is not possible for any reason, myocardial perfusion scintigraphy is indicated.

### **Advantages of the study (benefits)**

The results of this study can be used with a goal to determine possible risk of restenosis in a time, to prescribe medications according to patient's functional status, to correct possible risk factors, which can have negative influence on development of disease.

Regular patients' follow-ups increase the compliance of the patients and 'participation' in treatment process, thus increasing accuracy of data and foundation of the results.

### **Possible disadvantages (limits) of the study**

Irregular patients' follow-ups, thus exclusion from statistycal analysis.

Insufficient information about patient recurrent hospitalization at district hospital. Also the variability in medical treatment (general practitioner like "middle stage" between patient and cardiologist).

The lack of patients compliance in medication use, the lack of motivation in modification of their possible risk factors.

Low availability of myocardial perfusion scintigraphy, thus, not possibility to perform necessary investigation in position of diagnosis making and more detailed evaluation of clinical situation.

### **6. Conclusions**

1. Exercise test (veloergometry) is established as a safe method with high specificity, but lower sensitivity in evaluation of invasive and medical treatment efficiacy in patients with CAD, evolving different invasive treatment methods (with drug-eluting stents, balloon angioplasty, etc.)


### **7. References**

264 Advances in Electrocardiograms – Clinical Applications

The clinical course and prognosis of patients with coronary artery disease can be modified favourably by successful translation of recommendations for secondary coronary prevention into effective clinical care. Crucial to the outcome of any preventive strategy is patient attitude to lifestyle modification and compliance with drug therapies over the long term. We tend to claim that an aggressive follow-up programme could help us to reach the goals in secondary prevention of coronary artery disease set by European Societies and Latvian

The physical exercise test is a safe method with high specificity, but is limited by poor test sensitivity for the evaluation of efficacy of interventional and medical treatment for patients with coronary artery disease. A focussed exercise test performed on a regular basis indirectly influences clinical results and prognosis. Timely set diagnosis of restenosis provides necessary treatment measures, therefore, alienating adverse cardiac events such as unstable angina and myocardial infarction. Moreover, an exercise test performed on a regular basis provides essential corrections in drug therapy. After comparison of drug therapies recommended three months and one year after PCI, we can conclude that regular follow-ups provide the opportunity to sustain acceptable patient compliance and use of

A focussed follow-up programme with an exercise test allows to evaluate clinical status of patients as well as to determine timely possible risk of restenosis, to adapt medication doses, to reduce risk factors and to influence positively patient compliance. The exercise test provides accurate estimation of possible restenosis in patients with complete revascularization. In patients with incomplete revascularization the exercise test specificity is reduced. Taking into account the results of the current study, we are sure, that focussed physical exercise test should be advised for all patients after interventional treatment. It is of high importance to achieve submaximal heart rate during the exercise test. In all cases when this is not possible

The results of this study can be used with a goal to determine possible risk of restenosis in a time, to prescribe medications according to patient's functional status, to correct possible

Regular patients' follow-ups increase the compliance of the patients and 'participation' in

Insufficient information about patient recurrent hospitalization at district hospital. Also the variability in medical treatment (general practitioner like "middle stage" between patient

The lack of patients compliance in medication use, the lack of motivation in modification of

Low availability of myocardial perfusion scintigraphy, thus, not possibility to perform necessary investigation in position of diagnosis making and more detailed evaluation of

1. Exercise test (veloergometry) is established as a safe method with high specificity, but lower sensitivity in evaluation of invasive and medical treatment efficiacy in patients

Society of Cardiology.

medications even twelve months after an intervention.

for any reason, myocardial perfusion scintigraphy is indicated.

risk factors, which can have negative influence on development of disease.

Irregular patients' follow-ups, thus exclusion from statistycal analysis.

treatment process, thus increasing accuracy of data and foundation of the results.

**Advantages of the study (benefits)** 

and cardiologist).

clinical situation.

**6. Conclusions** 

their possible risk factors.

**Possible disadvantages (limits) of the study** 


The Role of Exercise Test After Percutaneous Coronary Intervention 267

Lauer M, Francis G, Okin P, Pashkow F, Snader C, Marwick T (1999) Impaired chronotropic Response to Exercise Stress Testing as a Predictor of Mortality. JAMA 281:524-529. Libby P, Theroux P (2005) Pathophysiology of Coronary Artery Disease. Circulation

Lombard A, Loperfido F, Pennestri F, Rossi e, Patrizi R, Cristinziani G, Catapano G, Maseri

Margonato A, Chierchia S, Xuereb R, Xuereb M, Fragasso G, Cappelletti A, Landoni C,

Mark D, Shaw L, Harrell F, Hlatky M, Lee K, Bengtson J, McCants C, Califf R, Pryor D (1991)

Mastouri R, Sawada SG, Mahenthiran J (2010) Current Noninvasive Imaging Technique for

Mehta S, Yusuf S, Peters R, Bertrand M, Lewis B, Natarjan M, Malmberg K, Rupprecht H,

Mercado N, Boersma E, Wijns W, Gersh B, Morillo C, de Valk V, van Es G, Grobbee D,

Michaelides A, Ryan J, Bacon J, Pozderac R, Toutouzas P, Boudoulas H (1995) Exercise-

Michaelides A, Ryan J, VanFoseen D, Pozderac R, Boudoulas H (1993) Exercise-Induced

Michaelides A, Triposkiadis F, Boudoulas H, Spanos A, Papadopoulos P, Kourouklis K,

Mintāle I, Erglis A. Recommendations of Latvian Society of Cardiology on Physical Exercise

Miranda C, Lehmann K, Froelicher V (1991) Correlation between Resting ST Segment

Miranda C, Lehmann K, Lachterman B, Coodley E, Froelicher V (1991) Comparison of Silent

Myers J, Prakash M, Froelicher V, Do D, Partington S, Atwood J (2002) Exercise Capacity and Mortality among Men referred for Exercise Testing. N Engl J Med 346:793-801.

Disease: a Marker of Myocardial Ischemia. J Cardiol 26:263-272.

Coronary Artery Disease. N Engl J Med 325:849-853.

Intervention: the PCI-CURE Study. Lancet 358:524-533.

Myocardial Ischemia. Am Heart J 126:1320-1325.

QRS Changes. Am Heart J 120:292-302.

Testing in Q Wave Myocardial Infarction. J Am Coll Cardiol 27:599-605. Mackay J, Mensah G (2004) Atlas of Heart Disease and Stroke. World Health Organization,

A (1996) Significance of Transient ST-T Segment Changes during Dobutamine

Lucignani G, Fazio F (1995) Specificity and Sensitivity of Exercise-Induced ST Segment Elevation for Detection of Residual Viability: Comparison with Fluorodeoxyglucose and Positron Emission Tomography. J Am Coll Cardiol

Prognostic Value of a Treadmill Exercise Score in Outpatients with Suspected

Detection of Coronary Artery Disease. Expert Review of Cardiovascular Therapy

Zhao F, Chrolavicius S (2001) Effects of Pretreatment with Clopidogrel and Aspirin followed by long-term Therapy in Patients undergoing Percutaneous Coronary

Serruys P (2001) Clinical and Quantitative Coronary Angiographic Predictors of Coronary Restenosis: a comparative Analysis from the Baloon-to-Stent Era. J Am

induced QRS Changes (Athens QRS Score) in Patients with Coronary Artery

QRS Prolongation in Patients with Coronary Artery Disease: a Marker of

Toutouzas P (1990) New Coronary Artery Disease Index based on Exercise-Induced

Depression, Exercise Testing, Coronary Angiography, and Long-Term Prognosis.

and Symptomatic Ischemia during Exercise Testing in Men. Ann Intern Med

111:3481-3488.

Geneva.

25:1032-1038.

8:77-91.

Coll Cardiol 38.

Tests, Riga, 2008.

114:649-656.

Am Heart J 122:1617-1628.

Statement for Healthcare Professionals from the American Heart Association. Writing Group. Circulation 91:580-615.


Foult J, Hoffman O, Ducardonnet A, Attal B, Uzan L, Weinmann P, Verdier J (2009) Exercise

Fox K, Garsia M, Ardissino D, Buszman P, Camici P, Crea F, Daly C, De Backer G, Hjemdahl

T (2006) Stable angina pectoris (Management of). ESC Practice guidelines. Gianrossi R, Detrano R, Mulvihill D, Lehmann K, Dubach P, Colombo A, McArthur D,

Gibbons R, Balady G, Bricker J, Chaitman B, Fletcher G, Froelicher V, Mark D, McCallister B,

Hamasaki S, Arima S, Tahara M, Kihara K, Shono H, Nakao S, Tanaka H (1996) Increase in the

Percutaneous Transluminal Coronary Angioplasty. Am J Cardiol 78:990-995. Hamasaki S, Abematsu H, Arima S, Tahara M, Kihara K, Shono H, Nakao S, Tanaka H. A

Hesse B, Morise A, Pothier C, Blackstone E, Lauer M (2005) Can we Reliably Predict Long-

Hjalmarson A, Goldstein S, Fagerberg B, Wedel H, Waagstein F, Kjekshus J, Wikstrand J,

Holmes DR, Vlietstra RE, Smith HC, Vetrovec GW, Kent KM, Cowley MJ, Faxon DP,

the National Heart, Lung, and Blood Institute Am J Cardiol 53:C77-C81. Imai K, Sato H, Hori M, Kusuoka H, Ozaki H, Yokoyama H, Takeda H, Inoue M, Kamada T

Kligfield P, Lauer M (2006) Exercise Electrocardiogram Testing. Beyond the ST Segment.

Kligfield P, Okin P (1994) Evolution of the Exercise Electrocardiogram. Am J Cardiol

Kwok J, Miller T, Christian T, Hodge D, Gibbons R (1999) Prognostic Value of a Treadmill

LaRosa J, Grundy S, Waters D, Shear C, Barter P, Fruchart J, Gotto A, Greten H, Kastelein J,

(MERIT-HF), MERIT-HF Study Group. JAMA 283:1295-1302.

Jawad E, Arora R (2008) Chronic Stable Angina Pectoris. Dis Mon 54:671-689.

with Stable Coronary Disease. N Engl J Med 352:1425-1435.

Artery Disease: a Meta -Analysis. Circulation 80:87-98.

Writing Group. Circulation 91:580-615.

ACC/AHA Practice Guidelines.

80: 411–415.

150:307-314.

1994:1529-1535.

73:1209-1210.

Circulation 114:2070-2082.

Resting ECG JAMA 282:1047-1053.

Testing in Cardiology.

Statement for Healthcare Professionals from the American Heart Association.

P, Lopez-Sendon H, Marco J, Morais J, Pepper J, Sechtem U, Simoons M, Thygesen

Froelicher V (1989) Exercise Induced ST Depression in the Diagnosis of Coronary

Mooss A, O'Reilly M, Winters W (2002) ACC/AHA 2002 Guideline Update for Exercise Testing. A Report of the American College of Cardiology/ American Heart Association Task Force on Practice Guidelines (Committee on Exercise Testing).

delta ST/ delta Heart Rate (HR) Index: a new Predictor of Restenosis after successful

new predictor of restenosis after successful percutaneous transluminal coronary angioplasty in patients with multivessel coronary artery disease. *Am J Cardiol.* 1997;

Term Mortality After Exercise Testing? An External Validation Am Heart J

AlAllaf D, Vitovec J, Aldershvile J (2000) Effects of controlled-release Metoprolol on total Mortality, Hospitalizations, and well-being in Patients with Heart Failure: the Metoprolol CR/XL Randomized Intervention Trial in Congestive Heart Failure

Gruentzig AR, Kelsey SF, Detre KM (1984) Restenosis after Percutaneous Transluminal Coronary Angioplasty (PTCA): A Report from the PTCA Registry of

(1994) Vagally mediated Heart Rate Recovery after Exercise is accelerated in Athletes but blunted in Patients with Chronic Heart Failure. J Am Coll Cardiol

Exercise Score in Symptomatic Patients with Nonspecific ST-T Abnormalities on

Shepherd J, Wenger N (2005) Intensive Lipid lowering with Atorvastatin in Patients


**Part 4** 

**Cardiotoxicology** 


## **Part 4**

**Cardiotoxicology** 

268 Advances in Electrocardiograms – Clinical Applications

Okin P, Kligfield P (1995) Heart Rate Adjustment of ST Segment Depression and

Okin P, Kligfield P (1995) Heart Rate Adjustment of ST Segment Depression and

Pepine C, Cohnm P, Deedwania P (1994) Effects of Treatment on Outcome in Mildly

Renkin J, Melin J, Robert A (1990) Detection of Restenosis after Successful Coronary

Ruygrok P, Webster M, De Valk V, van Es G, Ormiston J (2001) Clinical and Angiographic

Serruys PW, Luijten HE, Beatt KJ, Gauskens R, de Feyter PJ, van den Brand M, Reiber JH, ten

Sketch M, Mooss A, Butler M, Nair C, Mohiuddin S (1981) Digoxin-induced Positive Exercise Tests: their Clinical and Prognostic Significance. Am J Cardiol 48:655-659. Strauss HW, Miller D, Wittry M, Cerqueira MD, Garcia EV, Iskandrian AS, Schelbert HR,

Stuart RJ, Ellestad M (1980) National Survey of Exercise Stress Testing Facilities. Chest 77:94-

Viik J (2000) Diagnostic Properties of Exercise Electrocardiographic Leads and Variables in

Weinsaft J, Wong F, Walden J, Szulc M, Okin P, Kligfield P (2005) Anatomic Distribution of

Wenaweser P, Surmely J-F, Windecker S, Roffi M, Togni M, Billinger M, Cook S, Vogel R,

Whinnery J, Froelicher V, Stuart A (1997) The Electrocardiographic Response to Maximal

Yusuf S, Sleght P, Pogue J, Bosch J, Davies R, Dagenais G (2000) Effects of an Angiotensin-

Coronary Stent Implantation. Am J Cardiol 101:807-811.

Cardiol 25:1726-1735

Cardiol 25:1726-1735.

Study (ASIST). Circulation 90:762-768.

safe. J Am Coll Cardiol 42:1569-1573.

Intervention. Circulation 104:2289-2294.

Subgroups Circulation 98:1622-1630.

Journal of Nuclear Medicine 39:918-923.

97.

Publications.

Am J Cardiol 96:1356-1360.

Heart J 94:316-324.

Med 342:145-153.

Sheffield L (1991) Upsloping ST Segments. Circulation 84:426-428.

Performance of the Exercise Electrocardiogram: a Critical Evaluation. J Am Coll

Performance of the Exercise Electrocardiogram: a Critical Evaluation. J Am Coll

Symptomatic Patients with Ischemia during Daily Life: the Atenolol Silent Ischemia

Angioplasty: improved Clinical Decision Making with Use of a logistic Model Combining Procedural and Follow-Up Variables. J Am Coll Cardiol 16:1333-1340. Roffi M, Wenaweser P, Windecker S, Mehta H, Eberli F, Seiler C, Fleisch M, Garachemani A,

Pedrazzini G, Hess O, Meier B (2003) Early Exercise Test after Coronary Stenting is

Factors associated with Asymptomatic Restenosis after Percutaneous Coronary

Katen HJ, van Es GA, Hugenholtz PG (1988) Incidence of Restenosis after Successful Coronary Angioplasty: a time-related Phenomenon. A quantitative angiographic Study in 342 consecutive patients at 1, 2, 3, and 4 months. Circulation 77:361-371. Shaw L, Peterson E, Shaw L, Kesler K, DeLong E, Harrell F, Muhlbaier L, Mark D (1998) Use

of a Prognostic Treadmill Score in Identifying Diagnostic Coronary Disease

Wackers FJ (2002) Procedure Guideline for Myocardial Perfusion Imaging The

the Detection of Coronary Artery Disease. Tampere University of Technology

Myocardial Ischemia as a Determinant of Exercise-Induced ST-segment Depression

Seiler C, Hess O, Meier B (2008) Prognostic value of Early Exercise Testing After

Treadmill Exercise in Asymptomatic Men with Left Bundle Branch Block. Am

converting-Enzyme Inhibitor, Ramipril, on Cardiovascular Events in high-risk Patients. The Heart Outcomes Prevention Evaluation Study Investigators. N Engl J

**15** 

*Romania* 

**Toxic and Drug-Induced** 

**Changes of the Electrocardiogram** 

Catalina Lionte, Cristina Bologa and Laurentiu Sorodoc

There are numerous toxins and drugs that can cause, in overdose, electrocardiogram (ECG) changes, even in patients without history of cardiac pathology. The diagnosis and management of patients with an abnormal ECG encountered in a specific toxicity can challenge experienced physicians. One must have serious knowledge of basic cardiac physiology, in order to understand the ECG changes associated with various drugs and

The main mechanisms involved include membrane – depressant action (sodium channel blockers, slow calcium channel blockers, outward potassium (K+) channel blockers, and sodium-potassium adenosine-triphosphatase blockers), and action on autonomic nervous system and its sites of cardiovascular action (beta-adrenergic blockers and other sympathetic-inhibitors, sympathomimetic, anticholinergic and cholinomimetic substances). Many toxins and medications have actions that involve more than one of these mechanisms, including hypoxia, electrolyte and metabolic imbalances, and thus may result in a

In resting state, the myocardial cell membrane is impermeable to positively charged sodium ions (Na+). The Na+/K+ ATPase maintains a negative electric potential of approximately 90 mV in the myocyte. The rapid opening of Na+ channels and massive Na+ influx (phase 0 of action potential) explains depolarization of the cardiac cell membrane (fig.1), causing the rapid upstroke of the cardiac action potential, which is conducted through the ventricles and is expressed as the QRS complex of the ECG. The closure of Na+ channels and the transient opening of Ito K+ efflux channels (phase 1) mark the peak of the action potential. Then, phase 2 of the action potential occurs when the opening of slow calcium (Ca2+) channels produces an influx of positive ions with a steady maintenance of the membrane potential and myocardial contraction continues. The end of the cardiac cycle is marked by the closure of the Ca2+ channels and the activation of the K+ efflux channels, which allow the action potential to return to its resting potential of – 90 mV (phase 3). This K+ efflux from the myocardial cell is directly responsible for the QT interval on the ECG (Holstege et al., 2006). During phase 4 of the cardiac cell action potential, some cardiac fibers allow sodium ions to enter the cell, increasing the resting membrane potential, known as spontaneous diastolic depolarization. When the threshold in membrane potential is reached, the Na+ channels

**1. Introduction** 

combination of electrocardiographic changes.

open and another action potential is generated.

toxins.

*"Gr.T.Popa" University of Medicine and Pharmacy, Iasi,* 

## **Toxic and Drug-Induced Changes of the Electrocardiogram**

Catalina Lionte, Cristina Bologa and Laurentiu Sorodoc *"Gr.T.Popa" University of Medicine and Pharmacy, Iasi, Romania* 

### **1. Introduction**

There are numerous toxins and drugs that can cause, in overdose, electrocardiogram (ECG) changes, even in patients without history of cardiac pathology. The diagnosis and management of patients with an abnormal ECG encountered in a specific toxicity can challenge experienced physicians. One must have serious knowledge of basic cardiac physiology, in order to understand the ECG changes associated with various drugs and toxins.

The main mechanisms involved include membrane – depressant action (sodium channel blockers, slow calcium channel blockers, outward potassium (K+) channel blockers, and sodium-potassium adenosine-triphosphatase blockers), and action on autonomic nervous system and its sites of cardiovascular action (beta-adrenergic blockers and other sympathetic-inhibitors, sympathomimetic, anticholinergic and cholinomimetic substances). Many toxins and medications have actions that involve more than one of these mechanisms, including hypoxia, electrolyte and metabolic imbalances, and thus may result in a combination of electrocardiographic changes.

In resting state, the myocardial cell membrane is impermeable to positively charged sodium ions (Na+). The Na+/K+ ATPase maintains a negative electric potential of approximately 90 mV in the myocyte. The rapid opening of Na+ channels and massive Na+ influx (phase 0 of action potential) explains depolarization of the cardiac cell membrane (fig.1), causing the rapid upstroke of the cardiac action potential, which is conducted through the ventricles and is expressed as the QRS complex of the ECG. The closure of Na+ channels and the transient opening of Ito K+ efflux channels (phase 1) mark the peak of the action potential. Then, phase 2 of the action potential occurs when the opening of slow calcium (Ca2+) channels produces an influx of positive ions with a steady maintenance of the membrane potential and myocardial contraction continues. The end of the cardiac cycle is marked by the closure of the Ca2+ channels and the activation of the K+ efflux channels, which allow the action potential to return to its resting potential of – 90 mV (phase 3). This K+ efflux from the myocardial cell is directly responsible for the QT interval on the ECG (Holstege et al., 2006). During phase 4 of the cardiac cell action potential, some cardiac fibers allow sodium ions to enter the cell, increasing the resting membrane potential, known as spontaneous diastolic depolarization. When the threshold in membrane potential is reached, the Na+ channels open and another action potential is generated.

Toxic and Drug-Induced Changes of the Electrocardiogram 273

Cardiotoxins are responsible of ECG changes through a combination of membrane depressant effects, autonomic disturbances and metabolic changes. The severity of a toxicinduced conduction block varies depending on the toxin involved and its site of action.

Inhibition of the fast Na+ channels, in the phase 0 of the action potential (AP), decreases the rate of rise and amplitude of the AP in Purkinje fibers, and in atrial and ventricular myocardial cells. As a result, the upslope of depolarization is slowed and the QRS complex becomes wide. In a toxicological situation, QRS complex widening likely results directly from Na+ channel blockage or indirectly from toxin-induced hyperkalemia (Holstege et al.,






5. Toxins: Quinine, Saxitoxin, Tetrodotoxin

Ventricular tachycardia (VT) and ventricular fibrillation (VF)

ST/T changes consistent with ischemia (cocaine toxicity)



1. Cardiovascular drugs:

Procainamide)


Moricizine)

Maprotiline)

3. Other drugs: - Amantadine


Asystole

Table 1. Na+ channel blockers and the resulting ECG changes.

ECG changes QRS widening

\*mechanism not involving the beta-receptor.

4. Illicit drugs: Cocaine

Right bundle branch pattern R wave elevation in aVR lead Rightward deviation of QRS axis

Bradycardia with wide QRS complex


**2. Membrane – depressant drugs and toxins** 

**2.1 Sodium channel blockers** 

Inhibitors of fast Na+

channels

Fig. 1. Cardiac cycle action potential with corresponding ion changes across the membrane and electrocardiographic tracing. Dotted line indicates the changes associated with Na+ channel blocker toxicity. Dashed line indicates the changes associated with K+ efflux blocker toxicity. Ito= transient outward K+ current; ICa= L-type Ca2+ current; INa= late sodium channel current; IKr= rapidly activating delayed-rectifier K+ current; IKs= slowly activating delayed rectifier K+ current; IK1= inward rectifier K+ current (adapted from Holstege et al., 2005).

The atrial and ventricular myocardium contraction, and the conduction in the His-Purkinje system depend on sodium entry via the fast sodium channels in phase 0 of the action potential, while the conduction in sinoatrial node and atrioventricular (AV) node depend on Ca2+ entry during phase 0 via the slow Ca2+ channels (Patel & Benowitz, 2005).

Cardiac activity is controlled, among other mechanisms, by the autonomic nervous system. Sympathetic fibers increase the heart rate, the rate of AV nodal conduction and the contracility of the myocardium. The norepinephrine released by postganglionic fibers leads to an interaction with beta 1-adrenergic cardiac receptors, and increasing cells' permeability to Na+ and Ca2+, with an increase of contractility, excitability, and conduction. The parasympathetic postganglionic fibers innervate the sinus node and AV node. Stimulation of muscarinic receptors via releasing of acetylcholine decreases atrial excitability and slows the conduction of impulses to the ventricles (Patel & Benowitz, 2005).

In the setting of drug overdose or of a toxic exposure, ECG abnormalities, especially arrhythmias, are produced by direct or indirect sympathomimetic effects, anticholinergic effects, the effects of altered central nervous system (CNS) regulation of peripheral autonomic system, and myocardial membrane depression. Genesis of arrhythmias in the poisoned patient is based on the same three mechanisms as in an ischemic patient: abnormal impulse formation, abnormal impulse conduction, and triggered activity. Contributing factors to ECG changes are hypotension, hypoxia, acid-base and electrolyte imbalances.

### **2. Membrane – depressant drugs and toxins**

Cardiotoxins are responsible of ECG changes through a combination of membrane depressant effects, autonomic disturbances and metabolic changes. The severity of a toxicinduced conduction block varies depending on the toxin involved and its site of action.

### **2.1 Sodium channel blockers**

272 Advances in Electrocardiograms – Clinical Applications

Fig. 1. Cardiac cycle action potential with corresponding ion changes across the membrane and electrocardiographic tracing. Dotted line indicates the changes associated with Na+ channel blocker toxicity. Dashed line indicates the changes associated with K+ efflux blocker toxicity. Ito= transient outward K+ current; ICa= L-type Ca2+ current; INa= late sodium channel current; IKr= rapidly activating delayed-rectifier K+ current; IKs= slowly activating delayed rectifier K+ current; IK1= inward rectifier K+ current (adapted from Holstege et al., 2005).

The atrial and ventricular myocardium contraction, and the conduction in the His-Purkinje system depend on sodium entry via the fast sodium channels in phase 0 of the action potential, while the conduction in sinoatrial node and atrioventricular (AV) node depend on

Cardiac activity is controlled, among other mechanisms, by the autonomic nervous system. Sympathetic fibers increase the heart rate, the rate of AV nodal conduction and the contracility of the myocardium. The norepinephrine released by postganglionic fibers leads to an interaction with beta 1-adrenergic cardiac receptors, and increasing cells' permeability to Na+ and Ca2+, with an increase of contractility, excitability, and conduction. The parasympathetic postganglionic fibers innervate the sinus node and AV node. Stimulation of muscarinic receptors via releasing of acetylcholine decreases atrial excitability and slows

In the setting of drug overdose or of a toxic exposure, ECG abnormalities, especially arrhythmias, are produced by direct or indirect sympathomimetic effects, anticholinergic effects, the effects of altered central nervous system (CNS) regulation of peripheral autonomic system, and myocardial membrane depression. Genesis of arrhythmias in the poisoned patient is based on the same three mechanisms as in an ischemic patient: abnormal impulse formation, abnormal impulse conduction, and triggered activity. Contributing factors to ECG changes are hypotension, hypoxia, acid-base and electrolyte imbalances.

Ca2+ entry during phase 0 via the slow Ca2+ channels (Patel & Benowitz, 2005).

the conduction of impulses to the ventricles (Patel & Benowitz, 2005).

Inhibition of the fast Na+ channels, in the phase 0 of the action potential (AP), decreases the rate of rise and amplitude of the AP in Purkinje fibers, and in atrial and ventricular myocardial cells. As a result, the upslope of depolarization is slowed and the QRS complex becomes wide. In a toxicological situation, QRS complex widening likely results directly from Na+ channel blockage or indirectly from toxin-induced hyperkalemia (Holstege et al.,


\*mechanism not involving the beta-receptor.

Table 1. Na+ channel blockers and the resulting ECG changes.

Toxic and Drug-Induced Changes of the Electrocardiogram 275

sign of severe poisoning, indicating that the Na+ channel blockade is so profound that tachycardia does not occur, despite the clinical muscarinic antagonism or adrenergic agonism (Holstege et al., 2006). Nevertheless, bradycardia may occur because of slowed

All CCBs (Table 2) inhibit the voltage sensitive L-type Ca2+ channel within the cell membrane. In the pacemaker cells of the sinoatrial node and AV node, the primary ion channel, which controls depolarization, is the slow Ca2+ channel. When inhibited, there is a slowing or an inhibition of the specialized tissue to conduct a cardiac impulse (Patel &




depolarization of pacemaker cells that depend on entry of Na+ ions.

