**3.2.2 Results**

288 Dyslipidemia - From Prevention to Treatment

United States of America the results are also somewhat disappointing. In survivors of acute myocardial infarction or stroke, the control percentages for some primary risk factors are below expectations, particularly for smoking habits (18%), control of hypercholesterolemia

As in almost all chronic conditions, the real picture lags far behind the expectations and available resources. Regarding hypercholesterolemia, the current situation is even less understandable, given its clear and strong association with the prevailing causes of death and incapacity, and the public awareness of the problem and in consideration of the demonstrated effectiveness of the available lipid-lowering drugs, that may have a quite

Given the demonstrated role-playing of blood cholesterol in the atherosclerotic continuum, we designed two studies to ascertain the usefulness of the LDL-cholesterol/HDLcholesterol, Triglycerides/HDL-cholesterol and Total-cholesterol/HDL-cholesterol ratios in predicting cardiovascular risk, through its relation to cardiovascular events and peripheral

**3.2 Study 1 – Usefulness of the lipidic ratios predicting peripheral artery disease in** 

The importance of the lipidic profile is well established in atherosclerotic processes related to coronary artery disease. Its relation with atherosclerosis in other vascular territories, particularly the inferior limbs has also received strong support from several experimental settings and in different clinical contexts. In order to address wether the lipid ratios can predict the occurrence of obstructive peripheral artery disease (PAD) we conducted a crosssectional study in a sample of hypertensive patients. The study population consisted of 920

A total of 920 hypertensive patients (51.3% female, age 64.22 ± 12.01 years) were consecutively included in the study. None of the patients were taking drugs or were in situations known to affect lipoprotein metabolism. Total cholesterol, triglycerides and HDL cholesterol were measured. LDL cholesterol was obtained by Friedewald's formula (if triglycerides <3.39 mmol/l) or by ultracentrifugation. The LDL-Cholesterol/HDL-Cholesterol, Total Cholesterol/HDL-Cholesterol and Triglycerides/HDL-Cholesterol ratios were calculated in all patients. Blood pressure and heart rate were measured in standard conditions. Ankle-Brachial index (ABI) was estimated bilaterally as the ratio of ankle (left and right) systolic blood pressure and brachial (highest upper limb) systolic blood pressure. The normal range for ABI was 0.9-1.3 mmHg, and individuals with ABI<0.9 were classified

All data was processed using STATA for Windows, version 11.1. The distribution of the variables was tested for normality using the Kolmogorov-Smirnov test, and for homogeneity of variance by Levene's test. Simple descriptive statistics were used to characterize the sample and the distribution of variables. Logistic regression analysis was

used to determine the influence of the lipidic parameters on the occurrence of PAD.

(46%), diabetes (48%) and hypertension (53%) (Qureshi et al, 2001).

arterial disease (PAD) in two different clinical and experimental settings.

Portuguese nationals, aged between 20 and 91 years (mean 64.23 + 12.30 years).

favorable impact upon the prognosis of patients.

**hypertensive patients: A retrospective analysis** 

**3. Original research data** 

**3.1 Aim** 

**3.2.1 Methods** 

as having peripheral arterial disease.

The general characteristics of the studied population are summarized in Table 4. Mean age was 64.23±12.30, with a similar proportion of men versus women (49% and 51%, respectively).


PAD – peripheral artery disease; CV – cardiovascular events; SBP – systolic blood pressure; DBP – diastolic blood pressure; eGFR – estimated Glomerular Filtration Rate; ABI – Ankle-Brachial Index

Table 4. Characteristics of the study population, in general and stratified for the presence or absence of peripheral artery disease.

Mean body mass index was 28.79±11.85, indicating an overwheighted population. With regard to cardiovascular risk factors, all patients were hypertensive, 60% had dyslipidemia and 34% were diabetic; 11% were smokers and 12% had a personal history of cardiovascular events (mainly Stroke). About 37% were medicated for cardiovascular pathologies, with 13.6% of the patients undertaking statins. This factor was controlled in all the multivariable analysis. Peripheral artery disease (PAD) was encountered in 117 patients (12.7%). Patients with PAD were older, and had a worst metabolic and hemodynamic profile. The proportion of patients with a personal history of cardiovascular events was also greater in patients with PAD (25% versus 10%, p<0.01). The Ankle-Brachial Index (ABI) was also significantly lower

Dyslipidemia and Cardiovascular Risk:

PAD – peripheral artery disease

arterial disease.

