**2.3 Stanford Sleepiness Scale (SSS)**

122 Sleep Disorders

It is used especially in the assessment of type and of frequency about the tendency to fall asleep, micro-sleep episodes and monotony intolerance at work (especially in monitoring activities) and to capture the possibility of active participation in road traffic and other social

The *Epworth Sleepiness Scale* (ESS), the *Stanford Sleepiness Scale* (SSS), the *Multiple Sleep Latency Test* (MSLT) and the *Maintenance Wakefulness Test* (MWT) are among the methods that are most widely used for the investigation of daytime sleepiness in sleep disorders. The ESS reflects the global and subjective severity of daytime sleepiness in eight different situations and activities of daily living. The SSS is, however, to capture subjective circadian fluctuations of daytime sleepiness. To objective capture electrophysiological and standardized tests are often, such as the MSLT and the MWT used to determine the degree

If, on the basis of questionnaire data and medical history of sleeping on the basis of suspicion that a pathological daytime sleepiness (Table 1) exists, then objective analysis

The Epworth Sleepiness Scale (ESS) of Johns (1991) is very often used as a screening method for detecting the global daytime sleepiness and fall asleep in sleep disorders, especially used in hypersomnias. It is asked retrospectively, how high is the probability to fall asleep in eight everyday situations. The scale has a 4-step response format, in which values between 0 and 3 (0 = never to 3 = strongly agree) must be marked and results are added up a total

Following Johns (1991, 1992, Johns & Hocking 1997) a cut-off value ≥ 11 indicates a pathological daytime sleepiness. Standardization studies for the German-speaking countries were presented by Büttner et al. (2004c) and Sauter and colleagues (2007). The study found that 85% of healthy persons achieved a total value < 10, which corresponds to the calculated cut-off values in other studies (Johns 1991, Johns & Hocking 1997). The test-retest reliability of the ESS was calculated by Johns (1994) and based on a survey after five months in 87 healthy medical students. It was rtt = .82 (p <.001), even the quality of internal consistency

The ESS has in spite of it being subjective and a global assessment of daytime sleepiness (Johns 2000) has a very good validity. At a cut-off value > 10 it shows a high sensitivity of 93.5% and - high specificity 98.4%. The ESS is thus a highly reliable and valid procedure. The short implementation time and simple evaluation makes it very economical and cost effective. In addition, it can also be used for measuring the effectiveness of nCPAP therapy. Nevertheless the ESS does not lend itself to capture gradually different levels of sleepiness (Sangal et al. 1997b) and that four of the eight items have very low selectivity (Rühle et al.

Central nervous system activation

situations (Walsleben 1992, Weeß 2011).

of alertness on the basis of tonic activation.

Table 1. Sleepiness functions

maximum value of 24.

2005).

**2.2 Epworth Sleepiness Scale (ESS)** 

was confirmed (Cronbach's alpha = .88 (p <.001).

methods can be used to measure sleepiness-related functions.

Vigilance

Selected Attention Divided Attention The Stanford Sleepiness Scale (SSS) of Hoddes et al. (1973) is a scale on which momentary alertness can be assessed on a grading of 1 to 7 and thus serves to assess the circadian variations in daytime sleepiness. The scale describes gradual gradations of awareness; it varies between very alert and drowsy conditions. The alertness descriptions are also described, each with typical sensations (e.g. *some slack*, *slows*, *woozy*) characterized. Studies on the sensitivity of the scale showed that ratings in 15-minute intervals represent discrete changes in the degree of alertness. According to the response ratings point values are assigned for each time interval, which are then summated.

### **2.4 Multiple Sleep Latency Test (MSLT)**

The Multiple Sleep Latency Test by Carskadon and Dement (1977) recorded the sleep latency lying down and is recommended for the investigation of daytime sleepiness in OSAS patients in the ICSD-2. The MSLT is based on the assumption that a strong physiological sleepiness can reduce the sleep latency (Arand et al. 2005).