1. Dihydropyridines:

2. Phenylalkylamine:

Reflex tachycardia (ex. Nifedipine) Varying degrees of AV block

Sinus arrest with AV junctional rhythm

In CCB toxicity initially occurs a sinus bradycardia, followed by various degrees of AV block (fig.4), and junctional and ventricular bradydysrhythmias on ECG. Depending on the agent involved, other dysrhythmias may be seen (Gordon, 2006): sinus tachycardia (specifically Nifedipine), atrial arrhythmias, and junctional rhythms (fig. 5, 6,7). A wide QRS complex may appear, caused by ventricular escape rhythms or by CCB-induced Na+ channel blockade which delays of phase 0 of depolarization. Sudden shifts from

In addition, ECG changes associated with cardiac ischemia (fig.8) may occur as a result of the hypotension and changes in the cardiovascular status, especially in patients with pre-


**2.2 Slow Calcium Channel Blockers (CCB)** 

ECG changes Sinus bradycardia

Asystole

bradydysrhythmias to cardiac arrest have been reported.

Wide QRS complex ST/T changes

Table 2. Calcium channel blockers and the resulting ECG changes.

existing cardiac disease (Patel & Benowitz, 2005; Holstege et al., 2006).

Benowitz, 2005).

Inhibitors of slow Ca2+ channels

2005). Direct toxin-induced blockade of cardiac Na+ channels will cause QRS complex widening, and it has been described as a membrane stabilizing effect, a local anesthetic effect, or a quinidine-like effect. Some drugs in this category (Table 1) may also affect other myocardial ion transfers, such as the Ca2+ influx and K+ efflux (Holstege et al., 2006). Other abnormal QRS complex configurations are also possible. In the most severe cases, the QRS complex widening becomes so profound that the ultimate origin of the rhythm disturbance is impossible (fig. 2).

Fig. 2. Na+ channel blocker toxicity (patient with acute Propafenone overdose). Note the wide QRS complex, at a rate of 134/min, which suggests, at a first view, monomorphic VT.

R wave elevation in aVR ≥ 3 mm (fig.3) is the only ECG variable that significantly indicates the risk of seizures and arrhythmias in acute tricyclic antidepressant poisoning (Liebelt et al., 1995). In addition, QT interval prolongation can occur with tricyclic antidepressant poisoning, as well as rightward axis deviation of the terminal 40 msec of the frontal plane QRS axis, which is unknown in other Na+ channel blocking agents (Wolfe et al. 1989; Berkovitch et al, 1995). Continued prolongation of the QRS complex may result in a sine wave pattern and eventual asystole.

Fig. 3. Acute poisoning with Amitriptyline. ECG reveals sinus tachycardia 148/min, RBBB pattern, QRS complex ≥ 120 ms, as well as R wave elevation in aVR ≥ 3 mm.

Na+ channel blockers may determine slowed intraventricular conduction, unidirectional block, the development of a reentrant circuit, and a resulting VT as well as VF. Because many of the Na+ channel blocking agents have also anticholinergic or sympathomimetic effects, bradydisrrhythmias are rare. In Na+ channel blocker poisoning by anticholinergic and sympathomimetic drugs, the combination of a wide QRS complex and bradycardia is a

2005). Direct toxin-induced blockade of cardiac Na+ channels will cause QRS complex widening, and it has been described as a membrane stabilizing effect, a local anesthetic effect, or a quinidine-like effect. Some drugs in this category (Table 1) may also affect other myocardial ion transfers, such as the Ca2+ influx and K+ efflux (Holstege et al., 2006). Other abnormal QRS complex configurations are also possible. In the most severe cases, the QRS complex widening becomes so profound that the ultimate origin of the rhythm disturbance

Fig. 2. Na+ channel blocker toxicity (patient with acute Propafenone overdose). Note the wide QRS complex, at a rate of 134/min, which suggests, at a first view, monomorphic VT. R wave elevation in aVR ≥ 3 mm (fig.3) is the only ECG variable that significantly indicates the risk of seizures and arrhythmias in acute tricyclic antidepressant poisoning (Liebelt et al., 1995). In addition, QT interval prolongation can occur with tricyclic antidepressant poisoning, as well as rightward axis deviation of the terminal 40 msec of the frontal plane QRS axis, which is unknown in other Na+ channel blocking agents (Wolfe et al. 1989; Berkovitch et al, 1995). Continued prolongation of the QRS complex may result in a sine

Fig. 3. Acute poisoning with Amitriptyline. ECG reveals sinus tachycardia 148/min, RBBB

Na+ channel blockers may determine slowed intraventricular conduction, unidirectional block, the development of a reentrant circuit, and a resulting VT as well as VF. Because many of the Na+ channel blocking agents have also anticholinergic or sympathomimetic effects, bradydisrrhythmias are rare. In Na+ channel blocker poisoning by anticholinergic and sympathomimetic drugs, the combination of a wide QRS complex and bradycardia is a

pattern, QRS complex ≥ 120 ms, as well as R wave elevation in aVR ≥ 3 mm.

is impossible (fig. 2).

wave pattern and eventual asystole.

sign of severe poisoning, indicating that the Na+ channel blockade is so profound that tachycardia does not occur, despite the clinical muscarinic antagonism or adrenergic agonism (Holstege et al., 2006). Nevertheless, bradycardia may occur because of slowed depolarization of pacemaker cells that depend on entry of Na+ ions.

### **2.2 Slow Calcium Channel Blockers (CCB)**

All CCBs (Table 2) inhibit the voltage sensitive L-type Ca2+ channel within the cell membrane. In the pacemaker cells of the sinoatrial node and AV node, the primary ion channel, which controls depolarization, is the slow Ca2+ channel. When inhibited, there is a slowing or an inhibition of the specialized tissue to conduct a cardiac impulse (Patel & Benowitz, 2005).


Table 2. Calcium channel blockers and the resulting ECG changes.

In CCB toxicity initially occurs a sinus bradycardia, followed by various degrees of AV block (fig.4), and junctional and ventricular bradydysrhythmias on ECG. Depending on the agent involved, other dysrhythmias may be seen (Gordon, 2006): sinus tachycardia (specifically Nifedipine), atrial arrhythmias, and junctional rhythms (fig. 5, 6,7). A wide QRS complex may appear, caused by ventricular escape rhythms or by CCB-induced Na+ channel blockade which delays of phase 0 of depolarization. Sudden shifts from bradydysrhythmias to cardiac arrest have been reported.

In addition, ECG changes associated with cardiac ischemia (fig.8) may occur as a result of the hypotension and changes in the cardiovascular status, especially in patients with preexisting cardiac disease (Patel & Benowitz, 2005; Holstege et al., 2006).

Toxic and Drug-Induced Changes of the Electrocardiogram 277

Fig. 8. Same patient (Diltiazem poisoning), second day of evolution. ECG reveals atrioventricular dissociation, with escape junctional rhythm and some ventricular captures, and

Fig. 9. Drug-induced long QT interval, following interaction between Bisoprolol and Amiodarone. ECG shows long QT interval (0.80 sec), couples of PVB with R/T

phenomenon preceding a short episode of polymorphic VT.

Medications in the K+ efflux blocker category block the outward flow of K+ from intracellular to extracellular spaces. Blockade of the outward K+ currents may prolong the cardiac cycle action potential (Fig. 1). The primary electrocardiographic manifestation is QT interval prolongation (QTc interval greater than 0.45 seconds in men and 0.47 seconds in women). Delay of repolarization causes the myocardial cell to have less charge difference across its membrane, and result in the activation of the inward depolarization current (early after-depolarization), which is seen on ECG as prominent U waves (Murphy et al., 2007). This may promote triggered activity, which potentially can progress to re-entry and subsequent polymorphic VT. Toxin-induced blockade of K+ efflux channels during phase 3 of the action potential corresponding with repolarization and QT interval prolongation may place the patient at risk for polymorphic VT or torsades de pointes (Holstege et al., 2005; Holstege et al., 2006). The drugs and toxins reported to have this effect are listed in table 3. Risk factors for torsades de pointes (TdP) among patients treated with medications that prolong QT interval include, among others, female gender, hypokalemia, hypomagnesemia, bradycardia, overdose, drug interactions (fig.9), digitalis therapy, and background of the patient (preexisting cardiac disease, long QT interval or family history of long QT). (Murphy

signs of ischemia (in inferior and lateral leads).

**2.3 Outward potassium channel blockers** 

et al., 2007; Roden, 2004)

Fig. 4. Acute poisoning with Verapamil in a 61-years old female. ECG reveals sinus bradycardia 41/min, minor right bundle ranch block (RBBB), first-degree AV block (PR 0.32 sec) and long QT interval (0.64 sec).

Fig. 5. Acute poisoning with Norvasc 300 mg in a 31-years old female. ECG reveals sinus tachycardia 114/min, signs of ischemia in infero-lateral leads, 14 hours after ingestion (systolic blood pressure 70 mmHg).

Fig. 6. Same patient, two days after ingestion. ECG reveals accelerated junctional rhythm (sinus coronary rhythm, or Zahn rhythm) 114/min, with negative P waves in leads DII, DIII, aVF, and disappearance of ischemic changes present at admission.

Fig. 7. Acute Diltiazem poisoning at admission. ECG reveals accelerated junctional rhythm 75/min, retrograde atrial conduction (negative P waves after QRS), and atrial premature beats.

Fig. 4. Acute poisoning with Verapamil in a 61-years old female. ECG reveals sinus

sec) and long QT interval (0.64 sec).

(systolic blood pressure 70 mmHg).

bradycardia 41/min, minor right bundle ranch block (RBBB), first-degree AV block (PR 0.32

Fig. 5. Acute poisoning with Norvasc 300 mg in a 31-years old female. ECG reveals sinus tachycardia 114/min, signs of ischemia in infero-lateral leads, 14 hours after ingestion

Fig. 6. Same patient, two days after ingestion. ECG reveals accelerated junctional rhythm (sinus coronary rhythm, or Zahn rhythm) 114/min, with negative P waves in leads DII, DIII,

Fig. 7. Acute Diltiazem poisoning at admission. ECG reveals accelerated junctional rhythm 75/min, retrograde atrial conduction (negative P waves after QRS), and atrial premature beats.

aVF, and disappearance of ischemic changes present at admission.

Fig. 8. Same patient (Diltiazem poisoning), second day of evolution. ECG reveals atrioventricular dissociation, with escape junctional rhythm and some ventricular captures, and signs of ischemia (in inferior and lateral leads).

### **2.3 Outward potassium channel blockers**

Medications in the K+ efflux blocker category block the outward flow of K+ from intracellular to extracellular spaces. Blockade of the outward K+ currents may prolong the cardiac cycle action potential (Fig. 1). The primary electrocardiographic manifestation is QT interval prolongation (QTc interval greater than 0.45 seconds in men and 0.47 seconds in women). Delay of repolarization causes the myocardial cell to have less charge difference across its membrane, and result in the activation of the inward depolarization current (early after-depolarization), which is seen on ECG as prominent U waves (Murphy et al., 2007). This may promote triggered activity, which potentially can progress to re-entry and subsequent polymorphic VT. Toxin-induced blockade of K+ efflux channels during phase 3 of the action potential corresponding with repolarization and QT interval prolongation may place the patient at risk for polymorphic VT or torsades de pointes (Holstege et al., 2005; Holstege et al., 2006). The drugs and toxins reported to have this effect are listed in table 3. Risk factors for torsades de pointes (TdP) among patients treated with medications that prolong QT interval include, among others, female gender, hypokalemia, hypomagnesemia, bradycardia, overdose, drug interactions (fig.9), digitalis therapy, and background of the patient (preexisting cardiac disease, long QT interval or family history of long QT). (Murphy et al., 2007; Roden, 2004)

Fig. 9. Drug-induced long QT interval, following interaction between Bisoprolol and Amiodarone. ECG shows long QT interval (0.80 sec), couples of PVB with R/T phenomenon preceding a short episode of polymorphic VT.

Toxic and Drug-Induced Changes of the Electrocardiogram 279

Many of these drugs have other effects that can result in significant electrocardiographic changes, such as antipsychotics, that can cause muscarinic acetylcholine receptor and alphaadrenergic receptor blockade and cardiac cell K+, Na+, and Ca2+ channel blockade (Holstege et al., 2006). These effects lead to sinus tachycardia (secondary to anticholinergic effect) or

Cardiac glycosides (table 4) inhibit the Na+/K+ adenosine triphosphatase (Na+/K+ATPase) pump. As a result, there is an inhibition of the active transport of Na+ and K+ across cell membrane, intracellular Na+ increases and the Na+/Ca2+exchanger is secondary activated. The intracellular Ca2+ level increases, and augments myofibril activity in cardiac myocytes, which results in a positive inotropic effect, and increased automaticity. The cardiac glycosides also increase vagal tone that may lead to a direct atrioventricular (AV) nodal depression (Holstege et al., 2006). Digitalis derivatives in therapeutic doses are used to increase myocardial contractility or slow AV conduction. They modify ECG, changes known as ''digitalis effect'', expressed by abnormal inverted or flattened T waves coupled with ST segment depression (most pronounced in leads with tall R waves), QT interval shortening (as a result of decreased ventricular repolarization time), PR interval lengthening (increased vagal activity), and prominent U-waves. Sagging ST segments, inverted T waves, and normal or shortened QT intervals are sometimes identified by their similar appearance to "Salvador Dali's mustache" (Clancy**,** 2007). These ECG changes are seen with therapeutic

reflex tachycardia (secondary to alpha-adrenergic blockade).

Digoxin levels and do not represent toxicity (Chung, 1981).


2. Plants producing cardiac glycosides:

3. Animals producing cardiac glycosides: - Bufadienolide type: *Bufo marinus* toads ECG changes - Excitant activity: atrial and junctional premature beats, atrial

Table 4. Na+/K+ ATPase blocking agents and ECG changes in intoxication (adapted from



*Digitalis lanata* and *Digitalis purpurea* (foxglove) – digoxin, digitoxin, *Nerium oleander* (oleander) – oleandrin, *Convallaria majalis* (Lily of the valley), *Apocynum cannabinum* (Dogbane), *Asclepias species.*

tachycardia, atrial flutter (rare), AF (rare), accelerated junctional rhythms, PVB, bigeminy and multifocal, VT, bi-directional VT, VF. - Suppressant activity: sinus bradycardia, sinoatrial block, type I second degree AV block (Wenckebach), bundle branch blocks, complete AV block, type II second degree AV block (rare). - Combination of these: atrial tachycardia with AV block, sinus bradycardia with junctional tachycardia, Wenckebach with junctional premature beats, regularization of ventricular rhythm

1. Drugs:


with AF.

Gordon, 2006; Lapostolle & Borron, 2007).

Na+/K+ ATPase

blockers

**2.4 Sodium–potassium ATPase blockers** 


\*removed from the market; \*\* TdP reported.

Table 3. K+ efflux channel blockers, and the resulting ECG changes.

Fig. 10. Acute organophosphate poisoning, five days after ingestion. ECG shows long QT interval (0.60 sec), PVB with R/T phenomenon (arrows) preceding a short episode of TdP (TORS).



Doxepin, Imipramine, Nortriptyline, Maprotiline) - Other antidepressants (Citalopram, Venlafaxine)






\*\*)

1. Cardiovascular drugs:

Ibutilide, Sotalol, Vernakalant)


Terfenadine\*, Hydroxyzine)

Hydroxychloroquine, etc.)

4. Synthetic opioids: Levomethadyl 5. Opium alkaloids: Papaverine\*\*

Table 3. K+ efflux channel blockers, and the resulting ECG changes.


6. Toxins: Quinine, Organophosphates (fig.10)

Premature ventricular beats (PVB) followed by TdP

Fig. 10. Acute organophosphate poisoning, five days after ingestion. ECG shows long QT interval (0.60 sec), PVB with R/T phenomenon (arrows) preceding a short episode of TdP

Propafenone)

2. Psychiatric drugs:



T- or U-wave abnormalities

ECG changes QT interval prolongation

\*removed from the market; \*\* TdP reported.

(TORS).

Sinus tachycardia

Inhibitors of outward K+ channel

Many of these drugs have other effects that can result in significant electrocardiographic changes, such as antipsychotics, that can cause muscarinic acetylcholine receptor and alphaadrenergic receptor blockade and cardiac cell K+, Na+, and Ca2+ channel blockade (Holstege et al., 2006). These effects lead to sinus tachycardia (secondary to anticholinergic effect) or reflex tachycardia (secondary to alpha-adrenergic blockade).

### **2.4 Sodium–potassium ATPase blockers**

Cardiac glycosides (table 4) inhibit the Na+/K+ adenosine triphosphatase (Na+/K+ATPase) pump. As a result, there is an inhibition of the active transport of Na+ and K+ across cell membrane, intracellular Na+ increases and the Na+/Ca2+exchanger is secondary activated. The intracellular Ca2+ level increases, and augments myofibril activity in cardiac myocytes, which results in a positive inotropic effect, and increased automaticity. The cardiac glycosides also increase vagal tone that may lead to a direct atrioventricular (AV) nodal depression (Holstege et al., 2006). Digitalis derivatives in therapeutic doses are used to increase myocardial contractility or slow AV conduction. They modify ECG, changes known as ''digitalis effect'', expressed by abnormal inverted or flattened T waves coupled with ST segment depression (most pronounced in leads with tall R waves), QT interval shortening (as a result of decreased ventricular repolarization time), PR interval lengthening (increased vagal activity), and prominent U-waves. Sagging ST segments, inverted T waves, and normal or shortened QT intervals are sometimes identified by their similar appearance to "Salvador Dali's mustache" (Clancy**,** 2007). These ECG changes are seen with therapeutic Digoxin levels and do not represent toxicity (Chung, 1981).


Table 4. Na+/K+ ATPase blocking agents and ECG changes in intoxication (adapted from Gordon, 2006; Lapostolle & Borron, 2007).

Toxic and Drug-Induced Changes of the Electrocardiogram 281

In an acute poisoning, ECG changes, especially arrhythmias, can be explained by direct or indirect sympathomimetic effects, anticholinergic effects, and the effects of altered central nervous system (CNS) regulation of peripheral autonomic activity. Sympathetic fibers innervate most parts of the heart. Postganglionic fibers release norepinephrine, which interacts with the beta 1- adrenergic cardiac receptors, to increase permeability to Na+ and Ca2+, thus leading to increased excitability, conduction and contractility. The vagal postganglionic parasympathetic fibers locally release acetylcholine. Vagal stimulation of the muscarinic receptors primarily decreases excitability of the atria, and slows the conduction of impulse into the ventricles, to a complete blockade of transmission in the AV node, with



calcium channel blocking properties)

AV blocks (first-degree AV block is common)

Table 5. Beta-blockers and ECG changes in intoxication (adapted from Gordon, 2006;

Prolonged PR, and QTc intervals

Asystole (in severe poisoning) Mild tachycardia (Pindolol overdose)






Ventricular tachydysrhythmias (membrane stabilizing agents) - fig.14


Prolonged QRS complex (membrane stabilizing agents)

Multifocal ventricular extrasystoles, VT, VF (Sotalol)

**3. Drugs and toxins acting on autonomic nervous system** 

modest direct effects on contractility (Patel & Benowitz, 2005).


sympathomimetic activity)



1st generation


activity)


ECG changes Sinus or nodal bradycardia

Holstege et al., 2006; Brubacher, 2007)

Beta-adrenergic

blockers

Electrocardiographic abnormalities with cardiac glycoside toxicity are the result of increased automaticity (from increased intracellular Ca2+) accompanied by slowed conduction through the AV node. In 10% to 15% of cases, ectopic rhythms will be the first sign of intoxication. AV block or an increase in ventricular automaticity are the most common manifestations of Digoxin toxicity and have been shown to occur in 30%to 40%of verified cases of toxicity. The nonspecific dysrhythmias consist of premature ventricular contractions (especially bigeminal and multiform), first-, second-, and third degree AV block, sinus bradycardia (fig.11), sinus tachycardia, sino-atrial block or arrest, atrial fibrillation (AF) with slow ventricular response (fig.12), atrial tachycardia, junctional escape rhythm, AV dissociation, ventricular bigeminy and trigeminy, VT (fig.13), TdP, and VF. The more specific dysrhythmias are AF with slow, regular ventricular rate (AV dissociation), nonparoxysmal junctional tachycardia (rate 70-130), atrial tachycardia with block (atrial rate is usually 150-200), and bi-directional VT. Bidirectional VT is particularly characteristic of severe toxicity and is the result of alterations of intraventricular conduction, junctional tachycardia with aberrant intraventricular conduction, or, on rare occasions, alternating ventricular pacemakers. Typically, in the young, slow rhythms and conduction defects predominate over ventricular ectopy, which is more prominent in older Digoxin-toxic patients. (Litonjua et al., 2005)

Fig. 11. Sinus bradicardia 47/min, with ST/T changes, 20 hours after attempted suicide with 10 mg Digoxin, in an 18 years old man.

Fig. 12. AF 56/min, with a pause of 2.96 sec, 15 hours after attempted suicide with Digoxin, in a 68 years old woman.

Fig. 13. PVB and a short episode of non-sustained VT 135/min, in a 56 years old male with Digoxin overdose in attempted suicide.

Electrocardiographic abnormalities with cardiac glycoside toxicity are the result of increased automaticity (from increased intracellular Ca2+) accompanied by slowed conduction through the AV node. In 10% to 15% of cases, ectopic rhythms will be the first sign of intoxication. AV block or an increase in ventricular automaticity are the most common manifestations of Digoxin toxicity and have been shown to occur in 30%to 40%of verified cases of toxicity. The nonspecific dysrhythmias consist of premature ventricular contractions (especially bigeminal and multiform), first-, second-, and third degree AV block, sinus bradycardia (fig.11), sinus tachycardia, sino-atrial block or arrest, atrial fibrillation (AF) with slow ventricular response (fig.12), atrial tachycardia, junctional escape rhythm, AV dissociation, ventricular bigeminy and trigeminy, VT (fig.13), TdP, and VF. The more specific dysrhythmias are AF with slow, regular ventricular rate (AV dissociation), nonparoxysmal junctional tachycardia (rate 70-130), atrial tachycardia with block (atrial rate is usually 150-200), and bi-directional VT. Bidirectional VT is particularly characteristic of severe toxicity and is the result of alterations of intraventricular conduction, junctional tachycardia with aberrant intraventricular conduction, or, on rare occasions, alternating ventricular pacemakers. Typically, in the young, slow rhythms and conduction defects predominate over ventricular ectopy, which is more

Fig. 11. Sinus bradicardia 47/min, with ST/T changes, 20 hours after attempted suicide with

Fig. 12. AF 56/min, with a pause of 2.96 sec, 15 hours after attempted suicide with Digoxin,

Fig. 13. PVB and a short episode of non-sustained VT 135/min, in a 56 years old male with

prominent in older Digoxin-toxic patients. (Litonjua et al., 2005)

10 mg Digoxin, in an 18 years old man.

Digoxin overdose in attempted suicide.

in a 68 years old woman.

### **3. Drugs and toxins acting on autonomic nervous system**

In an acute poisoning, ECG changes, especially arrhythmias, can be explained by direct or indirect sympathomimetic effects, anticholinergic effects, and the effects of altered central nervous system (CNS) regulation of peripheral autonomic activity. Sympathetic fibers innervate most parts of the heart. Postganglionic fibers release norepinephrine, which interacts with the beta 1- adrenergic cardiac receptors, to increase permeability to Na+ and Ca2+, thus leading to increased excitability, conduction and contractility. The vagal postganglionic parasympathetic fibers locally release acetylcholine. Vagal stimulation of the muscarinic receptors primarily decreases excitability of the atria, and slows the conduction of impulse into the ventricles, to a complete blockade of transmission in the AV node, with modest direct effects on contractility (Patel & Benowitz, 2005).


Table 5. Beta-blockers and ECG changes in intoxication (adapted from Gordon, 2006; Holstege et al., 2006; Brubacher, 2007)

Toxic and Drug-Induced Changes of the Electrocardiogram 283

Fig. 16. Acute Sotalol poisoning presenting with bradyarrhythmias, with severe QT

Cardiac disturbances caused by sympathetic-inhibiting drugs are listed in table 6.

ECG changes Sinus, atrial, junctional and ventricular

and imidazoline receptor stimulation (Murphy et al., 2007; Wiley II, 2007).

Table 6. Sympathetic-inhibitors and the resulting ECG changes.

**3.2 Other sympathetic – inhibitors (other than BB)** 

Sympathetic-inhibiting agents Methyldopa

**3.3 Sympathomimetic toxicity** 

exposure to a sympathomimetic.

(leads V1-V3).

prolongation, and monomorphic PVB, R/T phenomenon, and a couple of polymorphic PVB

Reserpine, Guanethidine

bradyarrhythmias First degree AV block

These drugs are used for their antihypertensive action, explained by central and peripheral alpha 2-adrenergic agonist effects. In acute overdose they cause ECG changes, along with hypotension, and cardiac failure. Cardiac arrests have been described in adults with Clonidine poisoning. Over-the-counter topical decongestants commonly contain imidazoline derivatives (naphzoline, tetrahydrozoline, oxymetazoline, and xylometazoline), and can cause systemic toxicity after topical exposure, or ingestion, with sympatholytic effects, such as bradyarrhythmias and hypotension, related to central alpha 2-adrenergic

Sympathetic overactivity can be caused by a number of drugs and toxins (table 7), such as illicit drugs, and hydrocarbon solvents, but also by sedative drug withdrawal syndromes. The typical ECG changes are sinus and atrial tachycardia, and occasionally ventricular dysrrhythmias (in massive exposures). Sinus tachycardia may be the first manifestation of

Prazosin and other alpha-blockers

Ventricular tachyarrhythmias

Clonidine and other imidazoline derivatives

### **3.1 Beta-adrenergic blockers (BB)**

BBs competitively inhibit various β-adrenergic receptors, and are listed in table 5.

They cause in most cases sinus bradycardia. Serious arrhythmias result from purely anticholinergic compound poisoning, especially in patients with underlying ischemic heart disease (e.g. atrial tachycardia and PVB). In acute BB overdose, the most pronounced effects are bradycardia (from decreased sinoatrial node function), varying degrees of AV block, and hypotension. Beta-adrenergic antagonists competitively antagonize the effects of catecholamines at the beta-adrenergic receptor and blunt the chronotropic and inotropic response to catecholamines (Bird, 2007).

Inhibition of the conducting system most commonly causes first-degree AV block, but higher levels of toxicity can promote second- and third-degree AV block (fig.15), junctional rhythms, and intraventricular conduction delays (Anderson, 2008).

Three beta-blockers are known to prolong QTc intervals: Sotalol (fig.16), Propranolol, and Acebutolol. Sotalol blocks K+ channels, thereby prolonging the action potential and the repolarization duration. The prolongation of the QTc interval predisposes the patient to ventricular tachyarrhythmias and TdP, which have been described after both Sotalol overdose and therapeutic administration. Propranolol overdose has caused QTc prolongation and torsades on rare occasion (Delk et al., 2006).

Fig. 14. Acute poisoning 3 hours after ingestion of 2 grams of Propranolol, presenting with accelerated idioventricular rhythm 115/min, in a young female. Note the absence of P waves, wide QRS complex (0.12 msec).

Fig. 15. Acute Metoprolol poisoning presenting with third degree AV block, with narrow QRS complex at a rate of 35/min. Atrial activity is represented by P waves, 80/min.

Beta-adrenergic antagonists also cause myocardial depression, at least in part, by an action independent of either catecholamine antagonism or membrane-depressant activity (Brubacher, 2007).

They cause in most cases sinus bradycardia. Serious arrhythmias result from purely anticholinergic compound poisoning, especially in patients with underlying ischemic heart disease (e.g. atrial tachycardia and PVB). In acute BB overdose, the most pronounced effects are bradycardia (from decreased sinoatrial node function), varying degrees of AV block, and hypotension. Beta-adrenergic antagonists competitively antagonize the effects of catecholamines at the beta-adrenergic receptor and blunt the chronotropic and inotropic

Inhibition of the conducting system most commonly causes first-degree AV block, but higher levels of toxicity can promote second- and third-degree AV block (fig.15), junctional

Three beta-blockers are known to prolong QTc intervals: Sotalol (fig.16), Propranolol, and Acebutolol. Sotalol blocks K+ channels, thereby prolonging the action potential and the repolarization duration. The prolongation of the QTc interval predisposes the patient to ventricular tachyarrhythmias and TdP, which have been described after both Sotalol overdose and therapeutic administration. Propranolol overdose has caused QTc

Fig. 14. Acute poisoning 3 hours after ingestion of 2 grams of Propranolol, presenting with accelerated idioventricular rhythm 115/min, in a young female. Note the absence of P

Fig. 15. Acute Metoprolol poisoning presenting with third degree AV block, with narrow QRS complex at a rate of 35/min. Atrial activity is represented by P waves, 80/min.