OR – Odds Ratio

Framingham cardiovascular risk factors.

Lipid Ratios as Risk Factors for Cardiovascular Disease 291

Fig. 1. Representation of the comparative lipid ratios in patients with and without peripheral

Fig. 2. Adjusted Odds Ratios for Peripheral Artery Disease for the individual lipidic variables and for the lipidic ratios. The Odds Ratios are multi-adjusted to conventional

in patients with PAD, as expected. Interestingly, patients with PAD also had a significantly lower estimated glomerular filtration rate (76.94±20.88 ml/min/1.73m2 versus 85.86±23.40 ml/min/1.73m2 in patients without PAD).

Regarding the overall and the comparative lipidic profile (depicted in table 5), significant differences amongst patients with and without PAD were only observed for the three considered lipidic ratios, expressing higher values when PAD was present, and for the HDL-cholesterol, with the PAD patients reaching lower HDL levels (although tendencially, *p-value*=0.073).


Table 5. Lipid profile of the study population.

Figure 1 further ilustrates the differences in the lipidic ratios among patients with and without PAD, with all three considered ratios presenting significant differences between the considered groups.

A multivariable logistic regression analysis was also performed considering PAD as the dependent variable (dichotomized in normal/abnormal), and forcing each lipidic parameter (either individual lipis or lipid ratios) in a model adjusted for the conventional Framingham cardiovascular risk factors (age, sex, diabetes, blood pressure, smoking status and body mass index). The observed Odds Ratios (OR) with 95% confidence intervals is depicted in figure 2. Although there's an appreciable tendency of association with PAD in all lipid variables, it reaches statistical significancy only for the lipidic ratios. In fact, the OR for LDLcholesterol, Total-cholesterol, HDL-cholesterol and Triglycerides were respectively 1.004 (IC: 0.999-1.010, p=0.1), 1.001 (IC: 0.996-1.007, p=0.4), 0.993 (IC: 0.980-1.004, p=0.2) and 1.001 (IC: 0.998-1.004, p=0.2). For the LDL-cholesterol/HDL-cholesterol ratio, the multiadjusted OR was 1.06 (IC: 0.999-1.120, p=0.052), with a marginally significant association with PAD. For the Total-cholesterol/HDL-cholesterol and the Triglycerides/HDL-cholesterol ratios, the adjusted OR were respectively 1.051 (IC: 1.011-1.200, p=0.01) and 1.050 (IC: 1.002-1.110, p=0.04). A further analysis showed that the association of the lipid ratios with PAD was tendencially linear, particularly for the Total-cholesterol/HDL-cholesterol ratio.

PAD – peripheral artery disease

in patients with PAD, as expected. Interestingly, patients with PAD also had a significantly lower estimated glomerular filtration rate (76.94±20.88 ml/min/1.73m2 versus 85.86±23.40

Regarding the overall and the comparative lipidic profile (depicted in table 5), significant differences amongst patients with and without PAD were only observed for the three considered lipidic ratios, expressing higher values when PAD was present, and for the HDL-cholesterol, with the PAD patients reaching lower HDL levels (although tendencially,

Figure 1 further ilustrates the differences in the lipidic ratios among patients with and without PAD, with all three considered ratios presenting significant differences between the

A multivariable logistic regression analysis was also performed considering PAD as the dependent variable (dichotomized in normal/abnormal), and forcing each lipidic parameter (either individual lipis or lipid ratios) in a model adjusted for the conventional Framingham cardiovascular risk factors (age, sex, diabetes, blood pressure, smoking status and body mass index). The observed Odds Ratios (OR) with 95% confidence intervals is depicted in figure 2. Although there's an appreciable tendency of association with PAD in all lipid variables, it reaches statistical significancy only for the lipidic ratios. In fact, the OR for LDLcholesterol, Total-cholesterol, HDL-cholesterol and Triglycerides were respectively 1.004 (IC: 0.999-1.010, p=0.1), 1.001 (IC: 0.996-1.007, p=0.4), 0.993 (IC: 0.980-1.004, p=0.2) and 1.001 (IC: 0.998-1.004, p=0.2). For the LDL-cholesterol/HDL-cholesterol ratio, the multiadjusted OR was 1.06 (IC: 0.999-1.120, p=0.052), with a marginally significant association with PAD. For the Total-cholesterol/HDL-cholesterol and the Triglycerides/HDL-cholesterol ratios, the adjusted OR were respectively 1.051 (IC: 1.011-1.200, p=0.01) and 1.050 (IC: 1.002-1.110, p=0.04). A further analysis showed that the association of the lipid ratios with PAD was

tendencially linear, particularly for the Total-cholesterol/HDL-cholesterol ratio.

**No PAD Patients**  (n=803)

196.61±41.15 197.06±41.18 193.68±40.98 0.400

116.31±37.62 116.17±41.18 193.68±40.98 0.763

54.46±21.47 54.96±21.56 51.14±20.65 0.073

134.84±67.88 134.20±41.18 139.09±66.02 0.460

2.55±2.45 2.48±2.11 3.04±2.03 <0.01

4.15±2.95 4.06±2.48 4.80±2.05 <0.01

2.97±2.98 2.88±2.44 3.59±2.33 <0.01

**PAD Patients**  (n=117)

*p-value*  **(PAD versus No PAD)** 

(n=920)

ml/min/1.73m2 in patients without PAD).

 **Total** 

**Plasma Total Cholesterol,** 

**Plasma LDL-Cholesterol,** 

**Plasma HDL-Cholesterol,** 

**Plasma Triglicerides,** 

**LDL-Colesterol/HDL-Colesterol Racio** 

**Total Cholesterol/HDL-Colesterol Racio** 

Table 5. Lipid profile of the study population.

**Triglicerídeos/HDL-Colesterol Racio** 

considered groups.

*p-value*=0.073).

**mg/dl**

**mg/dl** 

**mg/dl** 

**mg/dl**

Fig. 1. Representation of the comparative lipid ratios in patients with and without peripheral arterial disease.

OR – Odds Ratio

Fig. 2. Adjusted Odds Ratios for Peripheral Artery Disease for the individual lipidic variables and for the lipidic ratios. The Odds Ratios are multi-adjusted to conventional Framingham cardiovascular risk factors.

Dyslipidemia and Cardiovascular Risk:

interval.

disease.

**Kg/m2**

**no:yes\***

**no:yes\***

**Body Mass Index,** 

**Family History,** 

**Tobacco Consumption,** 

**Plasma Creatinine,** 

**mg/dl**

**3.3.2 Results** 

Lipid Ratios as Risk Factors for Cardiovascular Disease 293

Groups were compared using the χ2 test for categorical variables and the Student's t test (2 groups) or ANOVA with the post-hoc Tukey test (3 groups) for quantitative variables. A value of P≤0.05 was taken as the criterion of statistical significance for a 95% confidence

The general characteristics of the study population are summarized in Table 6. Mean age was 46.33±13.77, indicating a relatively young sample, with similar proportions of men and women (59% and 41%, respectively). With regard to cardiovascular risk factors, 52% of the patients were hypertensive, 33% had dyslipidemia and 11% were diabetic; 17% were smokers and 15% had a family history of cardiovascular events. About 37% were medicated for cardiovascular pathologies, with 13.6% of the patients undertaking statins. This factor was controlled in all the multivariable analysis. Mean follow-up is currently 21.42±10.76 months. A total of 50 non-fatal MACE (2.2% of the sample) were recorded, including 27 cases of stroke, 19 of coronary events, 2 of renal failure and 2 of occlusive peripheral arterial

**Total Low Risk** 

**N,%** 2200 32% 66% 2%

**Patients**

**Age, years** 46.33±13.77 40.00±13.42 49.03±13,14 50.00±10,21 0.360 **Sex, men:women\*** 59:41 60:40 58:42 46:54 0.104

**Waist, cm** 89.82±11,05 86.83±10,30 90.63±11,00 90.00±13,06 0.917

**Hypertension, no:yes\*** 48:52 100:0 26:74 14:86 0.109 **Dyslipidemia, no:yes\*** 67:33 100:0 53:47 60:40 0.311 **Diabetes, no:yes\*** 89:11 100:0 85:15 86:14 0.941 **SBP, mmHg** 142.51±21.05 129.17±14.33 147.83±14.33 161.08±17.34 <0.001 **DBP, mmHg** 84.52±12.29 77.43±10.11 87.32±11.87 92.08±10.07 <0.001 **PP, mmHg** 57.99±15.29 51.74±11.90 60.05±15.86 66.20±12.93 <0.001 **MAP, mmHg** 103.85±14.02 94.68±1.26 107.48±13.52 117.14±11.43 <0.001 **Heart Rate, bpm** 70.56±12.24 68.21±12.58 71.49±11.87 78.20±13.01 0.001 **Plasma Glucose, mg/dl** 100.44±31.54 90.86±9.16 103.70±3.75 110.32±39.64 0.406

MACE – major acute cardiovascular events; SBP – systolic blood pressure; DBP – diastolic blood

Table 6. General characteristics of the study cohort, depending on the presence of MACE

pressure; PP – pulse pressure; MAP – mean blood pressure

and conventional cardiovascular risk factors.