Beta-adrenergic antagonists also cause myocardial depression, at least in part, by an action independent of either catecholamine antagonism or membrane-depressant activity

BBs competitively inhibit various β-adrenergic receptors, and are listed in table 5.

rhythms, and intraventricular conduction delays (Anderson, 2008).

prolongation and torsades on rare occasion (Delk et al., 2006).

**3.1 Beta-adrenergic blockers (BB)** 

response to catecholamines (Bird, 2007).

waves, wide QRS complex (0.12 msec).

(Brubacher, 2007).

Fig. 16. Acute Sotalol poisoning presenting with bradyarrhythmias, with severe QT prolongation, and monomorphic PVB, R/T phenomenon, and a couple of polymorphic PVB (leads V1-V3).

### **3.2 Other sympathetic – inhibitors (other than BB)**

Cardiac disturbances caused by sympathetic-inhibiting drugs are listed in table 6.


Table 6. Sympathetic-inhibitors and the resulting ECG changes.

These drugs are used for their antihypertensive action, explained by central and peripheral alpha 2-adrenergic agonist effects. In acute overdose they cause ECG changes, along with hypotension, and cardiac failure. Cardiac arrests have been described in adults with Clonidine poisoning. Over-the-counter topical decongestants commonly contain imidazoline derivatives (naphzoline, tetrahydrozoline, oxymetazoline, and xylometazoline), and can cause systemic toxicity after topical exposure, or ingestion, with sympatholytic effects, such as bradyarrhythmias and hypotension, related to central alpha 2-adrenergic and imidazoline receptor stimulation (Murphy et al., 2007; Wiley II, 2007).

### **3.3 Sympathomimetic toxicity**

Sympathetic overactivity can be caused by a number of drugs and toxins (table 7), such as illicit drugs, and hydrocarbon solvents, but also by sedative drug withdrawal syndromes. The typical ECG changes are sinus and atrial tachycardia, and occasionally ventricular dysrrhythmias (in massive exposures). Sinus tachycardia may be the first manifestation of exposure to a sympathomimetic.

Toxic and Drug-Induced Changes of the Electrocardiogram 285

Fig. 18. Atrial tachycardia 150/min, with variable AV block, in a female with a chronic

Either as a result of excessive circulating catecholamines observed with cocaine and sympathomimetics, or myocardial sensitization secondary to halogenated hydrocarbons or thyroid hormone, or increased second-messenger activity secondary to theophylline, the extreme inotropic and chronotropic effects cause dysrhythmias. Altered repolarization, increased intracellular Ca2+ concentrations, or myocardial ischemia may cause the dysrhythmia. Additionally, cocaine that produce focal myocardial ischemia, can lead to malignant ventricular dysrhythmias (Clancy, 2007). In high dose, along with its potent sympathomimetic action, cocaine blocks fast Na+ channels in the myocardium, with a depression of depolarization, and slowing of conduction velocity, manifested on ECG with

There are numerous and various anticholinergic drugs and toxins that may be ingested

Mandrake (Mandragora officinarum) - Toxic mushrooms: Amanita muscaria

Table 8. Drugs and toxins with anticholinergic effect, with induced ECG abnormalities.

They cause in most cases sinus tachycardia. Serious arrhythmias result from purely anticholinergic compound poisoning, especially in patients with underlying ischemic heart disease (e.g. atrial tachycardia and ventricular premature beats). Atropine, for example, increases myocardial oxygen demand secondary to tachycardia, and can lead to VT and fibrillation in patients after myocardial infarction. Patients presenting with anticholinergic



respiratory disease and Theophylline overdosage.

prolonged PR, QRS, and QT intervals (Murphy et al., 2007).

1. Drugs:

2. Toxins:

ECG changes Sinus and atrial tachycardia


Premature ventricular beats

**3.4 Anticholinergic toxicity** 

Anticholinergic drugs and toxins

(table 8) and produce ECG abnormalities.


Table 7. Sympathomimetic drugs and toxins and the resulting ECG changes.

Fig. 17. Paroxysmal supraventricular tachycardia 169/min, in a young female with acute cannabis and ethanol poisoning.

However, other supraventricular or ventricular dysrhythmias may develop if an abnormal rhythm is generated in another part of the heart.






Sinoatrial slowing with escape junctional or ventricular rhythms

Myocardial ischemia or infarction (cocaine, amphetamines, or

sympathomimetics (decongestants containing phenylephrine,


pseudoephedrine, ephedrine)

Sympathomimetic drugs and toxins

1. Drugs:


Terbutaline)




Fig. 17. Paroxysmal supraventricular tachycardia 169/min, in a young female with acute

However, other supraventricular or ventricular dysrhythmias may develop if an abnormal



3. Toxins: - Ethanol

VT, VF

(solvent inhalation) Atrial tachycardia – fig. 18 Ventricular premature beats

hydrocarbons ingestion)

Table 7. Sympathomimetic drugs and toxins and the resulting ECG changes.

ECG abnormalities Sinus tachycardia

cannabis and ethanol poisoning.

rhythm is generated in another part of the heart.

2. Illicit drugs: - Amphetamines

Fig. 18. Atrial tachycardia 150/min, with variable AV block, in a female with a chronic respiratory disease and Theophylline overdosage.

Either as a result of excessive circulating catecholamines observed with cocaine and sympathomimetics, or myocardial sensitization secondary to halogenated hydrocarbons or thyroid hormone, or increased second-messenger activity secondary to theophylline, the extreme inotropic and chronotropic effects cause dysrhythmias. Altered repolarization, increased intracellular Ca2+ concentrations, or myocardial ischemia may cause the dysrhythmia. Additionally, cocaine that produce focal myocardial ischemia, can lead to malignant ventricular dysrhythmias (Clancy, 2007). In high dose, along with its potent sympathomimetic action, cocaine blocks fast Na+ channels in the myocardium, with a depression of depolarization, and slowing of conduction velocity, manifested on ECG with prolonged PR, QRS, and QT intervals (Murphy et al., 2007).

### **3.4 Anticholinergic toxicity**

There are numerous and various anticholinergic drugs and toxins that may be ingested (table 8) and produce ECG abnormalities.


Table 8. Drugs and toxins with anticholinergic effect, with induced ECG abnormalities.

They cause in most cases sinus tachycardia. Serious arrhythmias result from purely anticholinergic compound poisoning, especially in patients with underlying ischemic heart disease (e.g. atrial tachycardia and ventricular premature beats). Atropine, for example, increases myocardial oxygen demand secondary to tachycardia, and can lead to VT and fibrillation in patients after myocardial infarction. Patients presenting with anticholinergic

Toxic and Drug-Induced Changes of the Electrocardiogram 287

rare feature is acute myocardial infarction (fig.20), with a complex mechanism (fig.21) explaining its presence (coronary spasm induced by parasympathetic hyperactivity, direct

Fig. 19. Acute organophosphate poisoning, five days after exposure. ECG shows QT interval prolongation (0.60 sec), short episode of ventricular flutter (FLV), a couple of PVB, and one

Patients with clinical signs of cholinesterase inhibition and abnormal ECG (including long QT interval) should be monitored continuously, because of the risk of developing ventricular arrhythmias (Lionte et al., 2007). Reversible acetylcholinesterase inhibitors, such as Donepezil, have a high selectivity for neuronal acetylcholinesterase, and in accidental

Fig. 20. Acute organophosphate poisoning, four days after exposure, serum cholinesterase normalized with antidote. ECG shows acute ST segment elevation anterior myocardial infarction (with increased cardiac enzymes), sinus tachycardia 106/min, QT interval 0.32

overdose were reported to produce sinus bradycardia (Murphy et al., 2007).

toxic effect of pesticide on myocardium).

PVB with R on T phenomenon (R/T).

sec.

toxidrome (mydriasis, diminished bowel sounds, urinary retention, dry mouth, flushed skin, tachycardia, and agitation) may have ingested antihistamine-sympathomimetic combinations, or tricyclic antidepressants and neuroleptics, having both anticholinergic and membrane-depressant effects, which can explain the presence of serious cardiovascular disturbances (such as widening of the QRS complex, and a rightward deflection of the terminal 40 msec of the QRS complex, with prolongation of the QTc, which creates a substrate for the development of TdP). (Murphy et al., 2007; Juurlink, 2007)

### **3.5 Cholinomimetic toxicity**

Poisoning by cholinomimetic drugs and toxins (table 9) lead to different ECG aspects. The most common type of cholinomimetic toxicity is poisoning with organophosphate and carbamate pesticides, resulting in the excessive inhibition of the cholinesterase. The ECG changes are unpredictable and often change over the time course of the poisoning.


Table 9. Cholinomimetic drugs and toxins, and ECG changes induced in acute exposure.

Early in the course, tachycardia is present, due to acetylcholine stimulation of nicotinic receptors, followed by bradycardia, secondary to muscarinic receptor stimulation. In severe poisonings, advanced AV block, bradydysrhythmias and asystole may occur (Murphy et al., 2007). Up to 5 days after exposure, QTc interval prolongation is followed by ventricular tachyarrhythmias (fig.19), including TdP (fig.10), due to persistent imbalance between sympathetic and parasympathetic influences on the heart, as well as dyselectrolytemias. A

toxidrome (mydriasis, diminished bowel sounds, urinary retention, dry mouth, flushed skin, tachycardia, and agitation) may have ingested antihistamine-sympathomimetic combinations, or tricyclic antidepressants and neuroleptics, having both anticholinergic and membrane-depressant effects, which can explain the presence of serious cardiovascular disturbances (such as widening of the QRS complex, and a rightward deflection of the terminal 40 msec of the QRS complex, with prolongation of the QTc, which creates a

Poisoning by cholinomimetic drugs and toxins (table 9) lead to different ECG aspects. The most common type of cholinomimetic toxicity is poisoning with organophosphate and carbamate pesticides, resulting in the excessive inhibition of the cholinesterase. The ECG






Sinus tachycardia (seen in early stages of cholinesterase inhibition and

changes are unpredictable and often change over the time course of the poisoning.

volatile anesthetics, Ketamine)


Aminopyridines, Carbachol, Guanidine)


Atrial, junctional, or ventricular bradycardia

nicotine poisoning due to ganglionic stimulation) VT associated with QT interval prolongation

Table 9. Cholinomimetic drugs and toxins, and ECG changes induced in acute exposure.

Early in the course, tachycardia is present, due to acetylcholine stimulation of nicotinic receptors, followed by bradycardia, secondary to muscarinic receptor stimulation. In severe poisonings, advanced AV block, bradydysrhythmias and asystole may occur (Murphy et al., 2007). Up to 5 days after exposure, QTc interval prolongation is followed by ventricular tachyarrhythmias (fig.19), including TdP (fig.10), due to persistent imbalance between sympathetic and parasympathetic influences on the heart, as well as dyselectrolytemias. A


substrate for the development of TdP). (Murphy et al., 2007; Juurlink, 2007)

**3.5 Cholinomimetic toxicity** 

1. Drugs:

Donepezil)

2. Toxins:


plant)

AV block

Asystole

Sinus bradycardia

Cholinomimetic drugs and toxins

ECG

abnormalities

rare feature is acute myocardial infarction (fig.20), with a complex mechanism (fig.21) explaining its presence (coronary spasm induced by parasympathetic hyperactivity, direct toxic effect of pesticide on myocardium).

Fig. 19. Acute organophosphate poisoning, five days after exposure. ECG shows QT interval prolongation (0.60 sec), short episode of ventricular flutter (FLV), a couple of PVB, and one PVB with R on T phenomenon (R/T).

Patients with clinical signs of cholinesterase inhibition and abnormal ECG (including long QT interval) should be monitored continuously, because of the risk of developing ventricular arrhythmias (Lionte et al., 2007). Reversible acetylcholinesterase inhibitors, such as Donepezil, have a high selectivity for neuronal acetylcholinesterase, and in accidental overdose were reported to produce sinus bradycardia (Murphy et al., 2007).

Fig. 20. Acute organophosphate poisoning, four days after exposure, serum cholinesterase normalized with antidote. ECG shows acute ST segment elevation anterior myocardial infarction (with increased cardiac enzymes), sinus tachycardia 106/min, QT interval 0.32 sec.

Toxic and Drug-Induced Changes of the Electrocardiogram 289

Myocardial injury from CO poisoning results from tissue hypoxia, and as damage at the cellular level. The affinity of hemoglobin for CO is 200 to 250 times greater than its affinity for oxygen. This results in competitive inhibition of oxygen release due to a shift in the oxygen-hemoglobin dissociation curve, reduced oxygen delivery, and subsequent tissue hypoxia. In CO poisoning, magnitude of ST/T changes doesn't correlate with the severity of the myocardial impairment, other tests, such as echocardiography, being necessary. All cardiovascular changes are more prominent in patients with underlying cardiac pathology

Fig. 22. ECG in acute CO poisoning, 24 hours after exposure (flattened T wave in DI, negative T wave in aVL, ST segment elevation and T wave changes in V1-V3).

Fig. 23. ECG in acute CO poisoning, same patient, 52 hours after exposure. Note the

Fig. 24. Same patient, 4 days after exposure. Note the evolution of ST/T changes in

evolution of ST/T changes in precordial leads.

precordial and lateral leads.

(Murphy et al., 2007; Satran et al., 2005).

Fig. 21. Cardiac manifestations of organophosphate poisoning and mechanisms involved.

### **4. Other substances inducing ECG changes**

### **4.1 Chemical asphyxiants**

Chemical asphyxiants act in one of two ways. Some prevent the uptake of oxygen in the blood. Carbon monoxide interferes with the transport of oxygen to the tissues by strongly binding with hemoglobin to form carboxyhemoglobin, which leaves inadequate hemoglobin available for oxygen transport. Hydrogen cyanide does not permit the normal oxygen transfer either from the blood to the tissues or within the cell itself, resulting in tissue hypoxia. Acute exposure to these agents leads to coma and metabolic acidosis, the last explaining in part ECG abnormalities (table 10) recorded in such poisoning.


Table 10. Major chemical asphyxiants and their effect on ECG in acute poisoning (adapted from Hall, 2007; Murphy et al., 2007; **Schraga** et al., 2008)

Fig. 21. Cardiac manifestations of organophosphate poisoning and mechanisms involved.

Chemical asphyxiants act in one of two ways. Some prevent the uptake of oxygen in the blood. Carbon monoxide interferes with the transport of oxygen to the tissues by strongly binding with hemoglobin to form carboxyhemoglobin, which leaves inadequate hemoglobin available for oxygen transport. Hydrogen cyanide does not permit the normal oxygen transfer either from the blood to the tissues or within the cell itself, resulting in tissue hypoxia. Acute exposure to these agents leads to coma and metabolic acidosis, the last

ST segment depression or elevation

Myocardial infarction (fig.22, 23, 24)

Table 10. Major chemical asphyxiants and their effect on ECG in acute poisoning (adapted

Arrhythmias occasionally (PVB, atrial fibrillation)

Erratic supraventricular and vntricular arrhythmias

Shortening of the ST segment with eventual fusion of the T wave

explaining in part ECG abnormalities (table 10) recorded in such poisoning.

Conduction disturbance

into the QRS complex.

Ventricular arrhythmias ST/T changes (fig.25)

Ischemic changes

Asystole

from Hall, 2007; Murphy et al., 2007; **Schraga** et al., 2008)

Bradicardia, heart block (late)

**4. Other substances inducing ECG changes** 

**4.1 Chemical asphyxiants** 

Chemical asphyxiants ECG changes

Hydrogen cyanide Tachycardia (early)

Zinc phosphide Atrial extrasystoles

Carbon monoxide (CO) T wave flattening or inversion

Myocardial injury from CO poisoning results from tissue hypoxia, and as damage at the cellular level. The affinity of hemoglobin for CO is 200 to 250 times greater than its affinity for oxygen. This results in competitive inhibition of oxygen release due to a shift in the oxygen-hemoglobin dissociation curve, reduced oxygen delivery, and subsequent tissue hypoxia. In CO poisoning, magnitude of ST/T changes doesn't correlate with the severity of the myocardial impairment, other tests, such as echocardiography, being necessary. All cardiovascular changes are more prominent in patients with underlying cardiac pathology (Murphy et al., 2007; Satran et al., 2005).

Fig. 22. ECG in acute CO poisoning, 24 hours after exposure (flattened T wave in DI, negative T wave in aVL, ST segment elevation and T wave changes in V1-V3).

Fig. 23. ECG in acute CO poisoning, same patient, 52 hours after exposure. Note the evolution of ST/T changes in precordial leads.

Fig. 24. Same patient, 4 days after exposure. Note the evolution of ST/T changes in precordial and lateral leads.

Toxic and Drug-Induced Changes of the Electrocardiogram 291

Atrial arrhythmias

Scombroid fish (scombrotoxin) Bradycardia, sinus tachycardia, ventricular

VF

acid, while others produce the poisoning because of improper fish handling.

Bradycardia

Table 11. ECG changes induced by natural products poisoning (adapted from Hahn, 2007;

In the pathogenesis of clinical syndrome produced by black widow spider bites was suggested to be also involved an excess of catecholamines, while in patients affected by

Fish-borne poisoning has multiple pathogenic mechanisms, depending on the toxin involved. Some of these toxins are heat stable, therefore unaffected by cooking, and gastric

Fig. 26. Scombroid fish poisoning, 1 hour after ingestion, in a young female, presenting chest pain and hypotension (BP 70/45 mmHg). ECG reveals signs of subendocardial myocardial

Aconite poisoning results from alkaloids contained in teas and herbs, which are not boiled enough before ingestion, such as aconitine and mesaconitine (Murphy et al., 2007).

infarction (sustained by enzymatic evidence), and sinus tachycardia 139/min.

changes with peaked T waves, QT prolongation, atrial and ventricular arrhythmias, myocardial infarction

Tachycardia, dysrhythmias, myocardial infarction

arrhythmias, acute myocardial infarction (fig.26)

VT (monomorphic or polymorphic)

Poisonous scorpion stings (venom) Sinus tachycardia, conduction abnormalities, ST/T

Poisonous animals and plants ECG abnormalities

Ciguatera fish (ciguatoxin) Bradycardia, arrhythmias Puffer fish (tetrodotoxin) Bradycardia, arrhythmias

Black widow spider bites (venom) Tachycardia

Lionte, 2010; Murphy et al., 2007; Wandersee, 2006).

*Hymenoptera* stings, histamine plays a role in pathogenesis.

*Hymenoptera* (bees and wasps)

Aconite poisoning (Chinese herbs or teas that contain *Aconitum carmichaelii* or *A. kusnezoffii*)

stings (venom)

In humans, cyanide produces histotoxic hypoxia by combining with the ferric ion in mitochondrial cytochrome oxidase, preventing electron transport in the cytochrome system and bringing oxidative phosphorylation and ATP production to a halt. The inhibition of oxidative metabolism puts increased demands on anaerobic glycolysis, which results in lactic acid production and may produce severe acid-base imbalance. Myocardial depression with decreased cardiac output produces stagnation hypoxia. Abnormal heartbeat can occur in cases of severe poisoning. Bradycardia, intractable low blood pressure, and death may result (Hall, 2007).

Fig. 25. Attempted suicide with zinc phosphide ingestion in a 19 years old female, presenting circulatory collapse and dyspnea. ECG (leads aVR, aVL, aVF, V1-V3) reveals tachycardia 160 /min with wide QRS complex, multiple PVC (trigeminy), some with R/T phenomenon, and ST/T changes.

Zinc phosphide is a highly effective insecticide and rodenticide. It is mediated by phosphine, which inhibits cytochrome C oxidase. It has been shown recently in nematodes that phosphine rapidly perturbs mitochondrial morphology, inhibits oxidative respiration by 70%, and causes a severe drop in mitochondrial membrane potential. This failure of cellular respiration is likely to be due to a mechanism other than inhibition of cytochrome C oxidase. In addition, phosphine and hydrogen peroxide can interact to form the highly reactive hydroxyl radical and phosphine also inhibits catalase and peroxidase; both mechanisms result in hydroxyl radical associated damage such as lipid peroxidation. The major lethal consequence of zinc phosphide ingestion, profound circulatory collapse, is secondary to factors including direct effects on cardiac myocytes, fluid loss, and adrenal gland damage. There is usually only a short interval between ingestion and the appearance of systemic toxicity. Phosphine-induced impairment of myocardial contractility and fluid loss leads to circulatory failure, and pulmonary edema. Metabolic acidosis, or mixed metabolic acidosis and respiratory alkalosis, are frequent, and contribute to ECG changes (Lionte et al., 2004; Proudfoot, 2009).

### **4.2 Natural products**

Many natural products and toxins have cardiovascular effects, leading to ECG changes (table 11). The clinical and myocardial histological manifestations of scorpion sting resemble those of catecholamine infusion. Myocardial infarction has been documented in scorpion envenomation with a pathophysiological mechanism involving transient myocardial "stunning" (Thomas et al., 2007).

In humans, cyanide produces histotoxic hypoxia by combining with the ferric ion in mitochondrial cytochrome oxidase, preventing electron transport in the cytochrome system and bringing oxidative phosphorylation and ATP production to a halt. The inhibition of oxidative metabolism puts increased demands on anaerobic glycolysis, which results in lactic acid production and may produce severe acid-base imbalance. Myocardial depression with decreased cardiac output produces stagnation hypoxia. Abnormal heartbeat can occur in cases of severe poisoning. Bradycardia, intractable low blood pressure, and death may

Fig. 25. Attempted suicide with zinc phosphide ingestion in a 19 years old female, presenting circulatory collapse and dyspnea. ECG (leads aVR, aVL, aVF, V1-V3) reveals tachycardia 160 /min with wide QRS complex, multiple PVC (trigeminy), some with R/T

Zinc phosphide is a highly effective insecticide and rodenticide. It is mediated by phosphine, which inhibits cytochrome C oxidase. It has been shown recently in nematodes that phosphine rapidly perturbs mitochondrial morphology, inhibits oxidative respiration by 70%, and causes a severe drop in mitochondrial membrane potential. This failure of cellular respiration is likely to be due to a mechanism other than inhibition of cytochrome C oxidase. In addition, phosphine and hydrogen peroxide can interact to form the highly reactive hydroxyl radical and phosphine also inhibits catalase and peroxidase; both mechanisms result in hydroxyl radical associated damage such as lipid peroxidation. The major lethal consequence of zinc phosphide ingestion, profound circulatory collapse, is secondary to factors including direct effects on cardiac myocytes, fluid loss, and adrenal gland damage. There is usually only a short interval between ingestion and the appearance of systemic toxicity. Phosphine-induced impairment of myocardial contractility and fluid loss leads to circulatory failure, and pulmonary edema. Metabolic acidosis, or mixed metabolic acidosis and respiratory alkalosis, are frequent, and contribute to ECG changes

Many natural products and toxins have cardiovascular effects, leading to ECG changes (table 11). The clinical and myocardial histological manifestations of scorpion sting resemble those of catecholamine infusion. Myocardial infarction has been documented in scorpion envenomation with a pathophysiological mechanism involving transient myocardial

result (Hall, 2007).

phenomenon, and ST/T changes.

(Lionte et al., 2004; Proudfoot, 2009).

"stunning" (Thomas et al., 2007).

**4.2 Natural products** 


Table 11. ECG changes induced by natural products poisoning (adapted from Hahn, 2007; Lionte, 2010; Murphy et al., 2007; Wandersee, 2006).

In the pathogenesis of clinical syndrome produced by black widow spider bites was suggested to be also involved an excess of catecholamines, while in patients affected by *Hymenoptera* stings, histamine plays a role in pathogenesis.

Fish-borne poisoning has multiple pathogenic mechanisms, depending on the toxin involved. Some of these toxins are heat stable, therefore unaffected by cooking, and gastric acid, while others produce the poisoning because of improper fish handling.

Fig. 26. Scombroid fish poisoning, 1 hour after ingestion, in a young female, presenting chest pain and hypotension (BP 70/45 mmHg). ECG reveals signs of subendocardial myocardial infarction (sustained by enzymatic evidence), and sinus tachycardia 139/min.

Aconite poisoning results from alkaloids contained in teas and herbs, which are not boiled enough before ingestion, such as aconitine and mesaconitine (Murphy et al., 2007).

Toxic and Drug-Induced Changes of the Electrocardiogram 293

Amphetamine and related drugs activate the sympathetic nervous system via central nervous system stimulation, peripheral release of catecholamines, inhibition of neuronal reuptake of catecholamines, and inhibition of monoamine oxidase. Fenfluramine and dexfenfluramine cause serotonin release and block neuronal serotonin uptake. Methamphetamine (crank, speed), 3,4-methylenedioxymethamphetamine (MDMA, ecstasy), and several other amphetamine derivatives (Lysergic Acid Diethylamide), as well as a number of prescription drugs, are used orally and intravenously as illicit stimulants. "Ice" is a smokable form of methamphetamine (Albertson, 2004). Direct catecholamine effects, and ischemic effects, after coronary vasospasm explain arrhythmias, in amphetamine use. The last is involved in the pathogenesis of myocardial infarction after amphetamine use, together with direct cardiac toxicity (myocarditis), and thrombus formation (Albertson et al.,

Most of the hallucinogens are indoleamine or phenylethylamine derivatives, structurally similar to the neurotransmitter serotonin. Commonly used hallucinogens include lysergic acid diethylamide (LSD), mescaline, 5-methoxy-N,N-diisopropyltriptamine, and psilocybin ("magic mushrooms"). They produce a variety of autonomic effects, both parasympathetic

Cocaine is one of the most popular drugs of abuse. Rapidly after smoking or intravenous injection (mediated by sympathetic overactivity) appear cardiovascular signs of toxicity. Coronary artery spasm and/or thrombosis may result in myocardial infarction, even in patients with no coronary disease. Chest pain with electrocardiographic evidence of ischemia or infarction in a young, otherwise healthy person suggests cocaine use (Benowitz, 2004). At low doses, occur sinus bradycardia and ectopic rhythms, based on cocaine's local anesthetic properties and its effects on catecholamines. At high doses, cocaine produces direct Na+ and K+ channel blockade (see section 3.3). Enhanced sympathetic stimulation will increase intracellular Ca2+ within myocardial cells, and will enhance automaticity, leading to

The cannabinoid delta 9-tetrahidrocannabinol (THC) is the principal psychoactive constituent of *Cannabis sativa* (marijuana consists of the leaves and flowering parts of the plant). Cardiovascular toxicity is dose-related, and is explained by stimulation of the autonomic nervous system, involving both parasympathetic and sympathetic pathways. The effects tend to be more serious in patients with preexisting cardiovascular pathology (for example, an increased risk of myocardial infarction was reported in the hour following marijuana use), but in young, healthy patients, these effects have no serious consequences

Opiates are a group of naturally occurring compounds derived from the juice of the poppy *Papaver somniferum*. The term opioid refers to these and other derivatives of naturally occurring opium (e.g., morphine, heroin, codeine, and hydrocodone) as well as new, totally synthetic opiate analogs (e.g., fentanyl, butorphanol, meperidine, methadone, and propoxyphene). In general, opioids share the ability to stimulate a number of specific opiate receptors in the CNS. With mild or moderate overdose**,** pulse rate is decreased. Cardiotoxicity similar to that seen with tricyclic antidepressants and quinidine can occur in patients with severe propoxyphene intoxication. Heroin and propoxyphene toxicity is associated with ECG changes, such as non-specific ST/T wave abnormalities, first-degree AV block, AF, prolonged QTc intervals, and ventricular dysrhythmias. In the pathogenesis of these cardiovascular findings contribute electrolyte, and metabolic derangements,

hypoxia, or adulterants (e.g. quinine) found in street drugs (Yip et al., 2007).

afterdepolarizations, and ectopic rhythms (Albertson et al., 2007, a).

2007, b).

(Delgado, 2007).

and sympathetic (Traub, 2007).

Aconitine has Na+ channel-binding properties (maintaining them in an open position), which explain, in part, its neurological and cardiovascular toxicity (cardiodepressant effects). Vagal stimulation is also involved in pathogenesis of aconitine poisoning. Aconitine has a propensity to cause early and delayed after-depolarizations in ventricular myocytes that may be due to increased intracellular Ca2+ and Na+. This explains the presence of biventricular tachycardia and TdP on ECG (Smolinske et al., 2007).

### **4.3 Drugs of abuse**

Some of the most commonly abused drugs are alcohol, nicotine, marijuana, amphetamines, cocaine, opium alkaloids and synthetic opioids, gamma-hydroxybutyrate, 3,4 methylenedioxymethamphetamine (MDMA, ecstasy), and phencyclidine. Drug abuse may lead to organ damage, addiction, and disturbed patterns of behavior. Some illicit drugs, such as heroin, lysergic acid diethylamide, and phencyclidine hydrochloride, have no recognized therapeutic effect in humans. Cardiovascular toxicity of illicit drugs relies on multiple pathophysiological mechanisms. Table 12 presents major categories of drugs, and the ECG abnormalities reported in acute setting.


Table 12. Drugs of abuse and consequent ECG changes in acute poisoning (adapted from Albertson, 2004; Albertson et al., 2007, a; Delgado, 2007; Quang, 2007; Traub, 2007; Yip et al., 2007).

Aconitine has Na+ channel-binding properties (maintaining them in an open position), which explain, in part, its neurological and cardiovascular toxicity (cardiodepressant effects). Vagal stimulation is also involved in pathogenesis of aconitine poisoning. Aconitine has a propensity to cause early and delayed after-depolarizations in ventricular myocytes that may be due to increased intracellular Ca2+ and Na+. This explains the presence of

Some of the most commonly abused drugs are alcohol, nicotine, marijuana, amphetamines, cocaine, opium alkaloids and synthetic opioids, gamma-hydroxybutyrate, 3,4 methylenedioxymethamphetamine (MDMA, ecstasy), and phencyclidine. Drug abuse may lead to organ damage, addiction, and disturbed patterns of behavior. Some illicit drugs, such as heroin, lysergic acid diethylamide, and phencyclidine hydrochloride, have no recognized therapeutic effect in humans. Cardiovascular toxicity of illicit drugs relies on multiple pathophysiological mechanisms. Table 12 presents major categories of drugs, and

> Myocardial ischemia, and infarction PVB and ventricular tachydysrhythmias

Supraventricular tachyarrhythmias

block, ST-segment elevation (pediatric)

First degree AV block, transient AF (rare)

Prolonged QT interval and TdP (methadone) QRS prolongation, dysrhythmias (Propoxyphene)

Non-specific ST/T changes PVC (occasional) and AF Cocaine Sinus bradycardia, complete heart block, bundle branch

Prolonged QTc, and QRS

VT or fibrillation Myocardial infarction

U waves (in adults)

Tachycardia (Tramadol) Table 12. Drugs of abuse and consequent ECG changes in acute poisoning (adapted from Albertson, 2004; Albertson et al., 2007, a; Delgado, 2007; Quang, 2007; Traub, 2007; Yip et al.,

Bradycardia

(heroin)

Sinus tachycardia and supraventricular tachyarrhythmias

Transient P wave inversion in DII, right bundle branch

Cardiac conduction abnormalities and dysrhythmias

biventricular tachycardia and TdP on ECG (Smolinske et al., 2007).

the ECG abnormalities reported in acute setting.

Cannabis (marijuana) Tachycardia (fig. 17)

block

Illicit drug ECG

**4.3 Drugs of abuse** 

Amphetamines and

(including "ecstasy")

Gamma-hydroxybutyrate

Hallucinogens Tachycardia

Opiates and opioids Bradycardia (opioids)

derivatives

(GHB)

2007).

Amphetamine and related drugs activate the sympathetic nervous system via central nervous system stimulation, peripheral release of catecholamines, inhibition of neuronal reuptake of catecholamines, and inhibition of monoamine oxidase. Fenfluramine and dexfenfluramine cause serotonin release and block neuronal serotonin uptake. Methamphetamine (crank, speed), 3,4-methylenedioxymethamphetamine (MDMA, ecstasy), and several other amphetamine derivatives (Lysergic Acid Diethylamide), as well as a number of prescription drugs, are used orally and intravenously as illicit stimulants. "Ice" is a smokable form of methamphetamine (Albertson, 2004). Direct catecholamine effects, and ischemic effects, after coronary vasospasm explain arrhythmias, in amphetamine use. The last is involved in the pathogenesis of myocardial infarction after amphetamine use, together with direct cardiac toxicity (myocarditis), and thrombus formation (Albertson et al., 2007, b).

Most of the hallucinogens are indoleamine or phenylethylamine derivatives, structurally similar to the neurotransmitter serotonin. Commonly used hallucinogens include lysergic acid diethylamide (LSD), mescaline, 5-methoxy-N,N-diisopropyltriptamine, and psilocybin ("magic mushrooms"). They produce a variety of autonomic effects, both parasympathetic and sympathetic (Traub, 2007).

Cocaine is one of the most popular drugs of abuse. Rapidly after smoking or intravenous injection (mediated by sympathetic overactivity) appear cardiovascular signs of toxicity. Coronary artery spasm and/or thrombosis may result in myocardial infarction, even in patients with no coronary disease. Chest pain with electrocardiographic evidence of ischemia or infarction in a young, otherwise healthy person suggests cocaine use (Benowitz, 2004). At low doses, occur sinus bradycardia and ectopic rhythms, based on cocaine's local anesthetic properties and its effects on catecholamines. At high doses, cocaine produces direct Na+ and K+ channel blockade (see section 3.3). Enhanced sympathetic stimulation will increase intracellular Ca2+ within myocardial cells, and will enhance automaticity, leading to afterdepolarizations, and ectopic rhythms (Albertson et al., 2007, a).

The cannabinoid delta 9-tetrahidrocannabinol (THC) is the principal psychoactive constituent of *Cannabis sativa* (marijuana consists of the leaves and flowering parts of the plant). Cardiovascular toxicity is dose-related, and is explained by stimulation of the autonomic nervous system, involving both parasympathetic and sympathetic pathways. The effects tend to be more serious in patients with preexisting cardiovascular pathology (for example, an increased risk of myocardial infarction was reported in the hour following marijuana use), but in young, healthy patients, these effects have no serious consequences (Delgado, 2007).

Opiates are a group of naturally occurring compounds derived from the juice of the poppy *Papaver somniferum*. The term opioid refers to these and other derivatives of naturally occurring opium (e.g., morphine, heroin, codeine, and hydrocodone) as well as new, totally synthetic opiate analogs (e.g., fentanyl, butorphanol, meperidine, methadone, and propoxyphene). In general, opioids share the ability to stimulate a number of specific opiate receptors in the CNS. With mild or moderate overdose**,** pulse rate is decreased. Cardiotoxicity similar to that seen with tricyclic antidepressants and quinidine can occur in patients with severe propoxyphene intoxication. Heroin and propoxyphene toxicity is associated with ECG changes, such as non-specific ST/T wave abnormalities, first-degree AV block, AF, prolonged QTc intervals, and ventricular dysrhythmias. In the pathogenesis of these cardiovascular findings contribute electrolyte, and metabolic derangements, hypoxia, or adulterants (e.g. quinine) found in street drugs (Yip et al., 2007).

Toxic and Drug-Induced Changes of the Electrocardiogram 295

Gordon, B. (2006). Cardiovascular drugs, In: *The Toxicology Handbook for Clinicians*, Harris, C.R. (Ed.), pp. 61-75, Mosby Elsevier, ISBN 1-56053-711-6, Philadelphia, USA. Hahn, I-H. (2007). Arthropods, In: *Goldfrank's Manual of Toxicologic Emergencies*, Hoffman

Hall, A.H. (2007). Cyanide and related compounds – sodium azide, In: *Haddad and* 

Holstege, C.; Baer, A. & Brady, W.J. (2005). The electrocardiographic toxidrome: the ECG

Holstege, C.P.; Eldridge, D.L. & Rowden, A.K. (2006). ECG manifestations: the poisoned

Juurlink, D.N. (2007). Antipsychotics, In: *Goldfrank's Manual of Toxicologic Emergencies*, Hoffman R.S. et al. (Eds.), pp. 583-590, The McGraw-Hill Companies, Inc., http://www.mhprofessional.com/product.php?cat=116&isbn=007144310X.

Lapostolle, F. & Borron, S.W. (2007). Digitalis, In: *Haddad and Winchester's Clinical* 

pp. 949-962, Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA. Liebelt, E.L.; Francis, P.D & Woolf, A.D. (1995). ECG Lead aVR versus QRS interval in

Lionte, C. (2010). An unusual cause of hypotension and abnormal electrocardiogram (ECG)

Lionte, C., et al. (2007). Electrocardiographic changes in acute organophosphate poisoning.

Lionte, C.; Sorodoc, L. & Laba, V. (2004). Zinc phosphide poisoning - diagnosis,

Litonjua, R.M., et al. (2005). Digoxin: The Monarch of Cardiac Toxicities. *Journal of Pharmacy* 

Murphy, N.G; Benowitz, N.L. & Goldschlager, N. (2007). Cardiovascular toxicology, In:

Patel, M.M. & Benowitz, N. (2005). Cardiac conduction and rate disturbances, In: *Critical care* 

(Eds.), pp. 241-260, Elsevier Mosby, ISBN 0-8151-4387-7, Philadelphia, USA. Proudfoot, A.T. (2009). Aluminium and zinc phosphide poisoning. *Clin Toxicol (Phila).* Vol.

*Practice,* Vol. 18, No. 3, (June 2005), pp.157–168, ISSN 0897-1900.

*Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 1309-1316, Saunders Elsevier, ISBN 978-0-7216-0693-4,

presentation of hydrofluoric acid ingestion. *American Journal of Emergency Medicine*,

patient. *Emergency Medicine Clinics of North America,* Vol.24, No.1, (Feb 2006),

*Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.),

predicting seizures and arrhythmias in acute tricyclic antidepressant toxicity. *Annals of Emergency Medicine*, Vol. 26, No.2, (Aug 1995), pp. 195-201, ISSN 0196-

— scombroid poisoning. *Central European Journal of Medicine***,** Vol. 5, No. 3, (June

*Revista Medico-Chirugicala a Societatii de Medici si Naturalisti Iasi*, Vol.111, No.4, (Oct-

complications, management. *Terapeutica, Farmacologie si Toxicologie Clinica*, Vol.

*Haddad and Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 133-165, Saunders Elsevier, ISBN 978-0-7216-0693-

*toxicology: diagnosis and management of the critically poisoned patient*, Brent, J. et al.

R.S. et al. (Eds.), pp. 901-911, The McGraw-Hill Companies, Inc., http://www.mhprofessional.com/product.php?cat=116&isbn=007144310X.

Vol. 23, No.2, (March 2005), pp.171-176. ISSN 0735-6757.

Philadelphia, USA.

0644.

pp.159-177, ISSN 0733-8627.

2010), pp. 292-297, ISSN 1895-1058.

4, Philadelphia, USA.

Dec, 2007), pp. 906-911, ISSN 0048-7848.

VIII, No. 4, (Dec 2004), pp. 87-91, ISSN 1583-0012.

47, No.2, (Feb 2009), pp. 89-100, ISSN 1556-3650.

### **5. Conclusions**

ECG is a valuable source of information in poisoned patients and has the potential to enhance and direct their care. Although it seems obvious that an ECG is required following exposure to a drug used for cardiovascular indications, many drugs with no overt cardiovascular effects from therapeutic dosing become cardiotoxic in overdose. An ECG should be examined extremely early in the initial evaluation of most poisoned patients.

### **6. References**


ECG is a valuable source of information in poisoned patients and has the potential to enhance and direct their care. Although it seems obvious that an ECG is required following exposure to a drug used for cardiovascular indications, many drugs with no overt cardiovascular effects from therapeutic dosing become cardiotoxic in overdose. An ECG should be examined extremely early in the initial evaluation of most poisoned patients.

Albertson, T.E. (2004). Amphetamines, In: *Poisoning & Drug Overdose, 4th Ed.*, Olson, K.R. et

Albertson, T.E.; Chan, A. & Tharratt, R.S. (2007). Cocaine, In: *Haddad and Winchester's Clinical* 

pp. 755-772, Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA. Albertson, T.E.; Kenyon, N.J. & Morrissey, B. (2007). Amphetamines and derivatives, In:

Anderson, A.C. (2008). Management of Beta-Adrenergic Blocker Poisoning. *Clinical Pediatric Emergency Medicine*, Vol. 9, No. 1, (March, 2008), pp.4-16, ISSN 1522-8401. Benowitz, N.L. (2004). Cocaine, In: In: *Poisoning & Drug Overdose, 4th Ed.*, Olson, K.R. et al.

Berkovitch, M., et al. (1995). Assessment of the terminal 40-millisecond QRS vector in

Bird, S.B.(2007). Beta-Adrenergic anatagonists, In: *Haddad and Winchester's Clinical* 

pp. 975-982, Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA. Brubacher, J.R. (2007). Beta-adrenergic antagonists, In: *Goldfrank's Manual of Toxicologic* 

Clancy, C. (2007). Electrocardiographic principles, In: *Goldfrank's Manual of Toxicologic* 

Delk, C.; Holstege, C.P. & Brady, W.J. (2007). Electrocardiographic abnormalities associated

*Care*, Vol.11, No.2, (Apr 1995), pp.75–77, ISSN 0749-5161.

Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA.

al. (Eds.), pp. 72-74, Lange Medical Books/McGraw-Hill, ISBN 0-8385-8172-2, New

*Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.),

*Haddad and Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 781-792, Saunders Elsevier, ISBN 978-0-7216-0693-

(Eds.), pp. 171-173, Lange Medical Books/McGraw-Hill, ISBN 0-8385-8172-2, New

children with a history of tricyclic antidepressant ingestion. *Pediatric Emergency* 

*Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.),

*Emergencies*, Hoffman R.S. et al. (Eds.), pp. 518-529, The McGraw-Hill Companies, Inc., http://www.mhprofessional.com/product.php?cat=116&isbn=007144310X. Chung, D.C. (1981). Anaesthetic problems associated with the treatment of cardiovascular

disease: I. Digitalis toxicity. *Can Anaesth Soc J.*, Vol. 28, No.1, (Jan 1981), pp.6-16,

*Emergencies*, Hoffman R.S. et al. (Eds.), pp. 29-40, The McGraw-Hill Companies, Inc., http://www.mhprofessional.com/product.php?cat=116&isbn=007144310X. Delgado, J. (2007). Marijuana, In: *Haddad and Winchester's Clinical Management of Poisoning* 

*and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 747-754, Saunders

with poisoning. *American Journal of Emergency Medicine*, Vol.25, No. 6, (July 2007),

**5. Conclusions** 

**6. References** 

York, USA.

York, USA.

ISSN 0008-2856.

pp. 672-687, ISSN 0735-6757.

4, Philadelphia, USA.


**16** 

*Taiwan* 

**Electrocardiogram (ECG) Abnormality** 

**Taiwan – A Study of Gene-Gene and** 

*National Health Research Institutes and National Taiwan University* 

Natural occurrence of arsenic in groundwater is found in the Americas, European, Western Africa, and Asia including Taiwan, Japan, southern Thailand and China where in some areas, drinking water supplies are primarily based on groundwater resources 2. For general population in southwestern coast Taiwan, the major arsenic exposure resource is the ingestion of arsenic contaminated groundwater. The residents have used high-arsenic contaminated well water for drinking and cooking for many decades since early 1910s. The tap water supply system was implemented in the early 1960s, however artesian well water

Because arsenic toxicity operates in a highly nonlinear manner and different levels of exposure measurements applied result in large discrepancy across studies and made it difficult to come up a reliable dose-response relationship for arsenic hazard. There is a long-standing observation of individual variability in susceptibility to arsenic toxicity 4 and this variation may be partly due to differences in age and sex distribution across areas, and also individual arsenic metabolism capabilities 5,6. Inter-individual differences in the speciation and amounts of arsenic metabolites have been reported among subjects chronically exposed to arsenic 7 and significant genetic determinants of arsenic metabolism was supported by epidemiological study 8. Toenail and hair arsenic has been reported to provided an integrated measure of internal arsenic exposure 9. However epidemiologic studies showed that external contamination lead to overestimation of internal dose and urinary arsenic concentration seems to be a better marker than concentrations in drinking water10. Growing epidemiological evidence also suggests that some factors such as age, sex and genetic susceptibility are related to its metabolism and can be important predictors for arsenic-related health hazard 11,12.

Heart disease or cardiovascular disease is defined as the class of disease that involved the cardiac or blood vessels including arteries and veins. Although the term technically refers to

has not been used for drinking or cooking until mid-1970s 3.

**1.1 Nature history of cardiovascular disease (CVD)** 

**1. Introduction** 

**Gene-Environment Interactions** 

Ya-Tang Liao, Wan-Fen Li, Chien-Jen Chen, Wei J. Chen, Hsiao-Yen Chen and Shu-Li Wang

**Among Residents in Arseniasis-Endemic** 

**and Non-Endemic Areas of Southwestern** 


http://emedicine.medscape.com/article/832840-overview.


### **Electrocardiogram (ECG) Abnormality Among Residents in Arseniasis-Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and Gene-Environment Interactions**

Ya-Tang Liao, Wan-Fen Li, Chien-Jen Chen, Wei J. Chen, Hsiao-Yen Chen and Shu-Li Wang *National Health Research Institutes and National Taiwan University Taiwan* 

### **1. Introduction**

296 Advances in Electrocardiograms – Clinical Applications

Quang, L.S. (2007). GHB and related compounds, In: *Haddad and Winchester's Clinical* 

pp. 803-823, Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA. Roden, D.M. (2004). Drug-induced prolongation of the QT interval. *New England Journal of Medicine*, Vol. 350, No.10, (March, 2004), pp.1013-1022, ISSN 0028-4793. Satran, D. et al. (2005). Cardiovascular Manifestations of Moderate to Severe Carbon

Schraga, E.D. et al. (Updated: Mar 11, 2008). Hydrogen Cyanide Poisoning, In: *Medscape* 

Smolinske, S.C.; Daubert, G.P. & Spoerke, D.G. (2007). Poisonous plants, In: *Haddad and* 

Thomas, J.D.; Thomas, K.E. & Kazzi, Z.N. (2007). Scorpion and stinging insects, In: *Haddad* 

Traub, S.J. (2007). Hallucinogens, In: *Haddad and Winchester's Clinical Management of* 

Wandersee, S. (2006). Toxic foods, In: *The Toxicology Handbook for Clinicians*, Harris, C.R. (Ed.), pp. 194-204, Mosby Elsevier, ISBN 1-56053-711-6, Philadelphia, USA. Wiley II, J.F. (2007). Clonidine and related imidazoline derivatives, In: *Haddad and* 

Wolfe, T.R.; Caravati, E.M. & Rollins, D.E. (1989). Terminal 40-ms frontal plane QRS axis as a

Yip, L.; Mégarbane, B. & Borron, S.W. (2007). Opioids, In: *Haddad and Winchester's Clinical* 

pp. 635-658, Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA.

Saunders Elsevier, ISBN 978-0-7216-0693-4, Philadelphia, USA.

No.4, (Apr 1989), pp.348–351, ISSN 0196-0644.

(June 2005), pp. 1513–1516, ISSN 0735-1097.

http://emedicine.medscape.com/article/832840-overview.

*Reference*, accessed 16 April, 2011,

Philadelphia, USA.

4, Philadelphia, USA.

Philadelphia, USA.

*Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.),

Monoxide Poisoning. *Journal of the American College of Cardiology,* Vol. 45, No. 9,

*Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 473-506, Saunders Elsevier, ISBN 978-0-7216-0693-4,

*and Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 440-454, Saunders Elsevier, ISBN 978-0-7216-0693-

*Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 793-802,

*Winchester's Clinical Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.), pp. 1001-1008, Saunders Elsevier, ISBN 978-0-7216-0693-4,

marker for tricyclic antidepressant overdose. *Annals of Emergency Medicine*, Vol.18,

*Management of Poisoning and Drug Overdose* (4th Ed.), Shannon, M.W. et al. (Eds.),

Natural occurrence of arsenic in groundwater is found in the Americas, European, Western Africa, and Asia including Taiwan, Japan, southern Thailand and China where in some areas, drinking water supplies are primarily based on groundwater resources 2. For general population in southwestern coast Taiwan, the major arsenic exposure resource is the ingestion of arsenic contaminated groundwater. The residents have used high-arsenic contaminated well water for drinking and cooking for many decades since early 1910s. The tap water supply system was implemented in the early 1960s, however artesian well water has not been used for drinking or cooking until mid-1970s 3.

Because arsenic toxicity operates in a highly nonlinear manner and different levels of exposure measurements applied result in large discrepancy across studies and made it difficult to come up a reliable dose-response relationship for arsenic hazard. There is a long-standing observation of individual variability in susceptibility to arsenic toxicity 4 and this variation may be partly due to differences in age and sex distribution across areas, and also individual arsenic metabolism capabilities 5,6. Inter-individual differences in the speciation and amounts of arsenic metabolites have been reported among subjects chronically exposed to arsenic 7 and significant genetic determinants of arsenic metabolism was supported by epidemiological study 8. Toenail and hair arsenic has been reported to provided an integrated measure of internal arsenic exposure 9. However epidemiologic studies showed that external contamination lead to overestimation of internal dose and urinary arsenic concentration seems to be a better marker than concentrations in drinking water10. Growing epidemiological evidence also suggests that some factors such as age, sex and genetic susceptibility are related to its metabolism and can be important predictors for arsenic-related health hazard 11,12.

### **1.1 Nature history of cardiovascular disease (CVD)**

Heart disease or cardiovascular disease is defined as the class of disease that involved the cardiac or blood vessels including arteries and veins. Although the term technically refers to

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

and arsenic biotransformation efficiency of the cell 34.

**1.6 Other genes may linked to arsenic-related CVD** 

**1.7 Arsenic-related early effect-biomarkers for CVD** 

enzymes.

**1.5 Atherosclerosis- and CVD-related genes** 

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 299

methylation levels in urine 31 which suggests that there are genetic factors in the regulation of the enzymes that metabolize arsenic, which may lead to difference in toxicity related to arsenic exposure. Not until recently have genes encoding enzymes that are responsible for arsenic metabolism been cloned and characterized. These genes include AS3MT and GSTO. The AS3MT gene directly encodes a cytosolic enzyme, arsenic methyltransferase, which catalyzes the multi-step process to convert inorganic arsenic to monomethyl arsenical (MMA) and dimethyl arsenical (DMA) 32. Glutathione S-transferases (GSTs) are Phase II detoxification enzymes that catalyze the conjugation of reduced glutathione (GSH) to a wide variety of endogenous and exogenous electrophilic compounds 33. A subfamily of GSTs, GST omega class,was shown to be identical with human monomethylarsonic acid (MMAV) reductase that is the rate-limiting enzyme for biotransformation of inorganic arsenic. Polymorphisms of GSTO genes were shown associated with the intracellular thiol status

High-density lipoprotein (HDL) is postulated to prevent the development of atherosclerosis by inhibiting the oxidation of low-density lipoprotein (LDL). Human paraoxonase (PON1) is a serum esterase/lactonase transported on HDL particles which is considered the major determinant of the antioxidant action of HDL 35. Both in vitro studies and animal studies using PON1-knockout mice have shown that PON1 can prevent both HDL and LDL oxidation and therefore a protective enzyme against the development of atherosclerosis 36-38. Our previous data also showed significant synergistic effects of genetic variations in the

PON gene cluster and chronic arsenic exposure on electrocardiogram abnormality 39.

A recent review article also pointed out that some other genes that associated with arsenic toxicity and altered gene expression in humans including genes involved in stress response, DNA damage response and apoptosis related genes, cell cycle regulatory genes and cell signaling and altered growth factor 40. Evidence from experimental studies had also suggested that arsenic increases the production of reactive oxygen species (ROS) 41-43. The induction of oxidative stress by arsenic may influence gene expression, inflammatory responses, and endothelial nitric oxide homeostasis 44, which play an important role in maintaining vascular tone. Genes involved in endogenous defenses against ROS thus may modify arsenic's effect. Genetic material is constantly being subjected to insult from a wide range of DNA damage agents and this damage is controlled by the action of DNA repair

Preclinical or subclinical disease was defined as the pathological changes in the heart and arteries that develop early in the course of cardiovascular disease before symptoms or morbid events occur. They were developed before the evaluation of any of the risk factors or the determination of the subsequent incidence and mortality and thus are unbiased. Persons with subclinical disease, regardless of whether other risk factors are present, are at greater risk of future cardiovascular events than are those without subclinical disease 45. Although standard methods of cardiac risk assessment from the physical examination, laboratory tests, and treadmill exercise testing are often used clinically in cardiovascular disease

stratification 46, evaluation of subclinical disease often not taken into account.

any disease that affects the cardiovascular system, it is usually to refer to those related to atherosclerosis and arterial disease since they shared similar conditions of causes, mechanisms and treatments13. The primary underlying disease process that leads to atherosclerosis is the deposition of lipid on the arterial surface progress to form plaques that reduced blood flow and induced blood clots that blocked flow entirely 14.

Most countries face high and increasing rates of cardiovascular disease. In United States, mortality from heart and hypertensive diseases was greater than mortality from neoplasm. In recent years, cardiovascular risk in women has been increasing and has killed more women than breast cancer 15. The estimated age-adjusted mortalities of cardiovascular disease in US is 152.1 per 100,000 in year 2002 and is 48.3 in Taiwan 2005 16. By the time that heart problems are detected, the underlying causes, atherosclerosis, is usually quite advanced, have progressed for decades 17. Therefore increased emphasis on preventing atherosclerosis by modifying risk factors is remained important.

### **1.2 Arsenic-related CVD**

Chronic arsenic exposure can lead to hyperkeratosis and loss of skin pigmentation as well as cancers of the skin, bladder, and lung. The international Agency for Research on Cancer 18 and the U.S. EPA 19 have classified arsenic as a group 1 and group A carcinogen based on human evidence. However, the mechanism of action for iAs-induced carcinogenicity is not known 20. Cardiovascular death is the major cause of mortality worldwide, and a small increased risk may imply a large quantity of excess mortality 21. Arsenic has also been shown to be a major risk factor of an unique form of peripheral vascular disease, Blackfoot disease (BFD) 22,23, especially in southwest Taiwan. Although the etiology of BFD development is still unclear, the dose-response relationships between arsenic and the prevalence of cardiovascular diseases have been documented including atherosclerosis, peripheral vascular disease (PVD), ischemic heart disease (IHD), hypertension, and cerebrovascular disease 21. IHD is a disease characterized by reduced blood supply to heart muscle, usually due to atherosclerosis of the coronary arteries. Its risk increases with age, smoking, hypercholesterolemia, diabetes, hypertension, and is more common in men and those who have close relatives with ischemic heart disease. Standard diagnosis of IHD including electrocardiogram, blood tests with cardiac enzymes, history and physical examinations.

Although both population-based and occupational studies had shown that long-term exposure to inorganic arsenic has significant toxic effects on the cardiovascular system and the maximum arsenic contamination level in drinking water has been lowered from 0.05 to 0.01 ppm by US Environmental Protection Agency in 2006. The allowable limit for arsenic in drinking water is 0.025 for Canada 24 and 0.05 ppm for Bengal, India, and Bangladesh 25. , the long-term association with chronic exposure to arsenic remains unclear and there is still epidemiological evidence needed for developing regulatory guidelines 26,27.

#### **1.3 Genetic factors associated with arsenic metabolism and CVD**

Although the mechanism of action for iAs-induced CVD hazards is not known 20. Proposed mechanisms include arsenic metabolism, pathogenesis of atherosclerosis and oxidative stress 28-30. More than one of these mechanisms may occur, and some may work together.