**No MACE MACE**

27.18±5.50 25.90±4.21 27.71±4.45 28.59±5.75 0.348

85:15 92:8 83:17 60:40 0.020

83:17 78:22 85:15 78:22 0.243

1.31±5.08 0.90±1.77 1.43±5.99 1.53±2.92 0.996

**Patients Patients** *p-value*

**(MACE vs No MACE)**

**3.3 Study 2 – Usefulness of the lipidic ratios in a low-to-moderate cardiovascular risk population: A sub-analysis of the EDIVA (Estudo de Distensibilidade Vascular) project**  The EDIVA project was an epidemiological study assessing cardiovascular risk through sequential Pulse Wave Velocity measurement (Maldonado, Pereira, Polónia *et al*, 2011), but since serum lipids were available for all the included patients, we re-analyzed the EDIVA database aiming to address the delineated objective: to ascertain the usefulness of The LDL-Cholesterol/HDL-Cholesterol, Total Cholesterol/HDL-Cholesterol and Triglycerides/HDL-Cholesterol ratios in the general population. The study population consisted of 2200 Portuguese nationals (1290 men and 910 women), aged between 18 and 91 years (mean 46.33±13.76 years). Of these, 668 had low cardiovascular risk, and 1532 were patients with hypertension, diabetes and/or dyslipidemia. Individuals defined as having low cardiovascular risk were those who had had no chronic disease, had never been prescribed chronic pharmacological therapy, and had a normal physical exam, electrocardiogram, blood and urine tests, these characteristics having remained unchanged for at least two annual assessments. The patient group was under pharmacological therapy for at least one of the above pathologies.

#### **3.3.1 Methods**

The study's aims were explained to all participants and their informed consent was obtained. The methodology used to collect the data was approved by the Portuguese Data Protection Commission and the study was approved by the Ethics Committees of the hospitals involved. Mean follow-up was 2 years.

This was a prospective, multicenter, observational study monitoring the occurrence of major adverse cardiovascular events (MACE) – death, stroke, transient ischemic attack, myocardial infarction, unstable angina, peripheral arterial disease, revascularization or renal failure. Follow-up of the patients consisted of annual assessments including, blood pressure (BP) measurement, laboratory tests, including serum lipids, and clinical observation. Total cholesterol, triglycerides and HDL cholesterol were measured. LDL cholesterol was obtained by Friedewald's formula (if triglycerides <3.39 mmol/l) or by ultracentrifugation. The LDL-Cholesterol/HDL-Cholesterol, Total Cholesterol/HDL-Cholesterol and Triglycerides/HDL-Cholesterol ratios were calculated in all patients. At each consultation, the subjects' weight and height were measured and body mass index (BMI) was calculated in kg/m2. Blood pressure and heart rate were measured in standard conditions, in a supine position and after a 10-minute resting period, by an experienced operator and using a clinically validated (class A) sphygmomanometer (Colson MAM BP 3AA1-2®; Colson, Paris) (Pereira & Maldonado, 2005).Three measurements were taken and the arithmetic mean was used in the analysis. All participants underwent routine fasting laboratory tests. At the first consultation they filled out a questionnaire concerning relevant personal and family history, smoking habits, alcohol consumption and medication.

Data from the sample subjects were processed using STATA for Windows, version 11.1. The distribution of the variables was tested for normality using the Kolmogorov-Smirnov test, and for homogeneity of variance by Levene's test. Simple descriptive statistics were used to characterize the sample and the distribution of variables. Cox proportional hazards analysis was used to determine the influence of the lipidic parameters on the occurrence of the specified cardiovascular events. C-Statistics was calculated to address the reliability of the lipidic parameters as prognostic variables.

Groups were compared using the χ2 test for categorical variables and the Student's t test (2 groups) or ANOVA with the post-hoc Tukey test (3 groups) for quantitative variables. A value of P≤0.05 was taken as the criterion of statistical significance for a 95% confidence interval.