#### **1.4 Arsenic metabolism genes**

In general, the distribution of arsenic in human urine is 10-30% iAs, 10-20% MMA(V), and 60-70% DMA(V) 6. However, some populations showed significant variation of arsenic methylation levels in urine 31 which suggests that there are genetic factors in the regulation of the enzymes that metabolize arsenic, which may lead to difference in toxicity related to arsenic exposure. Not until recently have genes encoding enzymes that are responsible for arsenic metabolism been cloned and characterized. These genes include AS3MT and GSTO. The AS3MT gene directly encodes a cytosolic enzyme, arsenic methyltransferase, which catalyzes the multi-step process to convert inorganic arsenic to monomethyl arsenical (MMA) and dimethyl arsenical (DMA) 32. Glutathione S-transferases (GSTs) are Phase II detoxification enzymes that catalyze the conjugation of reduced glutathione (GSH) to a wide variety of endogenous and exogenous electrophilic compounds 33. A subfamily of GSTs, GST omega class,was shown to be identical with human monomethylarsonic acid (MMAV) reductase that is the rate-limiting enzyme for biotransformation of inorganic arsenic. Polymorphisms of GSTO genes were shown associated with the intracellular thiol status and arsenic biotransformation efficiency of the cell 34.

### **1.5 Atherosclerosis- and CVD-related genes**

298 Advances in Electrocardiograms – Clinical Applications

any disease that affects the cardiovascular system, it is usually to refer to those related to atherosclerosis and arterial disease since they shared similar conditions of causes, mechanisms and treatments13. The primary underlying disease process that leads to atherosclerosis is the deposition of lipid on the arterial surface progress to form plaques that

Most countries face high and increasing rates of cardiovascular disease. In United States, mortality from heart and hypertensive diseases was greater than mortality from neoplasm. In recent years, cardiovascular risk in women has been increasing and has killed more women than breast cancer 15. The estimated age-adjusted mortalities of cardiovascular disease in US is 152.1 per 100,000 in year 2002 and is 48.3 in Taiwan 2005 16. By the time that heart problems are detected, the underlying causes, atherosclerosis, is usually quite advanced, have progressed for decades 17. Therefore increased emphasis on preventing

Chronic arsenic exposure can lead to hyperkeratosis and loss of skin pigmentation as well as cancers of the skin, bladder, and lung. The international Agency for Research on Cancer 18 and the U.S. EPA 19 have classified arsenic as a group 1 and group A carcinogen based on human evidence. However, the mechanism of action for iAs-induced carcinogenicity is not known 20. Cardiovascular death is the major cause of mortality worldwide, and a small increased risk may imply a large quantity of excess mortality 21. Arsenic has also been shown to be a major risk factor of an unique form of peripheral vascular disease, Blackfoot disease (BFD) 22,23, especially in southwest Taiwan. Although the etiology of BFD development is still unclear, the dose-response relationships between arsenic and the prevalence of cardiovascular diseases have been documented including atherosclerosis, peripheral vascular disease (PVD), ischemic heart disease (IHD), hypertension, and cerebrovascular disease 21. IHD is a disease characterized by reduced blood supply to heart muscle, usually due to atherosclerosis of the coronary arteries. Its risk increases with age, smoking, hypercholesterolemia, diabetes, hypertension, and is more common in men and those who have close relatives with ischemic heart disease. Standard diagnosis of IHD including electrocardiogram, blood tests with cardiac

Although both population-based and occupational studies had shown that long-term exposure to inorganic arsenic has significant toxic effects on the cardiovascular system and the maximum arsenic contamination level in drinking water has been lowered from 0.05 to 0.01 ppm by US Environmental Protection Agency in 2006. The allowable limit for arsenic in drinking water is 0.025 for Canada 24 and 0.05 ppm for Bengal, India, and Bangladesh 25. , the long-term association with chronic exposure to arsenic remains unclear and there is still

Although the mechanism of action for iAs-induced CVD hazards is not known 20. Proposed mechanisms include arsenic metabolism, pathogenesis of atherosclerosis and oxidative stress 28-30. More than one of these mechanisms may occur, and some may work together.

In general, the distribution of arsenic in human urine is 10-30% iAs, 10-20% MMA(V), and 60-70% DMA(V) 6. However, some populations showed significant variation of arsenic

epidemiological evidence needed for developing regulatory guidelines 26,27.

**1.3 Genetic factors associated with arsenic metabolism and CVD** 

reduced blood flow and induced blood clots that blocked flow entirely 14.

atherosclerosis by modifying risk factors is remained important.

**1.2 Arsenic-related CVD** 

enzymes, history and physical examinations.

**1.4 Arsenic metabolism genes** 

High-density lipoprotein (HDL) is postulated to prevent the development of atherosclerosis by inhibiting the oxidation of low-density lipoprotein (LDL). Human paraoxonase (PON1) is a serum esterase/lactonase transported on HDL particles which is considered the major determinant of the antioxidant action of HDL 35. Both in vitro studies and animal studies using PON1-knockout mice have shown that PON1 can prevent both HDL and LDL oxidation and therefore a protective enzyme against the development of atherosclerosis 36-38. Our previous data also showed significant synergistic effects of genetic variations in the PON gene cluster and chronic arsenic exposure on electrocardiogram abnormality 39.

### **1.6 Other genes may linked to arsenic-related CVD**

A recent review article also pointed out that some other genes that associated with arsenic toxicity and altered gene expression in humans including genes involved in stress response, DNA damage response and apoptosis related genes, cell cycle regulatory genes and cell signaling and altered growth factor 40. Evidence from experimental studies had also suggested that arsenic increases the production of reactive oxygen species (ROS) 41-43. The induction of oxidative stress by arsenic may influence gene expression, inflammatory responses, and endothelial nitric oxide homeostasis 44, which play an important role in maintaining vascular tone. Genes involved in endogenous defenses against ROS thus may modify arsenic's effect. Genetic material is constantly being subjected to insult from a wide range of DNA damage agents and this damage is controlled by the action of DNA repair enzymes.

### **1.7 Arsenic-related early effect-biomarkers for CVD**

Preclinical or subclinical disease was defined as the pathological changes in the heart and arteries that develop early in the course of cardiovascular disease before symptoms or morbid events occur. They were developed before the evaluation of any of the risk factors or the determination of the subsequent incidence and mortality and thus are unbiased. Persons with subclinical disease, regardless of whether other risk factors are present, are at greater risk of future cardiovascular events than are those without subclinical disease 45. Although standard methods of cardiac risk assessment from the physical examination, laboratory tests, and treadmill exercise testing are often used clinically in cardiovascular disease stratification 46, evaluation of subclinical disease often not taken into account.

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

**2. Materials and methods 2.1 Study area and population** 

and a total of 303 subjects were recruited.

**2.2 Measurement of arsenic exposure** 

association between cumulative arsenic exposure and the risk of CVD.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 301

impacts on cardiovascular disease through measuring ECG abnormality as subclinical phenotypes and to evaluate whether the arsenic methylation patterns modifies the

The study included a community-based cohort from previous arseniasis-endemic area in southwestern Taiwan and a non-exposed population recruited from documented nonendemic area in the same county with similar age, gender contribution and ecological status in 2002. The arseniasis-endemic area included Homei, Fusin and Hsinming villages in Putai Township on the southwestern coast of Taiwan which had been described previously 56-58. In short, residents in the study area consumed high-arsenic contaminated well water for decades since the 1910s because of the high salinity in shallow village wells 23. The arsenic concentration of artesian well water measured in the early 1960s was from 0.035 to 1.14 ppm, with a median of 0.78 ppm 59,60. An estimated total daily amount of arsenic ingested by local residents was as high as 1 mg, mainly from drinking water 61. A tap water supply system was implemented in the area in the early 1960s and the entire arseniasis-endemic area has been supplied with municipal water since the early 1970s. The arsenic concentration of tap water supplied in the study area was less than 0.01 ppm 62. The original cohort established in 1989 including 1571 residents and 1081 subjects provided informed consents and enrolled in the study cohort. In 1993, 732 residents from the villages had a 12 lead baseline Electrocardiogram (ECG) recorded. In 2002, after an average follow up period of eight years, 216 out of 380 subjects recruited provided a second ECG recording; 141 of them provided blood and urine specimens without an ECG recording; 229 were dead and their mortality determined through linkage with the national database; and the remaining 146 were lost to follow-up. Among the 121 residents with normal baseline ECGs, 42 developed an ECG abnormality at follow up. The non-exposed area was Chiali Township where the arsenic concentration of well water was very low according to the results of surveys conducted in 1960s and 1970s60,63. Climate, ethnic background (Han Chinese), urbanization degree and socioeconomic status were similar between Putai and Chiali. Frequency matching by age strata and gender were conducted for recruitment of resident

Arsenic level of well water for this study area was measured by the National Taiwan University group60. The water-contained arsenic recovery efficiencies were 95 percent or greater and were obtained using a PerkinElmer UV-VIS Spectrophotometer incorporating with Klett-Summerson Colorimeter. Detail validations of the water arsenic levels have been presented previously57,64. For villages which used more than one artesian well as a source of potable water, the medial levels of water arsenic contamination across those wells were assigned. The arsenic levels in artesian well water in this study area have been reported to be stable65. An index of cumulative arsenic exposure (micrograms per liter-years) were defined as the summation of products derived by multiplying the arsenic concentration (in micrograms per liter) in well water by duration of water consumption (in years) during

Epidemiology studies have showed that the population attributable fractions to CHD of subclinical disease was 36.8% and 42.5% for men and women respectively, which is much higher than for most of the known risk factors or combination of risk factors and further documents the importance of subclinical disease as a contributor to subsequent incident clinical disease 45. Among coronary artery disease (CAD) patients, the preoperative electrocardiogram (ECG) is shown to be predictive of long-term outcome independent of clinical findings and perioperative ischemia 47. Moreover, unrecognized silent myocardial infarction as diagnosed by electrographic changes is a major risk factor for subsequent myocardial infarction and coronary disease deaths 48, and it is also a useful additional tool for differentiating the x-lined form of hereditary cardiac myopathies 49. Subclinical disease and clinical disease shared similar risk factors and thus aggressive interventions to prevent clinical disease should be oriented to individuals with subclinical disease 50.

Various ECG abnormalities have been observed among cases of acute arsenic poisoning and in acute promyelocytic leukemia patients treated with arsenic trioxide. Individuals exposed to excess arsenic through drinking water showed some of the ECG abnormalities 51. QT prolongation and dispersion have been implicated in the genesis of ventricular arrhythmia and directly predictors of cardiovascular and all-cause mortality 52,53. The gradient relationship of chronic arsenic poisoning and prolonged QT interval and increased QT dispersion has been reported recently 54,55 and arsenic-induced QT dispersion was associated with atherosclerosis disease and predicted cardiovascular mortality. However, evidence was based on risk assessment on subjects with previous exposure to high arsenic level and biomarkers for methylation metabolism were not considered. Besides, the accuracy and reproducibility of ECG reading including QT dispersion measurement have been restricted by difficulties with reliable determination of T-waves offset. Further study with a standardize measurement of ECG reading is warranted for a reliable assessment of ECG abnormality.

Although an association between chronic arsenic exposure and CVD has been found in many studies, nearly all of these studies were limited by use of cross-sectional data, and longitudinal evidence by follow-up study was still limited. Besides, majority of previous studies were focus on the clinical arsenic-related cardiovascular disease, instead of the manifest of preclinical or subclinical detections. Morbidity and mortality from peripheral vascular disease, ischemic heart disease, and cerebral infarction are relative late clinical manifestations of chronic arsenic damage. These health effects may be the consequence of the interactions between predisposing and precipitating factors for cardiovascular diseases. The risk assessment based on these late cardiovascular events may be underestimated due to competing causes of death and the correctness in the diagnosis of the sudden death from cardiovascular diseases. Studies based on subclinical finding including ECG abnormality are needed to detect the early sign of chronic poisoning.

Furthermore, the variation in distribution of arsenic in human urine across areas 31 suggested that there are genetic factors in the regulation of the enzymes that metabolize arsenic, which may lead to difference in toxicity related to arsenic exposure. Association studies based on genetic polymorphisms have not provided consensus data that could generate a viable hypothesis on the molecular mechanism that determines the genetic basis of arsenic toxicity. The major objective of this study is to investigate the joint contribution of genetic factors including PON1, AS3MT, and GSTO gene families and the long-term arsenic

impacts on cardiovascular disease through measuring ECG abnormality as subclinical phenotypes and to evaluate whether the arsenic methylation patterns modifies the association between cumulative arsenic exposure and the risk of CVD.

### **2. Materials and methods**

300 Advances in Electrocardiograms – Clinical Applications

Epidemiology studies have showed that the population attributable fractions to CHD of subclinical disease was 36.8% and 42.5% for men and women respectively, which is much higher than for most of the known risk factors or combination of risk factors and further documents the importance of subclinical disease as a contributor to subsequent incident clinical disease 45. Among coronary artery disease (CAD) patients, the preoperative electrocardiogram (ECG) is shown to be predictive of long-term outcome independent of clinical findings and perioperative ischemia 47. Moreover, unrecognized silent myocardial infarction as diagnosed by electrographic changes is a major risk factor for subsequent myocardial infarction and coronary disease deaths 48, and it is also a useful additional tool for differentiating the x-lined form of hereditary cardiac myopathies 49. Subclinical disease and clinical disease shared similar risk factors and thus aggressive interventions to prevent

Various ECG abnormalities have been observed among cases of acute arsenic poisoning and in acute promyelocytic leukemia patients treated with arsenic trioxide. Individuals exposed to excess arsenic through drinking water showed some of the ECG abnormalities 51. QT prolongation and dispersion have been implicated in the genesis of ventricular arrhythmia and directly predictors of cardiovascular and all-cause mortality 52,53. The gradient relationship of chronic arsenic poisoning and prolonged QT interval and increased QT dispersion has been reported recently 54,55 and arsenic-induced QT dispersion was associated with atherosclerosis disease and predicted cardiovascular mortality. However, evidence was based on risk assessment on subjects with previous exposure to high arsenic level and biomarkers for methylation metabolism were not considered. Besides, the accuracy and reproducibility of ECG reading including QT dispersion measurement have been restricted by difficulties with reliable determination of T-waves offset. Further study with a standardize measurement of ECG reading is warranted for a reliable assessment of

Although an association between chronic arsenic exposure and CVD has been found in many studies, nearly all of these studies were limited by use of cross-sectional data, and longitudinal evidence by follow-up study was still limited. Besides, majority of previous studies were focus on the clinical arsenic-related cardiovascular disease, instead of the manifest of preclinical or subclinical detections. Morbidity and mortality from peripheral vascular disease, ischemic heart disease, and cerebral infarction are relative late clinical manifestations of chronic arsenic damage. These health effects may be the consequence of the interactions between predisposing and precipitating factors for cardiovascular diseases. The risk assessment based on these late cardiovascular events may be underestimated due to competing causes of death and the correctness in the diagnosis of the sudden death from cardiovascular diseases. Studies based on subclinical finding including ECG abnormality are

Furthermore, the variation in distribution of arsenic in human urine across areas 31 suggested that there are genetic factors in the regulation of the enzymes that metabolize arsenic, which may lead to difference in toxicity related to arsenic exposure. Association studies based on genetic polymorphisms have not provided consensus data that could generate a viable hypothesis on the molecular mechanism that determines the genetic basis of arsenic toxicity. The major objective of this study is to investigate the joint contribution of genetic factors including PON1, AS3MT, and GSTO gene families and the long-term arsenic

clinical disease should be oriented to individuals with subclinical disease 50.

ECG abnormality.

needed to detect the early sign of chronic poisoning.

### **2.1 Study area and population**

The study included a community-based cohort from previous arseniasis-endemic area in southwestern Taiwan and a non-exposed population recruited from documented nonendemic area in the same county with similar age, gender contribution and ecological status in 2002. The arseniasis-endemic area included Homei, Fusin and Hsinming villages in Putai Township on the southwestern coast of Taiwan which had been described previously 56-58. In short, residents in the study area consumed high-arsenic contaminated well water for decades since the 1910s because of the high salinity in shallow village wells 23. The arsenic concentration of artesian well water measured in the early 1960s was from 0.035 to 1.14 ppm, with a median of 0.78 ppm 59,60. An estimated total daily amount of arsenic ingested by local residents was as high as 1 mg, mainly from drinking water 61. A tap water supply system was implemented in the area in the early 1960s and the entire arseniasis-endemic area has been supplied with municipal water since the early 1970s. The arsenic concentration of tap water supplied in the study area was less than 0.01 ppm 62. The original cohort established in 1989 including 1571 residents and 1081 subjects provided informed consents and enrolled in the study cohort. In 1993, 732 residents from the villages had a 12 lead baseline Electrocardiogram (ECG) recorded. In 2002, after an average follow up period of eight years, 216 out of 380 subjects recruited provided a second ECG recording; 141 of them provided blood and urine specimens without an ECG recording; 229 were dead and their mortality determined through linkage with the national database; and the remaining 146 were lost to follow-up. Among the 121 residents with normal baseline ECGs, 42 developed an ECG abnormality at follow up. The non-exposed area was Chiali Township where the arsenic concentration of well water was very low according to the results of surveys conducted in 1960s and 1970s60,63. Climate, ethnic background (Han Chinese), urbanization degree and socioeconomic status were similar between Putai and Chiali. Frequency matching by age strata and gender were conducted for recruitment of resident and a total of 303 subjects were recruited.

### **2.2 Measurement of arsenic exposure**

Arsenic level of well water for this study area was measured by the National Taiwan University group60. The water-contained arsenic recovery efficiencies were 95 percent or greater and were obtained using a PerkinElmer UV-VIS Spectrophotometer incorporating with Klett-Summerson Colorimeter. Detail validations of the water arsenic levels have been presented previously57,64. For villages which used more than one artesian well as a source of potable water, the medial levels of water arsenic contamination across those wells were assigned. The arsenic levels in artesian well water in this study area have been reported to be stable65. An index of cumulative arsenic exposure (micrograms per liter-years) were defined as the summation of products derived by multiplying the arsenic concentration (in micrograms per liter) in well water by duration of water consumption (in years) during

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

polymorphism-specific restriction enzyme digestion and gel analysis.

SYNCHRON LX20 System (Beckman Coulter, Fullerton, CA).

chi-square analyses. Linkage disequilibrium (LD) as measured by D'

**2.6 Physical examination** 

**2.7 Statistical analysis** 

Cary, NC).

**3. Results** 

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 303

determined using a commercially designed TaqMan SNP Genotyping Assay (Applied Biosystems, USA). All other genotypes will be conducted by PCR amplification followed by

Resting twelve-lead conventional ECG recording was performed at the Beimen Branch, Shinyin Hospital. Minnesota standardized code classification 1 was evaluated for both baseline and follow-up ECG readings at Epidemiological Cardiology Research Center (EPICARE), Department of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA (blinded to all other study data). ECG readings were classified into normal and abnormal (including minor and major abnormality) according to the definition of cardiac function by myocardial infarction or ischemia (Q wave and STT change) (MC\_1, MC\_4, MC\_5, MC\_92), conduction defect (MC\_7), arrhythmias (MC\_6, MC\_81~MC\_88), atrial enlargement or ventricular hypertrophy (LVH\_MC3/LVH\_CV), and prolonged ventricular repolarization. Fasting plasma was analyzed for blood glucose, cholesterol, triglycerides, high- and low-density lipoproteins, and urine acid by Beckmen

Differences in demographic characteristics and cardiovascular risk factors between ECG normal and abnormal subjects were assessed. Continuous variables were expressed as mean standard deviation (SD) and evaluated by student's t or Wilcoxon rank-sum test. Categorical variables were expressed as proportions and compared using chi-square test or Fisher's exact test. Allele frequencies, genotype frequencies, and Hardy-Weinberg equilibrium were assessed separately in ECG abnormal and normal groups using SAS-genetics package. Relative distribution of polymorphisms in the ECG abnormal and ECG normal groups was assessed by

Haploview 4.0 (http://www.broad.mit.edu/mpg/haploview/). Haplotypes and tag SNPs were inferred using SAS. Logistic regression analysis was used to assess the effect of cardiovascular risk factors and genetic polymorphisms in relation to ECG abnormality. Arsenic exposure and ECG abnormality in between study subjects in Putai and Chiali areas were also compared. Arsenic exposure in Putai area were stratified into two categories by median levels and subjects in Chiali area were used as reference group, and a trend test was conducted to evaluate the dose-relationship. ANOVA was conducted to evaluate urinary arsenic species between subjects with normal and abnormal ECG reading. A *p* value <0.05 was considered statistically significant. Permutation test, a significance test used to obtain the unknown reference distribution by calculating all possible values of the test statistic under random rearrangements of the disease status on the observed study subjects , was used to control for type 1 error for multiple testing due to the limited sample size, and the empirical pvalues were reported 67. Statistical analyses were conducted using SAS version 9.1 (SAS, Inc.,

Baseline characteristics of arsenic exposure and cardiovascular risk factors among study subjects are summarized in Table 1. A total of 42 incident cases among the 121 baseline-

was assessed using

consecutive periods of living in the defacement villages. Both cumulative arsenic exposure and average arsenic concentrations in drinking water were calculated only for subjects who had complete information on arsenic exposure from drinking water throughout his or her lifetime.

### **2.3 Questionnaire interview**

At both baseline and follow-up, well trained public health nurses carried out the standardized personal interview based on a structured questionnaire to acquire information regarding demographic and socioeconomic characteristics, artesian well water usage, residential history, lifestyle variables, personal and family disease history of hypertension, diabetes, and cardiovascular diseases. Cumulative arsenic exposure (in ppm-years) was derived from the median arsenic concentration in artesian well water (ppm) in the village where the subject lived and the duration of consuming the artesian well water (years)while residing in the village. The human Ethical Committee of the National Health Research Institutes in Taiwan approved the study protocol which based on the ethical standards formulated from the Helsinki Declarations of 1964 and revised in 2000 66. Informed consent was provided to each subject before participation.

### **2.4 Biochemical measurements**

Fasting plasma was used for quantitative determination of blood glucose, cholesterol, triglycerides, high- and low-density lipoproteins, and urine acid were analyzed using the same instrument.

Urinary samples were collected from each subject for arsenic species analyses. Subjects were asked not to consume seafood three days before urine collection. Arsenic species in urine including arsenite (AsIII), arsenate (AsV), monomethyl arsenical (MMA) and dimethyl arsenical (DMA) were quantified using high-performance liquid chromatography (HPLC) coupled with flow injection atomic absorption spectrometry. The HPLC system consisted of a solvent delivery pump (PU-1580, Jasco, Tokyo, Japan) and a silica-based anion-exchange column (Nucleosil 10 SB, 250 mm×4.6 mm; Phenomenex, CA, USA) with a guard column packed with the same material. A flow injection analysis system (FIAS-400, PerkinElmer, CT, USA) was designed as the on-line interface to the continuous hydride generation system (Analyst 100, PerkinElmer, CT, USA) used in this study. With this method, the within-say and between-day precision (coefficient of variance, CV%) for AsIII, AsV, MMA, and DMA determinations ranged from 1.0 to 3.7% were observed. Furthermore, the recoveries for AsIII, AsV, MMA, and DMA were 99.0, 98.9, 99.0, and 99.0% while the detection limits were 0.75, 1.47, 1.19, and 0.76 μg/L, respectively. The primary methylation index (PMI) was defined as the ratio between MMA and iAs levels, and the secondary methylation index (SMI) was defined as the ratio between DMA and MMA.

### **2.5 Genotyping**

Eight functional polymorphisms: C-108T (promoter), L55M (exon 3) and Q192R (exon 6) of PON1, A148G (exon 5) and C311S (exon 9) of PON2, M287T (exon 9) of AS3MT, A140D (exon 4) of GSTO1, and N142D (exon 5) of GSTO2. SNPs were selected from NCBI's SNP database based on prior implication in disease and minor allele frequency. Genomic DNA was extracted from whole blood using standard techniques. The AS3MT M287T polymorphism was polymorphism-specific restriction enzyme digestion and gel analysis.

determined using a commercially designed TaqMan SNP Genotyping Assay (Applied Biosystems, USA). All other genotypes will be conducted by PCR amplification followed by

### **2.6 Physical examination**

302 Advances in Electrocardiograms – Clinical Applications

consecutive periods of living in the defacement villages. Both cumulative arsenic exposure and average arsenic concentrations in drinking water were calculated only for subjects who had complete information on arsenic exposure from drinking water throughout his or her

At both baseline and follow-up, well trained public health nurses carried out the standardized personal interview based on a structured questionnaire to acquire information regarding demographic and socioeconomic characteristics, artesian well water usage, residential history, lifestyle variables, personal and family disease history of hypertension, diabetes, and cardiovascular diseases. Cumulative arsenic exposure (in ppm-years) was derived from the median arsenic concentration in artesian well water (ppm) in the village where the subject lived and the duration of consuming the artesian well water (years)while residing in the village. The human Ethical Committee of the National Health Research Institutes in Taiwan approved the study protocol which based on the ethical standards formulated from the Helsinki Declarations of 1964 and revised in 2000 66. Informed consent

Fasting plasma was used for quantitative determination of blood glucose, cholesterol, triglycerides, high- and low-density lipoproteins, and urine acid were analyzed using the

Urinary samples were collected from each subject for arsenic species analyses. Subjects were asked not to consume seafood three days before urine collection. Arsenic species in urine including arsenite (AsIII), arsenate (AsV), monomethyl arsenical (MMA) and dimethyl arsenical (DMA) were quantified using high-performance liquid chromatography (HPLC) coupled with flow injection atomic absorption spectrometry. The HPLC system consisted of a solvent delivery pump (PU-1580, Jasco, Tokyo, Japan) and a silica-based anion-exchange column (Nucleosil 10 SB, 250 mm×4.6 mm; Phenomenex, CA, USA) with a guard column packed with the same material. A flow injection analysis system (FIAS-400, PerkinElmer, CT, USA) was designed as the on-line interface to the continuous hydride generation system (Analyst 100, PerkinElmer, CT, USA) used in this study. With this method, the within-say and between-day precision (coefficient of variance, CV%) for AsIII, AsV, MMA, and DMA determinations ranged from 1.0 to 3.7% were observed. Furthermore, the recoveries for AsIII, AsV, MMA, and DMA were 99.0, 98.9, 99.0, and 99.0% while the detection limits were 0.75, 1.47, 1.19, and 0.76 μg/L, respectively. The primary methylation index (PMI) was defined as the ratio between MMA and iAs levels, and the secondary methylation index

Eight functional polymorphisms: C-108T (promoter), L55M (exon 3) and Q192R (exon 6) of PON1, A148G (exon 5) and C311S (exon 9) of PON2, M287T (exon 9) of AS3MT, A140D (exon 4) of GSTO1, and N142D (exon 5) of GSTO2. SNPs were selected from NCBI's SNP database based on prior implication in disease and minor allele frequency. Genomic DNA was extracted from whole blood using standard techniques. The AS3MT M287T polymorphism was

lifetime.

**2.3 Questionnaire interview** 

was provided to each subject before participation.

(SMI) was defined as the ratio between DMA and MMA.

**2.4 Biochemical measurements** 

same instrument.

**2.5 Genotyping** 

Resting twelve-lead conventional ECG recording was performed at the Beimen Branch, Shinyin Hospital. Minnesota standardized code classification 1 was evaluated for both baseline and follow-up ECG readings at Epidemiological Cardiology Research Center (EPICARE), Department of Public Health Sciences, Wake Forest University School of Medicine, Winston-Salem, North Carolina, USA (blinded to all other study data). ECG readings were classified into normal and abnormal (including minor and major abnormality) according to the definition of cardiac function by myocardial infarction or ischemia (Q wave and STT change) (MC\_1, MC\_4, MC\_5, MC\_92), conduction defect (MC\_7), arrhythmias (MC\_6, MC\_81~MC\_88), atrial enlargement or ventricular hypertrophy (LVH\_MC3/LVH\_CV), and prolonged ventricular repolarization. Fasting plasma was analyzed for blood glucose, cholesterol, triglycerides, high- and low-density lipoproteins, and urine acid by Beckmen SYNCHRON LX20 System (Beckman Coulter, Fullerton, CA).

### **2.7 Statistical analysis**

Differences in demographic characteristics and cardiovascular risk factors between ECG normal and abnormal subjects were assessed. Continuous variables were expressed as mean standard deviation (SD) and evaluated by student's t or Wilcoxon rank-sum test. Categorical variables were expressed as proportions and compared using chi-square test or Fisher's exact test. Allele frequencies, genotype frequencies, and Hardy-Weinberg equilibrium were assessed separately in ECG abnormal and normal groups using SAS-genetics package. Relative distribution of polymorphisms in the ECG abnormal and ECG normal groups was assessed by chi-square analyses. Linkage disequilibrium (LD) as measured by D' was assessed using Haploview 4.0 (http://www.broad.mit.edu/mpg/haploview/). Haplotypes and tag SNPs were inferred using SAS. Logistic regression analysis was used to assess the effect of cardiovascular risk factors and genetic polymorphisms in relation to ECG abnormality. Arsenic exposure and ECG abnormality in between study subjects in Putai and Chiali areas were also compared. Arsenic exposure in Putai area were stratified into two categories by median levels and subjects in Chiali area were used as reference group, and a trend test was conducted to evaluate the dose-relationship. ANOVA was conducted to evaluate urinary arsenic species between subjects with normal and abnormal ECG reading. A *p* value <0.05 was considered statistically significant. Permutation test, a significance test used to obtain the unknown reference distribution by calculating all possible values of the test statistic under random rearrangements of the disease status on the observed study subjects , was used to control for type 1 error for multiple testing due to the limited sample size, and the empirical pvalues were reported 67. Statistical analyses were conducted using SAS version 9.1 (SAS, Inc., Cary, NC).

### **3. Results**

Baseline characteristics of arsenic exposure and cardiovascular risk factors among study subjects are summarized in Table 1. A total of 42 incident cases among the 121 baseline-

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

Gene SNP ECG status

PON1 Q192R

L55M

C-108T

PON2 C311S

A148G

AS3MT M287T

GSTO1 A140D

GSTO2 N142D

SNPs.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 305

 RR 18 14 1.00 (reference) QR 27 9 0.43 (0.15-1.20) QQ 11 3 0.35 (0.08-1.50)

 LL 48 21 1.00 (reference) LM 14 4 0.65 (0.19-2.22)

 CC 17 7 1.00 (reference) CT 30 15 1.21 (0.41-3.56) TT 15 6 0.97 (0.27-3.54)

 SS 31 18 1.00 (reference) CS 24 7 0.50 (0.18-1.40) CC 2 2 1.72 (0.22-13.30)

 AA 30 15 1.00 (reference) AG 28 10 0.71 (0.28-1.85) GG 3 2 1.33 (0.20-8.86)

 MM 67 30 1.00 (reference) MT 2 1 1.12 (0.10-12.80)

 AA 41 21 1.00 (reference) AD 22 9 0.80 (0.31-2.04) DD 5 2 0.78 (0.14-4.37)

 NN 35 16 1.00 (reference) ND 30 14 1.02 (0.43-2.43) DD 4 2 1.09 (0.18-6.60)

Figure 1 shows the related position and linkage disequilibrium (LD) between SNPs in the PON and GSTO gene clusters. Two SNPs within PON2 (C311S and A148G) and GSTO1- A140D and GSTO2-N142 were in high LD but SNPs within PON1 or adjacent SNPs between PON1 and PON2 (C-108T and C311s) had low LD measurements, implying they were not in the same LD block. Q192R, C-108T, C311S and A140D were identified as tag-

Hardy-Weinberg Equilibrium (HWE) test was conducted among all study subjects

Table 2. Association of SNPs and Hardy-Weinberg equilibrium test

TT 0 0 -

MM 0 0 -

Normal Abnormal OR (95% CI)

normal study subjects showed ECG deterioration at follow-up. Compared to ECG normal subjects, those with an ECG abnormality had significantly higher arsenic exposure as shown by both years of drinking artesian water and cumulative arsenic exposure index. Age and proportion of cigarette smoking in the ECG abnormal group tended to be higher but did not reach statistical significance. No differences were observed in other cardiovascular risk factors including gender, alcohol consumption, BMI, serum lipids, blood pressure, and plasma glucose.


Data are reported as mean ± SD or counts (%)

HDL: high density lipoprotein; LDL: low density lipoprotein; CHOL: total cholesterol levels; SBP: systolic blood pressure; DBP: diastolic blood pressure, AC: ante cibum, PC: post cibum

Table 1. Baseline characteristics of arsenic and CVD risk factors among baseline-normal study participants classified by ECG status at follow-up

### **3.1 Univariate SNPs association analysis**

Eight functional polymorphisms: C-108T, L55M and Q192R of PON1, A148G and C311S of PON2, M287T of AS3MT, A140D of GSTO1, and N142D of GSTO2 were screened for association with ECG abnormality and Hardy-Weinberg equilibrium (HWE). None reached statistical significance, suggesting no univariate SNP association in the analysis. Genotypic frequencies of M287T showed a significant departure from HWE but because of the limited number of participants carrying the T alleles in this study population, they were excluded from subsequent analysis.


normal study subjects showed ECG deterioration at follow-up. Compared to ECG normal subjects, those with an ECG abnormality had significantly higher arsenic exposure as shown by both years of drinking artesian water and cumulative arsenic exposure index. Age and proportion of cigarette smoking in the ECG abnormal group tended to be higher but did not reach statistical significance. No differences were observed in other cardiovascular risk factors including gender, alcohol consumption, BMI, serum lipids, blood pressure, and

Age (years) 62.0 ± 7.3 64.9 ± 8.7 Male (% ) 29 (36.7) 18 (42.9) Cigarette Smoking (%) 14 (17.7) 13 (31.0) Alcohol consumption (%) 10 (12.7) 6 (14.3) Residency (years) 41.5 ± 12.4 43.3 ± 14.3 Drinking Artesian Water (years) 19.2 ± 9.3 25.1 ± 9.7 Cumulative As exposure (ppm-years) 13.6 ± 8.4 18.0 ± 8.4 BMI (kg/m^2) 24.9 ± 3.13 24.5 ± 3.4 Triglycerides (mg/dl) 113.1 ± 64.7 132.3 ± 106.9 Total Cholesterol (mg/dl) 220.2 ± 45.0 211.7 ± 42.5 HDL (mg/dl) 60.3 ± 18.1 59.2 ± 14.3 LDL (mg/dl) 121.9 ± 48.1 124.6 ± 76.3 Cholesterol /HDL ratio 4.0 ± 1.6 3.8 ± 1.2 Uric acid (mg/dl) 6.1 ± 1.9 5.7 ± 1.8 SBP (mmHg) 124.8 ± 19.3 128.2 ± 17.3 DBP (mmHg) 82.2 ± 11.1 84.2 ± 8.6 AC glucose (mg/dl) 98.6 ± 23.1 104.4 ± 37.3 PC glucose (mg/dl) 127.3 ± 52.2 129.1 ± 78.1

HDL: high density lipoprotein; LDL: low density lipoprotein; CHOL: total cholesterol levels; SBP:

Table 1. Baseline characteristics of arsenic and CVD risk factors among baseline-normal

Eight functional polymorphisms: C-108T, L55M and Q192R of PON1, A148G and C311S of PON2, M287T of AS3MT, A140D of GSTO1, and N142D of GSTO2 were screened for association with ECG abnormality and Hardy-Weinberg equilibrium (HWE). None reached statistical significance, suggesting no univariate SNP association in the analysis. Genotypic frequencies of M287T showed a significant departure from HWE but because of the limited number of participants carrying the T alleles in this study population, they were excluded

systolic blood pressure; DBP: diastolic blood pressure, AC: ante cibum, PC: post cibum

(n=79)

ECG abnormal (n=42)

Variable ECG normal

Data are reported as mean ± SD or counts (%)

**3.1 Univariate SNPs association analysis** 

from subsequent analysis.

study participants classified by ECG status at follow-up

plasma glucose.

Hardy-Weinberg Equilibrium (HWE) test was conducted among all study subjects

Table 2. Association of SNPs and Hardy-Weinberg equilibrium test

Figure 1 shows the related position and linkage disequilibrium (LD) between SNPs in the PON and GSTO gene clusters. Two SNPs within PON2 (C311S and A148G) and GSTO1- A140D and GSTO2-N142 were in high LD but SNPs within PON1 or adjacent SNPs between PON1 and PON2 (C-108T and C311s) had low LD measurements, implying they were not in the same LD block. Q192R, C-108T, C311S and A140D were identified as tag-SNPs.

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

**3.2 Haplotype analysis and association with ECG abnormality** 

Haplotypes ECG normal ECG abnormal OR

R L T 46 (37.1%) 24 (42.9%) 1.00 (reference) Q L C 38 (30.7%) 10 (17.9%) 0.50 (0.22-1.18) R L C 18 (14.5%) 15 (26.8%) 1.60 (0.69-372) Q M C 8 (6.5%) 4 (7.1%) 0.96 (0.26-3.51) Q L T 8 (6.5%) 3 (5.4%) 0.72 (0.17-2.96)

R T 52 (41.9%) 24 (42.9%) 1.00 (reference) Q C 46 (37.1%) 14 (25.0%) 0.66 (0.31-1.42) R C 18 (14.5%) 15 (26.8%) 1.81 (0.78-4.18) Q T 8 (6.5%) 3 (5.4%) 0.81 (0.20-3.33)

S A 89 (71.8%) 40 (70.4%) 1.00 (reference) C G 32 (25.8%) 10 (17.9%) 0.70 (0.31-1.56)

A N 96 (69.5%) 43 (67.2%) 1.00 (reference) D D 28 (20.3%) 10 (15.6%) 0.80 (0.36-1.77) A D 10 (7.3%) 8 (12.5%) 1.79 (0.66-4.84)

R T S 47 (37.9%) 21 (37.5%) 1.00 (reference) Q C S 27 (21.7%) 9 (16.1%) 0.75 (0.30-1.86) R C S 14 (11.3%) 12 (21.4) 1.92 (0.76-4.85) Q C C 19 (15.3%) 5 (8.9%) 0.59 (0.19-1.79) Q T S 8 (6.5%) 3 (5.4%) 0.84 (0.20-3.48)

Table 3. Estimated haplotype frequencies and haplotypes association analysis with ECG

risk toward ECG abnormality.

Q192R L55M C-108T

abnormality

Q192R C-108T

C311S A148G

A140D N142D

Q192R C-108T C311S

Haplotypes with a frequency less than 5% were removed.

Empirical P-value: a Haplotype-specific test, b Haplotype-global test

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 307

Haplotypes of PON1, PON2, GSTO1, GSTO2 and tag-SNPs of the PON gene cluster were constructed, and those whose frequencies were <5% were excluded from association analysis (Table 3). Overall, the effects of these haplotypes on ECG abnormality were not statistically significant after 10,000 permutations; however, the haplotype R-C-S constructed by Q192R, C-108T and C311S had the highest odds, 1.92 (95% CI: 0.76-4.85) times increased

Fig. 1. Linkage disequilibrium (LD) plot of PON1 and GSTO gene clusters in 121 study subjects. The measure of LD (D') among all possible pairs of SNPs is shown graphically. Dark red represents high D' while white represents low D'

### **3.2 Haplotype analysis and association with ECG abnormality**

306 Advances in Electrocardiograms – Clinical Applications

Fig. 1. Linkage disequilibrium (LD) plot of PON1 and GSTO gene clusters in 121 study subjects. The measure of LD (D') among all possible pairs of SNPs is shown graphically.

Dark red represents high D' while white represents low D'

Haplotypes of PON1, PON2, GSTO1, GSTO2 and tag-SNPs of the PON gene cluster were constructed, and those whose frequencies were <5% were excluded from association analysis (Table 3). Overall, the effects of these haplotypes on ECG abnormality were not statistically significant after 10,000 permutations; however, the haplotype R-C-S constructed by Q192R, C-108T and C311S had the highest odds, 1.92 (95% CI: 0.76-4.85) times increased risk toward ECG abnormality.


Haplotypes with a frequency less than 5% were removed.

Empirical P-value: a Haplotype-specific test, b Haplotype-global test

Table 3. Estimated haplotype frequencies and haplotypes association analysis with ECG abnormality

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

ECG

a Permutation analysis adjusted by age, gender and cigarette smoking

ECG

a Permutation analysis adjusted by age, gender and cigarette smoking

Variable Non-exposed

R-C-S CAE ECG

R-C-S DAW ECG

**Urinary arsenic level** 

Σ(iAs + MMA + DMA)

**Urinary arsenic percentage** 

\*P-value < 0.05, \*\* P-value < 0.01

a P-value for ANOVA

Normal

CAE: Cumulative arsenic exposure (ppm-years) R-C-S: Q192R, C-108T and C311S R-C-S haplotype

Normal

DAW: Drinking artesian water (years) RCS: Q192R, C-108T & C311S R-C-S haplotype

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 309

Table 4a. Synergistic effects of Q192R, C-108T, and C311S R-C-S haplotypes carrier and high cumulative arsenic exposure (CAE) (> median of 14.7 ppm-years) on ECG abnormality


Table 4b. Synergistic effects of Q192R, C-108T, and C311S R-C-S haplotypes carrier and more years of drinking artesian water (DAW) (> mean of 21 years) on ECG abnormality

As (III) (μg/g creatinine)\*\* 2.07 (2.98) 4.61 (9.46) 4.73 (6.24) As (V) (μg/g creatinine) 2.78 (3.16) 3.13 (3.66) 2.21 (1.92) iAs (μg/g creatinine)\*\* 4.85 (4.66) 7.73 (11.54) 6.95 (6.74) MMA (μg/g creatinine)\*\* 3.13 (3.79) 4.48 (6.30) 4.21 (3.25) Σ(iAs +MMA) (μg/g creatinine)\*\* 7.98 (6.84) 12.21 (14.66) 11.15 (7.84) DMA (μg/g creatinine)\* 42.87 (34.48) 33.64 (27.34) 37.37 (24.32)

(μg/g creatinine) 50.85 (37.86) 45.87 (33.60) 48.58 (28.53) PMI (MMA/iAs) 0.87 (1.02) 0.84 (0.75) 0.85 (0.60) SMI (DMA/MMA)\*\* 23.68 (22.80) 15.44 (28.38) 15.94 (20.86)

As (III) %\*\* 4.59 (5.18) 10.08 (13.22) 9.18 (9.08) As (V) %\*\* 6.23 (5.62) 8.27 (7.76) 5.81 (7.58) iAs %\*\* 10.82 (8.37) 18.35 (16.50) 14.99 (10.73) MMA %\*\* 7.45 (8.59) 10.90 (9.05) 10.12 (7.57) DMA %\*\* 81.73 (13.84) 70.75 (20.23) 74.89 (13.99)

Table 5. Correlation of cumulative arsenic exposure and urinary metabolism capacity

(Chiali) (N=302)


Abnormal OR (95% CI) OR (95% CI)a Empirical

Abnormal OR (95% CI) OR (95% CI)a Empirical

≤ 14.7 (Putai) (N=191)

P-value a

P-value a

> 14.7 (Putai) (N=103)

The relative odds of lipid profiles for PON-haplotype R-C-S carrier compared with noncarriers are shown in Figure 2. The R-C-S haplotype was positively correlated with higher serum HDL-cholesterol, LDL-cholesterol, and triglyceride levels without statistical significance, but was significantly associated with increased total cholesterol levels (OR=2.91, 95% CI: 1.13-7.70).

Fig. 2. Odds ratios of lipid profiles for Q192R, C-108T and C311S R-C-S haplotype carrier among study subjects (N=121)

### **3.3 Synergistic association of PON haplotype and arsenic on ECG abnormality**

The synergistic associations between PON haplotype and arsenic exposure are summarized in Table 4. The PON R-C-S haplotype carrier with higher cumulative arsenic exposure (greater than the median value of 14.7 ppm-years) showed a >14.66 (95% CI: 1.83-117.64) increased risk for ECG abnormality compared to non-RCS haplotype carriers with low cumulative arsenic exposure (<14.7 ppm-years) (Table 4a). The PON R-C-S haplotype carrier with more years of drinking artesian water (greater than the median of 21 years) had a 10.83-fold (95% CI: 1.83-64.03) increased risk (Table 4b). These associations were even stronger after adjusting for age, gender, and cigarette smoking, when the odds increased to 19.19 (95% CI: 1.86-197.76) and 21.09 (95% CI: 2.77-160.35) for cumulative exposure index and drinking years, respectively.

Table 5 showed the correlation between cumulative arsenic exposure and urinary arsenic species from arsenic endemic and non-endemic areas in southwestern Taiwan. Subjects with higher cumulative arsenic exposure had significantly higher levels of As (III), iAs, MMA, Summation of iAs and MMA. However, DMA levels and SMI were significantly lower among subjects with high arsenic exposure. Similar pattern was observed when urinary arsenic was analyzed in percentage. Subjects with higher cumulative arsenic exposure had higher percentages of As(III), iAs and MMA in urinary and lower DMA percentage.


a Permutation analysis adjusted by age, gender and cigarette smoking CAE: Cumulative arsenic exposure (ppm-years) R-C-S: Q192R, C-108T and C311S R-C-S haplotype

Table 4a. Synergistic effects of Q192R, C-108T, and C311S R-C-S haplotypes carrier and high cumulative arsenic exposure (CAE) (> median of 14.7 ppm-years) on ECG abnormality


a Permutation analysis adjusted by age, gender and cigarette smoking DAW: Drinking artesian water (years)

RCS: Q192R, C-108T & C311S R-C-S haplotype

308 Advances in Electrocardiograms – Clinical Applications

The relative odds of lipid profiles for PON-haplotype R-C-S carrier compared with noncarriers are shown in Figure 2. The R-C-S haplotype was positively correlated with higher serum HDL-cholesterol, LDL-cholesterol, and triglyceride levels without statistical significance, but was significantly associated with increased total cholesterol levels

Fig. 2. Odds ratios of lipid profiles for Q192R, C-108T and C311S R-C-S haplotype carrier

The synergistic associations between PON haplotype and arsenic exposure are summarized in Table 4. The PON R-C-S haplotype carrier with higher cumulative arsenic exposure (greater than the median value of 14.7 ppm-years) showed a >14.66 (95% CI: 1.83-117.64) increased risk for ECG abnormality compared to non-RCS haplotype carriers with low cumulative arsenic exposure (<14.7 ppm-years) (Table 4a). The PON R-C-S haplotype carrier with more years of drinking artesian water (greater than the median of 21 years) had a 10.83-fold (95% CI: 1.83-64.03) increased risk (Table 4b). These associations were even stronger after adjusting for age, gender, and cigarette smoking, when the odds increased to 19.19 (95% CI: 1.86-197.76) and 21.09 (95% CI: 2.77-160.35) for cumulative exposure index

Table 5 showed the correlation between cumulative arsenic exposure and urinary arsenic species from arsenic endemic and non-endemic areas in southwestern Taiwan. Subjects with higher cumulative arsenic exposure had significantly higher levels of As (III), iAs, MMA, Summation of iAs and MMA. However, DMA levels and SMI were significantly lower among subjects with high arsenic exposure. Similar pattern was observed when urinary arsenic was analyzed in percentage. Subjects with higher cumulative arsenic exposure had higher percentages of As(III), iAs and MMA in urinary and lower DMA

**3.3 Synergistic association of PON haplotype and arsenic on ECG abnormality** 

(OR=2.91, 95% CI: 1.13-7.70).

among study subjects (N=121)

and drinking years, respectively.

percentage.

Table 4b. Synergistic effects of Q192R, C-108T, and C311S R-C-S haplotypes carrier and more years of drinking artesian water (DAW) (> mean of 21 years) on ECG abnormality


a P-value for ANOVA

\*P-value < 0.05, \*\* P-value < 0.01

Table 5. Correlation of cumulative arsenic exposure and urinary metabolism capacity

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

and smaller JT duration, JT index and RaVL (data not shown).

understand the underlying mechanism regarding arsenic-induced hazards.

underlying mechanisms.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 311

The major strength of this study was to apply a standardized Minnesota coding classification of ECG reading that ensures good quality assurance and control. Furthermore, detailed parameters regarding ECG abnormalities allowed us to evaluate the minor changes due to arsenic toxicity. In current analyses, higher duration of arsenic water consumption was associated with ECG abnormality, myocardial infarction or ischemia, atrial enlargement or ventricular hypertrophy in a dose-response relationship. Besides, it was also positively correlated to arrhythmia and prolonged ventricular repolarization without reached the statistical significance. Moreover, higher levels of cumulative arsenic exposure were also associated with ECG parameters including higher PR duration, QRS duration and QRS axis

There were still some limitations for this study. First, results from this study were based on a population with history of arsenic exposure. Insignificant association between urinary methylation capabilities might due to attrition of high-level arsenic exposed subjects, competing risks for CVD mortality was not considered in current analysis. Another limitation of the present study was that the measurement of urinary metabolism species and physical evaluation were conducted at a cross-sectional design. The causal-relationship between urinary species of arsenic and ECG abnormality could not be inferred given current evidence. However, the previous exposure status was significantly correlated with current urinary arsenic species implied it was more efficient among subjects after cessation of longterm exposure to high levels of arsenic. These results may have implications for arsenic mediation strategies in areas currently exposed to potentially harmful levels of arsenic in drinking water. Furthermore, CVD usually took years for disease development. Correlation may be biased due to unmeasured factors during a relative short follow-up period. Longer duration of follow-up with serial changes for ECG abnormality would help better

The major significance by this study was the assessment of arsenic risk from subjects without exposure of inorganic arsenic to moderate and relatively high levels of cumulative arsenic exposure. Causal inference can be strengthened by the dose-response relationship by the stronger effect in a susceptible subgroup of the population. Besides, this study demonstrated significant gene-gene and gene-environment interactions by showing PON1 gene cluster including polymorphisms of PON1: Q192R, PON1: C-108T, and PON2: C311S and latent effect of arsenic exposure on incidence of ECG abnormality. Besides, PON2: C311S was independently associated with LDH elevation and further predicted future CVD mortality independent to other conventional risk factors including age, gender, cigarette smoking, hypertension and diabetes mellitus. After cessation of arsenic-contaminated water consumption for decades, biomarkers for CVD mortality and morbidity was still associated with reduced risks for arsenic and attributable to underlying genetic predisposition. Such data may also help risk assessment in the population and provide knowledge about the

HDL has been shown to prevent atherogenesis in vivo and in vitro through anti-oxidative and anti-inflammatory activities35. The major part of anti-atherogenic properties associated with HDL is explained by the activity of Paraoxonase 168. Both PON1 and PON2 belong to the protein family of Paraoxonase 1 that includes PON3 which has been suggested that involved in CVD69. PON1 directly form part of HDL particles whereas PON2 found in endothelial cells, smooth muscle cells and macrophages that possesses antioxidant properties similar to PON1 by delays cellular oxidative stress and prevents apoptosis in

Distribution between levels of cumulative arsenic exposure and ECG abnormality was summarized in Table 6. Significant dose-response relationships were observed between higher levels of cumulative arsenic exposure and ECG reading regarding abnormalities, myocardial infarction or ischemia disease, and also atrial enlargement and ventricular hypertrophy. Increased cumulative arsenic exposure was also correlated with a higher proportion of abnormality in prolonged ventricular repolarization however not reach statistical significance.


Table 6. Distribution of ECG reading by cumulative arsenic exposure

### **4. Discussion**

Various ECG abnormalities have been observed among cases of acute arsenic poisoning and in acute promyelocytic leukemia patients treated with arsenic trioxide. Individuals exposed to excess arsenic through drinking water showed some of the ECG abnormalities51. Several epidemiologic studies showed that QT prolongation and increased CVD mortality among high levels of arsenic-exposed subjects. However, the results might not be applicable in subjects with low to moderate arsenic. Our data replicated this association in an arseniasisendemic area and a well-matched control area which no previous history of water contamination. We highlighted the correlation between previous chronic arsenic exposure and ECG abnormalities after cessation of arsenic-contaminated water consumption for decades.

Distribution between levels of cumulative arsenic exposure and ECG abnormality was summarized in Table 6. Significant dose-response relationships were observed between higher levels of cumulative arsenic exposure and ECG reading regarding abnormalities, myocardial infarction or ischemia disease, and also atrial enlargement and ventricular hypertrophy. Increased cumulative arsenic exposure was also correlated with a higher proportion of abnormality in prolonged ventricular repolarization however not reach

 0: Normal 155 (51.3) 96 (50.3) 40 (38.8) 1: Minor abnormal 122 (40.3) 70 (36.7) 40 (38.8) 2: Major abnormal 25 (8.3) 25 (13.0) 23 (22.4)

 0: Normal 216 (71.5) 136 (71.2) 68 (66.0) 1: Minor abnormal 72 (23.8) 37 (19.4) 20 (19.4) 2: Major abnormal 14 (4.6) 18 (9.4) 15 (14.6)

 0: Normal 261 (86.4) 174 (91.1) 86 (83.5) 1: Abnormal 41 (13.6) 17 (8.9) 17 (16.5)

 0: Normal 265 (87.8) 170 (89.0) 97 (94.2) 1: Abnormal 37 (12.3) 21 (11.0) 6 (5.8)

 0: Normal 286 (94.7) 167 (87.4) 85 (82.5) 1: Abnormal 16 (5.3) 24 (12.6) 18 (17.5)

 0: Normal 299 (99.0) 187 (98.0) 99 (96.1) 1: Abnormal 3 (1.0) 4 (2.0) 4 (3.9)

Various ECG abnormalities have been observed among cases of acute arsenic poisoning and in acute promyelocytic leukemia patients treated with arsenic trioxide. Individuals exposed to excess arsenic through drinking water showed some of the ECG abnormalities51. Several epidemiologic studies showed that QT prolongation and increased CVD mortality among high levels of arsenic-exposed subjects. However, the results might not be applicable in subjects with low to moderate arsenic. Our data replicated this association in an arseniasisendemic area and a well-matched control area which no previous history of water contamination. We highlighted the correlation between previous chronic arsenic exposure and ECG abnormalities after cessation of arsenic-contaminated water consumption for

(Chiali) (N=302)

≤ 14.7 (Putai) (N=191)

> 14.7 (Putai) (N=103)

Variable Non-exposed

Atrial enlargement/Ventricular hypertrophy (Hypertrophy)\*\*

Table 6. Distribution of ECG reading by cumulative arsenic exposure

Myocardial Infarction or Ischemia (MC\_MI)\*

Conduction defect (BBB; Bundle Branch Block)

Prolonged ventricular repolarization (Long\_QT)

statistical significance.

ECG reading (ECG group)\*\*

Arrhythmia (Arrhythmia)

a P-value for trend test

**4. Discussion** 

decades.

\*P-value < 0.05, \*\* P-value < 0.01

The major strength of this study was to apply a standardized Minnesota coding classification of ECG reading that ensures good quality assurance and control. Furthermore, detailed parameters regarding ECG abnormalities allowed us to evaluate the minor changes due to arsenic toxicity. In current analyses, higher duration of arsenic water consumption was associated with ECG abnormality, myocardial infarction or ischemia, atrial enlargement or ventricular hypertrophy in a dose-response relationship. Besides, it was also positively correlated to arrhythmia and prolonged ventricular repolarization without reached the statistical significance. Moreover, higher levels of cumulative arsenic exposure were also associated with ECG parameters including higher PR duration, QRS duration and QRS axis and smaller JT duration, JT index and RaVL (data not shown).

There were still some limitations for this study. First, results from this study were based on a population with history of arsenic exposure. Insignificant association between urinary methylation capabilities might due to attrition of high-level arsenic exposed subjects, competing risks for CVD mortality was not considered in current analysis. Another limitation of the present study was that the measurement of urinary metabolism species and physical evaluation were conducted at a cross-sectional design. The causal-relationship between urinary species of arsenic and ECG abnormality could not be inferred given current evidence. However, the previous exposure status was significantly correlated with current urinary arsenic species implied it was more efficient among subjects after cessation of longterm exposure to high levels of arsenic. These results may have implications for arsenic mediation strategies in areas currently exposed to potentially harmful levels of arsenic in drinking water. Furthermore, CVD usually took years for disease development. Correlation may be biased due to unmeasured factors during a relative short follow-up period. Longer duration of follow-up with serial changes for ECG abnormality would help better understand the underlying mechanism regarding arsenic-induced hazards.

The major significance by this study was the assessment of arsenic risk from subjects without exposure of inorganic arsenic to moderate and relatively high levels of cumulative arsenic exposure. Causal inference can be strengthened by the dose-response relationship by the stronger effect in a susceptible subgroup of the population. Besides, this study demonstrated significant gene-gene and gene-environment interactions by showing PON1 gene cluster including polymorphisms of PON1: Q192R, PON1: C-108T, and PON2: C311S and latent effect of arsenic exposure on incidence of ECG abnormality. Besides, PON2: C311S was independently associated with LDH elevation and further predicted future CVD mortality independent to other conventional risk factors including age, gender, cigarette smoking, hypertension and diabetes mellitus. After cessation of arsenic-contaminated water consumption for decades, biomarkers for CVD mortality and morbidity was still associated with reduced risks for arsenic and attributable to underlying genetic predisposition. Such data may also help risk assessment in the population and provide knowledge about the underlying mechanisms.

HDL has been shown to prevent atherogenesis in vivo and in vitro through anti-oxidative and anti-inflammatory activities35. The major part of anti-atherogenic properties associated with HDL is explained by the activity of Paraoxonase 168. Both PON1 and PON2 belong to the protein family of Paraoxonase 1 that includes PON3 which has been suggested that involved in CVD69. PON1 directly form part of HDL particles whereas PON2 found in endothelial cells, smooth muscle cells and macrophages that possesses antioxidant properties similar to PON1 by delays cellular oxidative stress and prevents apoptosis in

EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

1989;40(6):547-558.

Oct 1996;5(10):849-852.

Suppl):349-357.

2007;25(1):1-22.

*statistics*2005.

1993.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 313

[3] Tseng WP. Blackfoot disease in Taiwan: a 30-year follow-up study. *Angiology.* Jun

[5] Watanabe C, Inaoka T, Kadono T, et al. Males in rural Bangladeshi communities are

[6] Vahter M. Genetic polymorphism in the biotransformation of inorganic arsenic and its

[7] Loffredo CA, Aposhian HV, Cebrian ME, Yamauchi H, Silbergeld EK. Variability in

[8] Chung JS, Kalman DA, Moore LE, et al. Family correlations of arsenic methylation

[9] Karagas MR, Morris JS, Weiss JE, Spate V, Baskett C, Greenberg ER. Toenail samples as

[10] Cohen SM, Arnold LL, Eldan M, Lewis AS, Beck BD. Methylated arsenicals: the

[11] Tseng CH. Metabolism of inorganic arsenic and non-cancerous health hazards

[12] Tseng CH. Arsenic methylation, urinary arsenic metabolites and human diseases:

[13] Maton A. *Human Biology and Health* Englewood Cliffs, New Jersey: Prentice Hall; 1993. [14] McGill HC, Jr., McMahan CA, Zieske AW, et al. Associations of coronary heart disease

[15] CDC, US. Chronic Disease Overview: Center for Disease control and Prevention 1999. [16] Department of Health, Executive Yuan, Taiwan. *Health and Vital statistics-vital* 

[17] Rainwater DL, McMahan CA, Malcom GT, et al. Lipid and apolipoprotein predictors of

[18] IARC. Arsenic and arsenic compounds. In: Cancer IAfRo, ed*.* Vol 23: IARC Monogr

[19] EPA US. Inorganic Arsenic. Washington, DC, U.S.: Integrated Risk Information System;

[20] Hughes MF. Biomarkers of exposure: a case study with inorganic arsenic. *Environ Health* 

more susceptible to chronic arsenic poisoning than females: analyses based on

patterns in children and parents exposed to high concentrations of arsenic in

an indicator of drinking water arsenic exposure. *Cancer Epidemiol Biomarkers Prev.* 

implications of metabolism and carcinogenicity studies in rodents to human risk

associated with chronic exposure in humans. *J Environ Biol.* Apr 2007;28(2

current perspective. *J Environ Sci Health C Environ Carcinog Ecotoxicol Rev.* Jan-Mar

risk factors with the intermediate lesion of atherosclerosis in youth. The Pathobiological Determinants of Atherosclerosis in Youth (PDAY) Research Group.

atherosclerosis in youth: apolipoprotein concentrations do not materially improve prediction of arterial lesions in PDAY subjects. The PDAY Research Group.

[4] NRC. Arsenic in drinking water. Washington, DC: National Academy Press; 1999.

urinary arsenic. *Environ Health Perspect.* Dec 2001;109(12):1265-1270.

human metabolism of arsenic. *Environ Res.* Jun 2003;92(2):85-91.

drinking water. *Environ Health Perspect.* Jul 2002;110(7):729-733.

role in toxicity. *Toxicol Lett.* Mar 15 2000;112-113:209-217.

assessment. *Crit Rev Toxicol.* Feb 2006;36(2):99-133.

*Arterioscler Thromb Vasc Biol.* Aug 2000;20(8):1998-2004.

*Arterioscler Thromb Vasc Biol.* Mar 1999;19(3):753-761.

Eval Carcinog Risks Hum; 1987:100-103.

*Perspect.* Nov 2006;114(11):1790-1796.

vascular endothelial cells70,71. Regarding three common polymorphisms in coding region of the human PON1 gene, the frequencies of Q192R, L55M, and C-108T for Taiwanese population were similar to those reported in the literature for the Chinese population 72,73. Besides, we confirmed that paraoxonase, diazoxonase and arylesterase activities were directly influenced by the Q192R and C-108T polymorphisms. Previous studies had also shown that 311 C allele in PON2 was associated with increased risks of coronary artery disease, MI and also diabetic nephropathy74-76. Our data confirmed the significance of PON2: C311S polymorphism during pathogenesis of CVD among chronic arsenic exposed subject. This finding could help to identify subjects at higher risk of cardiovascular damage for arsenic toxicity.

However, some other factors that might have influenced the arsenic methylation profiles were not considered. Nutritional status and dietary intake may also be uncontrollable factors. Besides, we could not obtain the accurate data on allele distribution three polymorphisms: PON1: C-108T, GSTO1:A140D, and AS3MT: L55M in our samples. We still could not rule out the intra- and inter-individual variability to arsenic methylation and also their impacts on pathogenesis of CVD morbidity and mortality. Besides, the arsenic levels in rice growing in the arsenic contaminated area or inorganic arsenic from fish intake may be elevated. This might potentially increase arsenic exposure in the endemic area77-80. Future study of exposure assessment is needed. In addition, arsenic exposure has also been shown to alter the methylation level of both global DNA and certain genes in studies that analyzed a limited number of epigenetic endpoints81. Therefore, it is necessary to enlarge sample size for the evaluation of genetic association and ECG abnormality and other confounders that may be directly related to arsenic risks.

Genome-wide association studies (GWAS) have been applied in the search for suscepitibility genes to coronary artery disease, myocardial infarction and heart failure82-84. However, none of the candidate regions and genes showed a powerful association with CVD at genomewide significance and the molecular and biological mechanisms remains unclear. Atherosclerosis is a multifactorial disease that may lead to myocardial infarction or heart failure. A conservative estimate would be that at least 100 genes have the potential to affect the modifying factors including atherosclerosis, myocardial infarction and congestive heart failure with each having a genetic contribution of as much as 2% to the phenotype84. Since CVD usually took years to develop, correlation may be biased due to unmeasured factors during a relative short follow-up period or relatively underestimated by competing risk. Our studies with longer duration of follow-up with serial changes for ECG abnormality did help better understand the underlying mechanism and duration regarding arsenic-induced hazards. These findings emphasize the importance of long term arsenic effect, along with the necessity of intensive follow-up for preclinical or subclinical phenotypes such as ECG abnormality for preventing excessive CVD mortality.

### **5. References**


vascular endothelial cells70,71. Regarding three common polymorphisms in coding region of the human PON1 gene, the frequencies of Q192R, L55M, and C-108T for Taiwanese population were similar to those reported in the literature for the Chinese population 72,73. Besides, we confirmed that paraoxonase, diazoxonase and arylesterase activities were directly influenced by the Q192R and C-108T polymorphisms. Previous studies had also shown that 311 C allele in PON2 was associated with increased risks of coronary artery disease, MI and also diabetic nephropathy74-76. Our data confirmed the significance of PON2: C311S polymorphism during pathogenesis of CVD among chronic arsenic exposed subject. This finding could help to identify subjects at higher risk of cardiovascular damage

However, some other factors that might have influenced the arsenic methylation profiles were not considered. Nutritional status and dietary intake may also be uncontrollable factors. Besides, we could not obtain the accurate data on allele distribution three polymorphisms: PON1: C-108T, GSTO1:A140D, and AS3MT: L55M in our samples. We still could not rule out the intra- and inter-individual variability to arsenic methylation and also their impacts on pathogenesis of CVD morbidity and mortality. Besides, the arsenic levels in rice growing in the arsenic contaminated area or inorganic arsenic from fish intake may be elevated. This might potentially increase arsenic exposure in the endemic area77-80. Future study of exposure assessment is needed. In addition, arsenic exposure has also been shown to alter the methylation level of both global DNA and certain genes in studies that analyzed a limited number of epigenetic endpoints81. Therefore, it is necessary to enlarge sample size for the evaluation of genetic association and ECG abnormality and other confounders that

Genome-wide association studies (GWAS) have been applied in the search for suscepitibility genes to coronary artery disease, myocardial infarction and heart failure82-84. However, none of the candidate regions and genes showed a powerful association with CVD at genomewide significance and the molecular and biological mechanisms remains unclear. Atherosclerosis is a multifactorial disease that may lead to myocardial infarction or heart failure. A conservative estimate would be that at least 100 genes have the potential to affect the modifying factors including atherosclerosis, myocardial infarction and congestive heart failure with each having a genetic contribution of as much as 2% to the phenotype84. Since CVD usually took years to develop, correlation may be biased due to unmeasured factors during a relative short follow-up period or relatively underestimated by competing risk. Our studies with longer duration of follow-up with serial changes for ECG abnormality did help better understand the underlying mechanism and duration regarding arsenic-induced hazards. These findings emphasize the importance of long term arsenic effect, along with the necessity of intensive follow-up for preclinical or subclinical phenotypes such as ECG

[1] Prineas RJ, Crow RS, Blackburn H. The Minnesota Code Manual of Electrocardiographic

[2] Kapaj S, Peterson H, Liber K, Bhattacharya P. Human health effects from chronic arsenic

poisoning--a review. *J Environ Sci Health A Tox Hazard Subst Environ Eng.* 

Findings. In: John Wright PSB Inc.,1st edition. Boston: MA; 1982.

for arsenic toxicity.

**5. References** 

may be directly related to arsenic risks.

abnormality for preventing excessive CVD mortality.

2006;41(10):2399-2428.


EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

*Pharmacol.* Sep 1 2009;239(2):178-183.

1996;169(2):256-268.

2004;198(3):444-449.

Apr 1996;27(5):1007-1019.

1984;311(18):1144-1147.

*Genet.* Mar 1972;9(1):8-12.

Jun 2006;24(2):221-227.

*Circulation.* Feb 10 1998;97(5):467-472.

*Am Heart J.* Feb 2006;151(2):508-513.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 315

[37] Watson AD, Berliner JA, Hama SY, et al. Protective effect of high density lipoprotein

[38] Shih DM, Gu L, Xia YR, et al. Mice lacking serum paraoxonase are susceptible to

[39] Liao YT, Li WF, Chen CJ, et al. Synergistic effect of polymorphisms of paraoxonase gene

[40] Ghosh P, Banerjee M, Giri AK, Ray K. Toxicogenomics of arsenic: classical ideas and

[41] Barchowsky A, Klei LR, Dudek EJ, Swartz HM, James PE. Stimulation of reactive

[44] Simeonova PP, Luster MI. Arsenic and atherosclerosis. *Toxicol Appl Pharmacol.* Aug 1

[45] Kuller LH, Shemanski L, Psaty BM, et al. Subclinical disease as an independent risk factor for cardiovascular disease. *Circulation.* Aug 15 1995;92(4):720-726. [46] Califf RM, Armstrong PW, Carver JR, D'Agostino RB, Strauss WE. 27th Bethesda

[47] Jeger RV, Probst C, Arsenic R, et al. Long-term prognostic value of the preoperative 12-

[48] Kannel WB, Abbott RD. Incidence and prognosis of unrecognized myocardial

[49] Emery AE. Abnormalities of the electrocardiogram in hereditary myopathies. *J Med* 

[50] Kuller L, Borhani N, Furberg C, et al. Prevalence of subclinical atherosclerosis and

[51] Ahmad SA, Khatun F, Sayed MH, et al. Electrocardiographic abnormalities among

[52] de Bruyne MC, Hoes AW, Kors JA, Hofman A, van Bemmel JH, Grobbee DE. QTc

Health Study. *Am J Epidemiol.* Jun 15 1994;139(12):1164-1179.

low levels of arsenite. *Free Radic Biol Med.* Dec 1999;27(11-12):1405-1412. [42] Chen YC, Lin-Shiau SY, Lin JK. Involvement of reactive oxygen species and caspase 3 activation in arsenite-induced apoptosis. *J Cell Physiol.* Nov 1998;177(2):324-333. [43] Wang TS, Kuo CF, Jan KY, Huang H. Arsenite induces apoptosis in Chinese hamster

low density lipoprotein. *J Clin Invest.* Dec 1995;96(6):2882-2891.

recent advances. *Mutat Res.* Sep-Oct 2008;659(3):293-301.

associated paraoxonase. Inhibition of the biological activity of minimally oxidized

organophosphate toxicity and atherosclerosis. *Nature.* Jul 16 1998;394(6690):284-287.

cluster and arsenic exposure on electrocardiogram abnormality. *Toxicol Appl* 

oxygen, but not reactive nitrogen species, in vascular endothelial cells exposed to

ovary cells by generation of reactive oxygen species. *J Cell Physiol.* Nov

Conference: matching the intensity of risk factor management with the hazard for coronary disease events. Task Force 5. Stratification of patients into high, medium and low risk subgroups for purposes of risk factor management. *J Am Coll Cardiol.* 

lead electrocardiogram before major noncardiac surgery in coronary artery disease.

infarction. An update on the Framingham study. *N Engl J Med.* Nov 1

cardiovascular disease and association with risk factors in the Cardiovascular

arsenic-exposed persons through groundwater in Bangladesh. *J Health Popul Nutr.* 

dispersion predicts cardiac mortality in the elderly: the Rotterdam Study.


[21] Wang CH, Hsiao CK, Chen CL, et al. A review of the epidemiologic literature on the

[22] Ch'i IC, Blackwell RQ. A controlled retrospective study of Blackfoot disease, an endemic peripheral gangrene disease in Taiwan. *Am J Epidemiol.* Jul 1968;88(1):7-24. [23] Wu HY, Chen KP, Tseng WP, Hsu CL. Epidemiologic studies on blackfoot disease:

[24] Grantham DA, Jones FJ. Arsenic contamination of water wells in Nova Scotia. *Journal of* 

[25] Rahman MM, Chowdhury UK, Mukherjee SC, et al. Chronic arsenic toxicity in

[26] Smith AH, Lopipero PA, Bates MN, Steinmaus CM. Public health. Arsenic

[27] Navas-Acien A, Sharrett AR, Silbergeld EK, et al. Arsenic exposure and cardiovascular

[28] Hughes MF. Arsenic toxicity and potential mechanisms of action. *Toxicol Lett.* Jul 7

[29] Kitchin KT. Recent advances in arsenic carcinogenesis: modes of action, animal model

[30] Rossman TG. Mechanism of arsenic carcinogenesis: an integrated approach. *Mutat Res.* 

[31] Vahter M, Concha G, Nermell B, Nilsson R, Dulout F, Natarajan AT. A unique

[32] Lin S, Shi Q, Nix FB, et al. A novel S-adenosyl-L-methionine:arsenic(III)

[33] Townsend D, Tew K. Cancer drugs, genetic variation and the glutathione-S-transferase

[34] Tanaka-Kagawa T, Jinno H, Hasegawa T, et al. Functional characterization of two

[35] Aviram M, Rosenblat M, Bisgaier CL, Newton RS, Primo-Parmo SL, La Du BN.

[36] Mackness MI, Arrol S, Durrington PN. Paraoxonase prevents accumulation of lipoperoxides in low-density lipoprotein. *FEBS Lett.* Jul 29 1991;286(1-2):152-154.

gene family. *Am J Pharmacogenomics.* 2003;3(3):157-172.

*the American Water Works Association.* 1977;69:653-657.

*Pharmacol.* Aug 1 2007;222(3):315-326.

1961;7:1-18.

2146.

*Toxicol.* 2001;39(7):683-700.

2005;162(11):1037-1049.

2002;133(1):1-16.

2001;172(3):249-261.

1995;293(4):455-462.

1998;101(8):1581-1590.

10803.

Dec 10 2003;533(1-2):37-65.

*Commun.* Feb 7 2003;301(2):516-520.

role of environmental arsenic exposure and cardiovascular diseases. *Toxicol Appl* 

prevalence and incidence of the disease by age, sex, year, occupation, and geographic distribution. *Memoirs of College of Medicine (National Taiwan University).* 

Bangladesh and West Bengal, India--a review and commentary. *J Toxicol Clin* 

epidemiology and drinking water standards. *Science.* Jun 21 2002;296(5576):2145-

disease: a systematic review of the epidemiologic evidence. *Am J Epidemiol.* Dec 1

systems, and methylated arsenic metabolites. *Toxicol Appl Pharmacol.* May 1

metabolism of inorganic arsenic in native Andean women. *Eur J Pharmacol.* Dec 7

methyltransferase from rat liver cytosol. *J Biol Chem.* Mar 29 2002;277(13):10795-

variant human GSTO 1-1s (Ala140Asp and Thr217Asn). *Biochem Biophys Res* 

Paraoxonase inhibits high-density lipoprotein oxidation and preserves its functions. A possible peroxidative role for paraoxonase. *J Clin Invest.* Apr 15


EElectrocardiogram (ECG) Abnormality Among Residents in Arseniasis-

*Atherosclerosis.* Dec 1993;104(1-2):129-135.

2001;276(48):44444-44449.

2008;11(5):544-549.

Jan 2001;44(1):104-107.

15 2003;37(2):229-234.

*Perspect.* Aug 2006;114(8):1233-1236.

*J Clin Invest.* Jan 2004;34(1):14-20.

*Plant Sci.* Sep 2004;9(9):415-417.

Apr 2008;71(8):1522-1530.

*Perspect.* Jan 2011;119(1):11-19.

2007;8 Suppl 1:S5.

*Arterioscler Thromb Vasc Biol.* Feb 1 2003;23(2):328-334.

Endemic and Non-Endemic Areas of Southwestern Taiwan – A Study of Gene-Gene and… 317

[68] Mackness MI, Arrol S, Abbott C, Durrington PN. Protection of low-density lipoprotein

[69] Wheeler JG, Keavney BD, Watkins H, Collins R, Danesh J. Four paraoxonase gene

[70] Horke S, Witte I, Wilgenbus P, Kruger M, Strand D, Forstermann U. Paraoxonase-2

[72] Li WF, Pan MH, Chung MC, Ho CK, Chuang HY. Lead exposure is associated with

[73] Wang X, Fan Z, Huang J, et al. Extensive association analysis between polymorphisms

[74] Jalilian A, Javadi E, Akrami M, et al. Association of cys 311 ser polymorphism of

[75] Martinelli N, Girelli D, Olivieri O, et al. Interaction between smoking and PON2

[76] Pinizzotto M, Castillo E, Fiaux M, Temler E, Gaillard RC, Ruiz J. Paraoxonase2

[77] Meharg AA, Rahman MM. Arsenic contamination of Bangladesh paddy field soils:

[78] Meharg AA. Arsenic in rice--understanding a new disaster for South-East Asia. *Trends* 

[79] Ruiz-Chancho MJ, Lopez-Sanchez JF, Schmeisser E, Goessler W, Francesconi KA, Rubio

[80] Peshut PJ, Morrison RJ, Brooks BA. Arsenic speciation in marine fish and shellfish from

[81] Ren X, McHale CM, Skibola CF, Smith AH, Smith MT, Zhang L. An emerging role for

[82] Larson MG, Atwood LD, Benjamin EJ, et al. Framingham Heart Study 100K project:

American Samoa. *Chemosphere.* Mar 2008;71(3):484-492.

analysis of 43 studies. *Lancet.* Feb 28 2004;363(9410):689-695.

against oxidative modification by high-density lipoprotein associated paraoxonase.

polymorphisms in 11212 cases of coronary heart disease and 12786 controls: meta-

reduces oxidative stress in vascular cells and decreases endoplasmic reticulum stress-induced caspase activation. *Circulation.* Apr 17 2007;115(15):2055-2064. [71] Ng CJ, Wadleigh DJ, Gangopadhyay A, et al. Paraoxonase-2 is a ubiquitously expressed

protein with antioxidant properties and is capable of preventing cell-mediated oxidative modification of low density lipoprotein. *J Biol Chem.* Nov 30

decreased serum paraoxonase 1 (PON1) activity and genotypes. *Environ Health* 

of PON gene cluster with coronary heart disease in Chinese Han population.

paraoxonase-2 gene with the risk of coronary artery disease. *Arch Iran Med.* Sep

Ser311Cys polymorphism as a determinant of the risk of myocardial infarction. *Eur* 

polymorphisms are associated with nephropathy in Type II diabetes. *Diabetologia.* 

implications for rice contribution to arsenic consumption. *Environ Sci Technol.* Jan

R. Arsenic speciation in plants growing in arsenic-contaminated sites. *Chemosphere.* 

epigenetic dysregulation in arsenic toxicity and carcinogenesis. *Environ Health* 

genome-wide associations for cardiovascular disease outcomes. *BMC Med Genet.* 


[53] Zabel M, Klingenheben T, Franz MR, Hohnloser SH. Assessment of QT dispersion for

[54] Wang CH, Chen CL, Hsiao CK, et al. Increased risk of QT prolongation associated with

[55] Wang CH, Chen CL, Hsiao CK, et al. Arsenic-induced QT dispersion is associated with

[56] Chen CJ, Chuang YC, Lin TM, Wu HY. Malignant neoplasms among residents of a

[57] Chen CJ, Hsueh YM, Lai MS, et al. Increased prevalence of hypertension and long-term

[58] Tseng CH, Chong CK, Tseng CP, et al. Long-term arsenic exposure and ischemic heart

[59] Chen KP, Wu TC, Wu TC. Epidemiologic studies on blackfoot disease in Taiwan. 3.

[61] Blackwell RQ. Estimation total arsenic ingested by residents in the endemic blackfoot

[62] Chen WY, Chen KP. Study on arsenic artesian well water and blackfoot disease

[63] Lo MC, Hsen YC, Lin BK. *Arsenic content of underground water in Taiwan: Second* 

[64] Hsueh YM, Lin P, Chen HW, et al. Genetic polymorphisms of oxidative and antioxidant

[65] MC L, YC H, BK L. *Arsenic content of underground water in Taiwan: Second report.*

[66] World Medical Association. Declaration of Helsinki: Ethical Principles for Medical Research Involving Human Subjects. *Bulletin of Medical Ethics* Vol 1622000:8-11. [67] Potter DM. A permutation test for inference in logistic regression with small- and

moderate-sized data sets. *Stat Med.* Mar 15 2005;24(5):693-708.

*Poisoning* Institute of Pathology, National Taiwan University; 1964.

Taiwan. *Toxicol Appl Pharmacol.* Sep 15 2009;239(3):320-324.

cancers. *Cancer Res.* Nov 1985;45(11 Pt 2):5895-5899.

arsenic exposure. *Hypertension.* Jan 1995;25(1):53-60.

area. *J Formosan Med. Assoc.* 1961;60:1143-1144.

2550.

Mar 2010;10(1):17-26.

2003;137(1-2):15-21.

1975.

Sanitation;1977.

2005;68(17-18):1471-1484.

Taichung, Taiwan1977.

prediction of mortality or arrhythmic events after myocardial infarction: results of a prospective, long-term follow-up study. *Circulation.* Jun 30 1998;97(25):2543-

atherosclerotic diseases in arseniasis-endemic area in southwestern coast of

atherosclerotic diseases and predicts long-term cardiovascular mortality in subjects with previous exposure to arsenic: A 17-Year follow-up study. *Cardiovasc Toxicol.* 

blackfoot disease-endemic area in Taiwan: high-arsenic artesian well water and

disease in arseniasis-hyperendemic villages in Taiwan. *Toxicol Lett.* Jan 31

Physicochemical characteristics of drinking water in endemic blackfoot disease areas. *Memoirs of College of Medicine (National Taiwan University).* 1962;8:115-129. [60] Kuo TL. *Arsenic Content of Artesian Well Water in Endemic Area of Chronic Arsenic* 

progress. *Taichung, Taiwan: Taiwan Provincial Institute of Environmental Sanitation*: Taichung, Taiwan: Taiwan Provincial Institute of Environmental Sanitation;

*report.* Taichung, Taiwan: Taiwan Provincial Institute of Environmental

enzymes and arsenic-related hypertension. *J Toxicol Environ Health A.* Sep


**17** 

*Morocco* 

**Abnormal Electrocardiogram in** 

*Medical Intensive Care Unit - Faculty of Medicine and Pharmacy,* 

Aluminum phosphide (AlP) is used throughout the world as pesticides to protect stored grains from rodents and other pests **(Cienki, 2001)**. The chemical is usually formulated in pellets, granules or as a dust. Upon contact with moisture in the environment, AlP undergoes a chemical reaction yielding phosphine gas (PH3), which is the active pesticidal component and a very toxic system poison, makes acute AlP poisoning (AAlPP) extremely

AAlPP has been reported in literature since 1985. The majority of cases of AlP poisoning involving intentional suicide acts. It is a major health problem with a high mortality rate especially in developing countries where AlP is low cost and easily accessible **(Mehpour et al., 2008; Louriz et al., 2009).** Patients who intend to commit suicide take tablets. Once mixed with hydrochloric acid in the stomach, PH3 is immediately released and absorbed rapidly via the lungs causing systemic poisoning **(Proudfoot, 2009).** However, accidental exposure to AlP is a relatively common cause of poisoning from agriculture chemical exposure in many countries. The manufacture and application of AlP fumigants pose risks of inhalation

During the past 35 years, high mortality rates have been reported following significant exposures to aluminum phosphide. This mortality rates vary from 40% to 80% **(Chugh et al.,** 

The toxicity of AlP is attributed to the liberation of PH3 gas which is cytotoxic and causes

PH3 a nucleophile, acts as a strong reducing agent capable of inhibiting cellular enzymes involved in several metabolic processes. Early studies on PH3 demonstrated specific inhibitory effects on mitochondrial cytochrome c oxidase. Experimental and observational studies have subsequently demonstrated that the inhibition of cytochrome c oxidase and other enzymes leads to the generation superoxide radicals and cellular peroxides. Cellular

**1. Introduction** 

dangerous **(Sasser et al., 1998).** 

exposure to PH3 **(Sudakin, 2005).**

**2. Toxicology of phosphine gas** 

free radical mediated injury **(Sudakin, 2005).** 

**1991).**

**Patients with Acute Aluminum** 

**Phosphide Poisoning** 

*University Mohamed V, Rabat,* 

Amine Ali Zeggwagh and Maha Louriz


## **Abnormal Electrocardiogram in Patients with Acute Aluminum Phosphide Poisoning**

Amine Ali Zeggwagh and Maha Louriz *Medical Intensive Care Unit - Faculty of Medicine and Pharmacy, University Mohamed V, Rabat, Morocco* 

### **1. Introduction**

318 Advances in Electrocardiograms – Clinical Applications

[83] Keating BJ, Tischfield S, Murray SS, et al. Concept, design and implementation of a

[84] Dorn GW, Cresci S. Genome-wide association studies of coronary artery disease and heart failure: where are we going? *Pharmacogenomics.* Feb 2009;10(2):213-223.

studies. *PLoS One.* 2008;3(10):e3583.

cardiovascular gene-centric 50 k SNP array for large-scale genomic association

Aluminum phosphide (AlP) is used throughout the world as pesticides to protect stored grains from rodents and other pests **(Cienki, 2001)**. The chemical is usually formulated in pellets, granules or as a dust. Upon contact with moisture in the environment, AlP undergoes a chemical reaction yielding phosphine gas (PH3), which is the active pesticidal component and a very toxic system poison, makes acute AlP poisoning (AAlPP) extremely dangerous **(Sasser et al., 1998).** 

AAlPP has been reported in literature since 1985. The majority of cases of AlP poisoning involving intentional suicide acts. It is a major health problem with a high mortality rate especially in developing countries where AlP is low cost and easily accessible **(Mehpour et al., 2008; Louriz et al., 2009).** Patients who intend to commit suicide take tablets. Once mixed with hydrochloric acid in the stomach, PH3 is immediately released and absorbed rapidly via the lungs causing systemic poisoning **(Proudfoot, 2009).** However, accidental exposure to AlP is a relatively common cause of poisoning from agriculture chemical exposure in many countries. The manufacture and application of AlP fumigants pose risks of inhalation exposure to PH3 **(Sudakin, 2005).**

During the past 35 years, high mortality rates have been reported following significant exposures to aluminum phosphide. This mortality rates vary from 40% to 80% **(Chugh et al., 1991).**

### **2. Toxicology of phosphine gas**

The toxicity of AlP is attributed to the liberation of PH3 gas which is cytotoxic and causes free radical mediated injury **(Sudakin, 2005).** 

PH3 a nucleophile, acts as a strong reducing agent capable of inhibiting cellular enzymes involved in several metabolic processes. Early studies on PH3 demonstrated specific inhibitory effects on mitochondrial cytochrome c oxidase. Experimental and observational studies have subsequently demonstrated that the inhibition of cytochrome c oxidase and other enzymes leads to the generation superoxide radicals and cellular peroxides. Cellular

Abnormal Electrocardiogram in Patients with Acute Aluminum Phosphide Poisoning 321

clinical, biological, electrical and histological observations suggest that myocardial

Increased CK with raised cardiac marker CK-MB fraction has been previously reported point to severe myocardial damage **(Nakakita et al., 2009).** Controversies exist about the magnesium level and prognosis of AAlPP. Some studies seem to suggest that there is hypomagnesaemia associated with AAlPP and that there is a direct relationship between abnormal electrocardiographic findings and low magnesium levels. They report reduced mortality rates with magnesium therapy in these patients **(Chugh et al., 1991)**. However other studies have shown no such benefits and some have even demonstrated hypermagnesaemia in patients with AAlPP **(Singh et al., 1991).** The pathogenesis of

PH3 also has corrosive effects on tissues. Louriz et al investigated microscopic changes in vital organs of the body, liver, heart and kidneys **(Louriz et al., 2009).** These changes were found to be suggestive of cellular hypoxia. Other recent studies with more patients were performed and showed congestion, edema and leukocytie or leukocyte infiltration in the liver, kidneys, heart,

Nakakita studied the histology of the heart. Histopathological finding of myocyte vacuolation and myocytolysis and degeneration are both suggestive of myocardial injury. The areas of increased waviness of myocardial fibers indicate an episode of myocardial

The toxicity of AlP is systemic and can affect all organs, but particularly cardiac and vascular tissues. Myocardial injury following AlP poisoning has been documented on electrocardiograms in several studies. AlP induced cardiotoxicity was responsible for a high level of mortality. Cardiac toxicity due to AlP and PH3 exposure is represented by a depression in myocardial cellular metabolism, as well as myocardial necrosis due to the

Several studies noted electric abnormalities in 38% to 91% of cases **(Chugh et al., 1991; Karla et al., 1991).** There are conduction disorders such as right and left bundle branch block (25%), atrioventricular block (8%) and rarely, sinoatrial block **(Karla et al., 1991).** On the other hand, cardiac dysrhythmias were described as atrial fibrillation (4% to 61%), junctional rhythm (4% to 100%), ventricular and atrial extrasyxtoles (18%) and ventricular fibrillation (2%) **(Gupta et al., 1995; Louriz et al, 2009).** Finally, re-polarization disorders were also reported, such as ST segment depression (12% to 65%), ST segment elevation (4% to 65%)

Indeed, in Shadnia's study, 39 patients admitted to the ICU with AAlPP were studied. Average time elapsed between poisoning and admission at the hospital was 3.4 ± 3.5 hours. Average ingested amount was 1.4 ± 0.9 tablets. ECG abnormalities were found in 17 (43.6%) cases at the time of admission with ST-T changes in 8 cases. Ischemic change in 3 cases and dysrhythmias in 6 cases. The nature of these dysrhythmias was not described. The mortality rate was high, about 67%. In this study, ECG abnormalities were a prognostic factor

stomach, lungs, brain and adrenals **(Arora et al., 1995; Sinha et al., 2005).**

and T wave inversion (36%) induced by AAlPP **(Lall et al., 1997).**

involvement is responsible for the acute circulatory insufficiency **(Lall et al., 1997).** 

**3.2 Biological abnormalities in AAlPP** 

magnesium level abnormalities was not clear.

**3.3 Histological injury in AAlPP** 

infarction **(Nakakita et al., 2009)**.

release of reactive oxygen intermediates.

**3.4 Cardiotoxicity of AlP** 

**(Shadnia et al., 2010).**

Fig. 1. Aluminum phosphide and liberation of phosphine gas. Al≡P : aluminum phosphide ; PH3 : phosphine gas

injury subsequently occurs through lipid peroxidation and other oxidant mechanisms **(Chugh et al., 1996).** Indeed, significant decreases in glutathione concentrations were shown in different tissues during AlP poisoning **(Hsu et al., 2002).** Glutathione is known to be an important factor protecting against oxidation by catalyzing the reduction of the oxygen peroxide in O2 and H2O.

Mutagenic effects resulting from oxidative damage to DNA have been reported in vitro. A cytogenetic study of phosphide fumigant applicators reported a significantly higher incidence of chromatid gaps and deletions in comparison to controls **(Hsu et al., 1998).** 

Very little information is available relating to the toxicokinetics of PH3 in humans. In an investigation of a series of patients with acute AlP poisoning , indicators of oxidative stress (malonydialdehyde levels, superoxide dismutase and catalase activity) appeared to peak within 48 hours of exposure, with normalization of most indicators occurring by day 5 **(Chugh et al., 1996a).** Chugh et al. reported that, serum PH3 levels correlate positively with the severity of poisoning and levels equal to or less than 1.067 ± 0.16 mg % appear to be the limit of PH3 toxicity **(Chugh et al., 1996b).**

A bedside test has been described for the diagnosis of AlP ingestion, using gastric aspirates and paper strips impregnated with silver nitrate. The test was found to be positive in 100% of cases of AlP ingestion **(Chugh et al., 1994).** The same test was also investigated to detect PH3 gas in the exhaled air of patients with intoxication with AlP ingestion, and the results were positive in 50 % of cases.

### **3. Acute aluminum phosphide poisoning**

### **3.1 Clinical features of AAlPP**

AAlPP results in the rapid onset of gastrointestinal signs and symptoms, including epigastric pain and recurrent, profuse vomiting. Characteristic garlic smell of PH3 in the patient's expired breath. Cardiovascular manifestations include hypotention and profound circulatory collapse. Neurological manifestations following acute poisoning include headache, anxiety and dizziness, frequently accompanied by a normal mental state **(National Institute of occupational Safety and Health [NIOSH], 2003).** Pulmonary injury and oedema have been described. Acute renal and liver injury can also develop. The prognosis from suicidal ingestion of AlP is poor.

The major lethal consequence of AAlPP, myocardial suppression with profound circulatory collapse, is reportedly secondary to toxins generated, which lead to direct effects on cardiac myocytes, fluid loss induced by several episodes of vomiting and adrenal gland damage. The AAlPP is involving young patients without history of cardiac diseases. However, clinical, biological, electrical and histological observations suggest that myocardial involvement is responsible for the acute circulatory insufficiency **(Lall et al., 1997).** 

### **3.2 Biological abnormalities in AAlPP**

320 Advances in Electrocardiograms – Clinical Applications

Fig. 1. Aluminum phosphide and liberation of phosphine gas. Al≡P : aluminum phosphide ;

injury subsequently occurs through lipid peroxidation and other oxidant mechanisms **(Chugh et al., 1996).** Indeed, significant decreases in glutathione concentrations were shown in different tissues during AlP poisoning **(Hsu et al., 2002).** Glutathione is known to be an important factor protecting against oxidation by catalyzing the reduction of the oxygen

Mutagenic effects resulting from oxidative damage to DNA have been reported in vitro. A cytogenetic study of phosphide fumigant applicators reported a significantly higher incidence of chromatid gaps and deletions in comparison to controls **(Hsu et al., 1998).**  Very little information is available relating to the toxicokinetics of PH3 in humans. In an investigation of a series of patients with acute AlP poisoning , indicators of oxidative stress (malonydialdehyde levels, superoxide dismutase and catalase activity) appeared to peak within 48 hours of exposure, with normalization of most indicators occurring by day 5 **(Chugh et al., 1996a).** Chugh et al. reported that, serum PH3 levels correlate positively with the severity of poisoning and levels equal to or less than 1.067 ± 0.16 mg % appear to be the

A bedside test has been described for the diagnosis of AlP ingestion, using gastric aspirates and paper strips impregnated with silver nitrate. The test was found to be positive in 100% of cases of AlP ingestion **(Chugh et al., 1994).** The same test was also investigated to detect PH3 gas in the exhaled air of patients with intoxication with AlP ingestion, and the results

AAlPP results in the rapid onset of gastrointestinal signs and symptoms, including epigastric pain and recurrent, profuse vomiting. Characteristic garlic smell of PH3 in the patient's expired breath. Cardiovascular manifestations include hypotention and profound circulatory collapse. Neurological manifestations following acute poisoning include headache, anxiety and dizziness, frequently accompanied by a normal mental state **(National Institute of occupational Safety and Health [NIOSH], 2003).** Pulmonary injury and oedema have been described. Acute renal and liver injury can also develop. The

The major lethal consequence of AAlPP, myocardial suppression with profound circulatory collapse, is reportedly secondary to toxins generated, which lead to direct effects on cardiac myocytes, fluid loss induced by several episodes of vomiting and adrenal gland damage. The AAlPP is involving young patients without history of cardiac diseases. However,

PH3 : phosphine gas

peroxide in O2 and H2O.

limit of PH3 toxicity **(Chugh et al., 1996b).**

**3. Acute aluminum phosphide poisoning** 

prognosis from suicidal ingestion of AlP is poor.

were positive in 50 % of cases.

**3.1 Clinical features of AAlPP** 

Increased CK with raised cardiac marker CK-MB fraction has been previously reported point to severe myocardial damage **(Nakakita et al., 2009).** Controversies exist about the magnesium level and prognosis of AAlPP. Some studies seem to suggest that there is hypomagnesaemia associated with AAlPP and that there is a direct relationship between abnormal electrocardiographic findings and low magnesium levels. They report reduced mortality rates with magnesium therapy in these patients **(Chugh et al., 1991)**. However other studies have shown no such benefits and some have even demonstrated hypermagnesaemia in patients with AAlPP **(Singh et al., 1991).** The pathogenesis of magnesium level abnormalities was not clear.

### **3.3 Histological injury in AAlPP**

PH3 also has corrosive effects on tissues. Louriz et al investigated microscopic changes in vital organs of the body, liver, heart and kidneys **(Louriz et al., 2009).** These changes were found to be suggestive of cellular hypoxia. Other recent studies with more patients were performed and showed congestion, edema and leukocytie or leukocyte infiltration in the liver, kidneys, heart, stomach, lungs, brain and adrenals **(Arora et al., 1995; Sinha et al., 2005).**

Nakakita studied the histology of the heart. Histopathological finding of myocyte vacuolation and myocytolysis and degeneration are both suggestive of myocardial injury. The areas of increased waviness of myocardial fibers indicate an episode of myocardial infarction **(Nakakita et al., 2009)**.

### **3.4 Cardiotoxicity of AlP**

The toxicity of AlP is systemic and can affect all organs, but particularly cardiac and vascular tissues. Myocardial injury following AlP poisoning has been documented on electrocardiograms in several studies. AlP induced cardiotoxicity was responsible for a high level of mortality. Cardiac toxicity due to AlP and PH3 exposure is represented by a depression in myocardial cellular metabolism, as well as myocardial necrosis due to the release of reactive oxygen intermediates.

Several studies noted electric abnormalities in 38% to 91% of cases **(Chugh et al., 1991; Karla et al., 1991).** There are conduction disorders such as right and left bundle branch block (25%), atrioventricular block (8%) and rarely, sinoatrial block **(Karla et al., 1991).** On the other hand, cardiac dysrhythmias were described as atrial fibrillation (4% to 61%), junctional rhythm (4% to 100%), ventricular and atrial extrasyxtoles (18%) and ventricular fibrillation (2%) **(Gupta et al., 1995; Louriz et al, 2009).** Finally, re-polarization disorders were also reported, such as ST segment depression (12% to 65%), ST segment elevation (4% to 65%) and T wave inversion (36%) induced by AAlPP **(Lall et al., 1997).**

Indeed, in Shadnia's study, 39 patients admitted to the ICU with AAlPP were studied. Average time elapsed between poisoning and admission at the hospital was 3.4 ± 3.5 hours. Average ingested amount was 1.4 ± 0.9 tablets. ECG abnormalities were found in 17 (43.6%) cases at the time of admission with ST-T changes in 8 cases. Ischemic change in 3 cases and dysrhythmias in 6 cases. The nature of these dysrhythmias was not described. The mortality rate was high, about 67%. In this study, ECG abnormalities were a prognostic factor **(Shadnia et al., 2010).**

Abnormal Electrocardiogram in Patients with Acute Aluminum Phosphide Poisoning 323

C)

D1)

B)

A)

B)

C)

Abnormal Electrocardiogram in Patients with Acute Aluminum Phosphide Poisoning 325

In Louriz's study, 49 patients were enrolled. The ingested dose was 1.2 ± 0.7 grams. The time between ingestion and admission to the medical ICU was 9.1 ± 10.7 hours. The ECG was abnormal in 28 cases (58.7%) at the time of admission with myocardial ischemia in 21 cases, atrial fibrillation in 6 cases, ventricular extrasystoles in 9 cases and ventricular fibrillation in one case. The mortality rate was 49%. In this study, ECG abnormalities were also a

In Mathai's study, 27 patients with AAlPP were admitted into the ICU. One and half grams of poison was consumed. There was a mean delay of 2.1 ± 1.55 hours before presenting to the hospital. Thirteen (48.1%) patients had dysrhythmia at admission of which the majority (69%) of supraventricular origin. Ventricular arythmias was found in 4 cases. The mortality rate was 59.3%. In this study, the presence of ECG abnormalities did not predict mortality

In these 3 studies cited above, there wasn't any association between the dose of poison consumed or the time delay in presentation to the hospital with the mortality. All the ECG abnormalities found in these studies were recorded at the admission to the hospital.

Indeed, Bogle et al described a case of a lethal AAlPP caused by deliberate ingestion of AlP. ECG showed a sinus tachycardia 2 hours after ingestion of a 10 g sachet of pesticide with 56% of AlP. ECG recorded 12 hours after ingestion showed extreme widening of the QRS complexe despite amiodarone therapy. The rate of ECG abnormalities resulting of AAlPP

In several studies, echocardiography showed a global hypokinesis of the left ventrivule

A) B)

Fig. 3. Echocardiographically parasternal long axis depicted improvement of ventricle function following aluminum phosphide poisoning . A) Echocardiogram obtained on the second day after admission depicts global hypokinesis with LVEF of 30% and dilatation of LV. B) Echocardiogram taken eight days after admission showed improvement of ventricle

However, other ECG changes could be found during the hospitalization.

Some electrical abnormalities noted in our practice are reported in Figure 2.

prognostic factor **(Louriz et al., 2009).**

might be under estimated **(Bogle et al., 2006).**

**(Akkaoui et al., 2007; Bajaj et al., 1988)** (Figure 3).

**(Mathai et al., 2010).**

function.

D2)

D3)

Fig. 2. Electrocardiographic changes following aluminium phosphide poisoning. 12-leads surface ECG recorded on admission showing: (A) Sinusal bradycardia (B) Ventricular extrasystoles (C) Sinusal tachycardia with ST segment depression in the leads D1, aVL, V5 and V6 (D1, D2 &D3) Sinusal tachycardia with ST-T changes.

D2)

D3) Fig. 2. Electrocardiographic changes following aluminium phosphide poisoning. 12-leads surface ECG recorded on admission showing: (A) Sinusal bradycardia (B) Ventricular extrasystoles (C) Sinusal tachycardia with ST segment depression in the leads D1, aVL, V5

and V6 (D1, D2 &D3) Sinusal tachycardia with ST-T changes.

In Louriz's study, 49 patients were enrolled. The ingested dose was 1.2 ± 0.7 grams. The time between ingestion and admission to the medical ICU was 9.1 ± 10.7 hours. The ECG was abnormal in 28 cases (58.7%) at the time of admission with myocardial ischemia in 21 cases, atrial fibrillation in 6 cases, ventricular extrasystoles in 9 cases and ventricular fibrillation in one case. The mortality rate was 49%. In this study, ECG abnormalities were also a prognostic factor **(Louriz et al., 2009).**

In Mathai's study, 27 patients with AAlPP were admitted into the ICU. One and half grams of poison was consumed. There was a mean delay of 2.1 ± 1.55 hours before presenting to the hospital. Thirteen (48.1%) patients had dysrhythmia at admission of which the majority (69%) of supraventricular origin. Ventricular arythmias was found in 4 cases. The mortality rate was 59.3%. In this study, the presence of ECG abnormalities did not predict mortality **(Mathai et al., 2010).**

In these 3 studies cited above, there wasn't any association between the dose of poison consumed or the time delay in presentation to the hospital with the mortality. All the ECG abnormalities found in these studies were recorded at the admission to the hospital. However, other ECG changes could be found during the hospitalization.

Indeed, Bogle et al described a case of a lethal AAlPP caused by deliberate ingestion of AlP. ECG showed a sinus tachycardia 2 hours after ingestion of a 10 g sachet of pesticide with 56% of AlP. ECG recorded 12 hours after ingestion showed extreme widening of the QRS complexe despite amiodarone therapy. The rate of ECG abnormalities resulting of AAlPP might be under estimated **(Bogle et al., 2006).**

Some electrical abnormalities noted in our practice are reported in Figure 2.

In several studies, echocardiography showed a global hypokinesis of the left ventrivule **(Akkaoui et al., 2007; Bajaj et al., 1988)** (Figure 3).

Fig. 3. Echocardiographically parasternal long axis depicted improvement of ventricle function following aluminum phosphide poisoning . A) Echocardiogram obtained on the second day after admission depicts global hypokinesis with LVEF of 30% and dilatation of LV. B) Echocardiogram taken eight days after admission showed improvement of ventricle function.

Abnormal Electrocardiogram in Patients with Acute Aluminum Phosphide Poisoning 327

Bajaj R, Wasir HS, Agarwal R, Malhotra A, Chopra P & Bhatia ML. (1988). Aluminum

Bogle RG, Theron P, Brooks P, Dargan PI & Redhead J. (2006). Aluminium phosphide

Chugh SN, Dushayant K, Ram S & Arora B. (1991). Incidence and outcome of aluminum phosphide poisoning in a hospital study. *Indian J Med Res*, No.94, pp. 232-35. Chugh SN, Chung K, Ram S & Malhotra KC. (1991). Electrocardiographic abnormalities in

Chugh SN, Kamar P, Sharma A, Chugh K, Mittal A & Arora B. (1994). Magnesium status

Chugh SN, Arora V, Sharma A & Chugh K (1996). Free radical scavengers and lipid

Chugh SN, Pal R, Singh V& Seth S. (1996). Serial blood phosphine levels in acute aluminum phosphide poisoning. *J Assoc Physicians India* 1996b, No.44, pp. 184-5 Cienki JJ. (2001). *Non-anticoagulant rodenticides*. (Ford MD, Delaney KA, Ling LJ, Erickson T).

Gupta MS, Malik A & Shama VK. (1995). Cardiovascular manifestations in aluminum

Hsu CH, Quistad GB & Casida JE. (1998). Phosphine-induced oxidative stress in Hepa

Hsu C-H, CHiB-C, Liu J-H, Chen C-J & Chen R-Y. (2002). phosphine induced oxidative

Karla GS, Anand IS, Jit I, Bushnurmath B &Wahi PL. (1991) Aluminum phosphide poisoning

Lall SB,Sinha K, Mittra S & Seth SD. (1997). An experimental study on cardiotoxicity of

Louriz M, Dendane T, Abidi K, Madani N, Abouqal R & Zeggwagh AA. (2009). Prognostic

Mathai A & Bhanu MS. (2010). Acute aluminium phosphide poisoning: can we predict

Mehrpour O, Alfred S, Shadnia S, Keyler DE, Soltaninejad K & Chalaki N. (2008).

Nakakita, H.a , Katsumata, Y.b & Ozawa, T.b . (2009). The effect of phosphine on respiration of rat liver mitochondria. *Indian Journal of Critical Care Medicine*, No.13, pp. 41-43 National Institute of occupational Safety and Health. NIOSH alert: preventing phosphine

factors of acute aluminum phosphide poisoning. *Indian J Med Sci*, No.63, pp. 227-34

Hyperglycemia in acute aluminum phosphide poisoning as a potential prognostic

poisoning and explosions during fumigation. Columbus, OH: NIOSH,2003:99-

damage in rats: Role of glutathione. *Toxicology*, No.179, pp.1-8

haemodynamic observations. *Indian Heart J*, No.43, pp. 175-8

aluminum phosphide. *Indian J Exp Biol*, No.35, pp. 1060-1064

mortality ? *Indian J Anaesth*, No.54, pp. 302-307

factor. *Hum Exp Toxicol ,* No*.*27, pp.591-5.

patients. *Natl Med J India*, No.1, pp. 270-274

intoxication. *Magnes Res*, No. 7, pp. 289-94.

*Clinical Toxicology*, 858. Philadelphia.

*Physicians India*, No.43, pp. 774-780

1c1c7cells. *Toxicol Sci*, No.46, pp. 204

poisoning. *Emerg Med J*, No.23, pp.e3

437-9.

No104, pp. 190-3.

phosphide poisoning.Clinical toxicity and outcome in eleven intensively monitored

aluminum phosphide poisoning with special reference to its incidence, pathogenesis, mortality and hisptopathology. *J Indian Med Assoc ,*No.89, pp. 32-35 Chugh SN, Jaggal KL, Sharma A, Arora B & Malhotra KC. (1991). Magnesium levels in acute

cardiotoxicity due to aluminum phosphide poisoning. *Indian J Med Res,* No.94, pp.

and parenteral magnesium sulphate therapy in acute aluminium phosphide

peroxidation in acute aluminum phosphide poisoning. *Indian J Med Res* 1996a,

phosphide poisoning with especial reference to echocardiographic changes. *J Assoc* 

Indeed, Bahsin et al showed a generalized hypokinesis of left ventricle wall and interventricular septum in 80% of their cases. This study revealed akinesis and pericarditis in 3% and 35% of the cases, respectively **(Bhasin et al., 1991).** Thus, Bajaj et al showed a global hypokinesis of the left ventricle in the three patients that underwent serial ventriculography **(Bajaj et al., 1988).** However, other authors have described a focal myocardial necrosis **(Singh et al., 1991; Wilson et al., 1980).** At a patient hospitalized in our intensive care unit, we observed an unusual and important dilation of right cardiac cavities explained partially by hypokinesia of the left ventricle but probably also by the direct toxicity of the AlP on the right ventricle (Figure 4).

Fig. 4. Echocardiography parasternal long axis depicted improvement of ventricule function. Echocardiogram obtained at admission depicts global hypokinesis with left ventricle ejection fraction (LVEF) of 40% and important dilation of right ventricle and auricle.

### **4. Conclusion**

The severity of the poisoning is judged by the cardiac failure and the unavailability of an antidote. Myocardial injury following AAlPP is responsible for significant mortality. Despite all intensive medical care efforts in supportive therapy, the prognosis of AAlPP is poor. Therefore the use and availability of the pesticide aluminium phosphide should be restricted as much as possible.

### **5. References**


Indeed, Bahsin et al showed a generalized hypokinesis of left ventricle wall and interventricular septum in 80% of their cases. This study revealed akinesis and pericarditis in 3% and 35% of the cases, respectively **(Bhasin et al., 1991).** Thus, Bajaj et al showed a global hypokinesis of the left ventricle in the three patients that underwent serial ventriculography **(Bajaj et al., 1988).** However, other authors have described a focal myocardial necrosis **(Singh et al., 1991; Wilson et al., 1980).** At a patient hospitalized in our intensive care unit, we observed an unusual and important dilation of right cardiac cavities explained partially by hypokinesia of the left ventricle but probably also by the direct toxicity of the AlP on the

Fig. 4. Echocardiography parasternal long axis depicted improvement of ventricule function.

The severity of the poisoning is judged by the cardiac failure and the unavailability of an antidote. Myocardial injury following AAlPP is responsible for significant mortality. Despite all intensive medical care efforts in supportive therapy, the prognosis of AAlPP is poor. Therefore the use and availability of the pesticide aluminium phosphide should be

Abder-Rahman H. (1999). Effect of aluminum phosphide on blood glucose level. *Vet Human* 

Akkaoui M, Achour S, Abidi K, Himdi B, Madani N, Zeggwagh AA & Abouqal R. (2007).

Arora B, Punia RS, Karla R, Chugh SN & Arora DR. (1995). Histopathological changes in aluminum phosphide poisoning. *J Indian Med Assoc*, No. 93, pp. 380-81.

Reversible myocardial injury associated with aluminum phosphide poisoning. *Clin* 

Echocardiogram obtained at admission depicts global hypokinesis with left ventricle ejection fraction (LVEF) of 40% and important dilation of right ventricle and auricle.

right ventricle (Figure 4).

**4. Conclusion** 

**5. References** 

restricted as much as possible.

*Toxicol,* No.41, pp.31-32

*Toxicol*, No.45, pp. 728-31


126,1-16Proudfoot AT. (2009). Aluminum and Zinc phosphide poisoning. *Cli Toxicol* , No.47, pp. 89-100


Sasser SM. (1998). Rodenticides. In: Viccellio P, BaniaT, Brent J, Hoffman RS, Kulig KW,

Shadnia S, Mehrpour O & Soltaninejad K. (2010). A simplified acute physiology score in the

Sinha US, Kapoor AK, Singh AK, Gupta A & Mehrotra R. (2005). Histopathological changes

Singh RB, Singh RG & Singh U. (1991). Hyperglicemia following aluminum phosphide

Sudakin DL. (2005). Occupational exposure to aluminum phosphide and phosphine gas ? A

Wilson R, Lovejoy FH, Jaeger RJ & Landrigan PL. (1980). Acute phosphine poisoning aboard

poisoning. *Int J Clin Pharmacol Ther Toxicol*, No.29, pp. 82-85

*Toxicol* , No.47, pp. 89-100

Lippincott-Ravec; P. 433.

*toxicology*, No.24, pp. 27-33

No.64, pp. 532-539

180

pp. 148-150

126,1-16Proudfoot AT. (2009). Aluminum and Zinc phosphide poisoning. *Cli* 

Mofenson HC, et al, editors. Emergency toxicology. 2end ed. Philadelphia, PA:

prediction of acute aluminium phosphide poisoning outcome. *Indian J Med Sci* ,

in casesof aluminum phosphide poisoning. *Indian J Pathol Microbiol*, No.48, pp. 177-

suspected case report and review of the literature. *Human and experimental* 

a grain freighter Epidemiologic, clinical and pathological finding. *JAMA*, No.244,

### *Edited by Richard M. Millis*

Electrocardiograms have become one of the most important, and widely used medical tools for diagnosing diseases such as cardiac arrhythmias, conduction disorders, electrolyte imbalances, hypertension, coronary artery disease and myocardial infarction. This book reviews recent advancements in electrocardiography. The four sections of this volume, Cardiac Arrhythmias, Myocardial Infarction, Autonomic Dysregulation and Cardiotoxicology, provide comprehensive reviews of advancements in the clinical applications of electrocardiograms. This book is replete with diagrams, recordings, flow diagrams and algorithms which demonstrate the possible future direction for applying electrocardiography to evaluating the development and progression of cardiac diseases. The chapters in this book describe a number of unique features of electrocardiograms in adult and pediatric patient populations with predilections for cardiac arrhythmias and other electrical abnormalities associated with hypertension, coronary artery disease, myocardial infarction, sleep apnea syndromes, pericarditides, cardiomyopathies and cardiotoxicities, as well as innovative interpretations of electrocardiograms during exercise testing and electrical pacing.

Advances in Electrocardiograms - Clinical Applications

Advances in

Electrocardiograms

Clinical Applications

*Edited by Richard M. Millis*

Photo by Dutko / iStock