For a long time the MSLT has been considered a gold standard for the investigation of daytime sleepiness (Carskadon et al. 1986). The MSLT (as well as the Maintenance of Wakefulness Test (MWT)) is often used to determine the alertness with expert's investigations, e.g. to assess the driving ability (Poceta et al. 1992). Five times a day electrophysiological recordings (C3/A2, C4/A1, EOG, EMG) are performed in 2-hour intervals. The first time of measurement should be from 1.5 to 3 hours after waking. The patient lies in a darkened room and is asked to fall asleep. During the test procedure, the patient is monitored with a video recording.

A pathological fall asleep exists, when the medium sleep latency is < 5 minutes (Richardson et al. 1982). The gray area is between 5-10 minutes and > 10-20 minutes is a normal finding. But are also divergent standard values of 5-8 minutes; thereby establishing of normal values is equivalent to a kind of "rule of thumb" (Guilleminault et al. 1994, van den Hoed et al. 1981, Johns 2000). Although the MSLT perform and should be evaluated strictly according to objective criteria and standardized, it seems to have low implementation objectivity, because the results of individual tests vary greatly (Danker-Hopfe et al. 2006). As other reasons for the inconsistent individual test results Thorpy (1992) describes the different day times and measuring times and not objectified sleep deprivation and sedative or stimulating effects of drugs. In spite of these influences, however satisfactory test-retest reliabilities of rtt = .65 to .97 (van den Hoed et al. 1981, Zwyghuizen-Doorenbos et al. 1998) have been found.

Another problem of MSLT is the limited external generalization of daytime sleepiness in everyday situations (Johns 1994). The assumption that the MSLT describe daytime sleepiness - as reflection of everyday life - Johns (2000) keeps being wrong. As a predictor of MSLT is therefore not own, regardless how strict standards and criteria were met. In considering of the relationship between ESS and MSLT are unsatisfactory correlation of r = .27 (p <.001) or on those that are not significant (Mitler et al 1998.). Reasons for the inconsistent correlations are different: Either there are satisfactory (significant) correlations when all patients fell asleep in all MSLT times or when the patients rarely slept or not fell asleep (Chua et al. 1998).

### **2.5 Maintenance of Wakefulness Test (MWT)**

The Maintenance of Wakefulness Test of Poceta et al. (1992) examines the ability to stay awake in a sleep-inducing situation. The patient sits in a darkened room on a comfortable

The Effects of Sleep-Related Breathing Disorders on Waking Performance 125

daytime sleepiness in (Table 2). In a normal population (n = 349) between 20 and 60 years, was found a mean value for ln PUI of 1.50 ± 0.39 mm/min. Thus, abnormal values are obtained from ln PUI > 1.89 and pathological values from ln PUI > 2.28. The cut-off value of > 6.64 was found for 84.1% of a healthy sample (Wilhelm et al 2001), which was established that this is independent of gender and age (r = .85 to .94). The PUI correlated low, but significantly with the subjective estimates of daytime sleepiness in SSS (r = .29, p < .010). The implementation objectivity and evaluation objectivity seem to be sufficiently given, because the change in pupil size can be deliberately manipulated. The reliability was tested in healthy control subjects and is satisfactory (r = .64, p < .001) (Weeß et al. 2000).

**Value range Mean-2SD Mean-SD Mean Mean+SD Mean+2SD**  ln PUI (mm/min) 0.73 1.11 1.50 1.89 2.28 Percentile 2.3% 15.9% 50.0% 84.1% 97.7% PUI (mm/min) 2.07 3.05 4.50 6.64 9.80

In the first version of the *Reading test*, it was up to the patients and healthy controls, to select a passage according to their interests. Therefore, it was possible that the individual level of activation of OSAS patients may have influenced the excitement level of the books. For this reason, the story "One day, maybe one night" by Arnold Stadler (2003) was selected. This is a retrospective narrative. Due to the low excitement level of the narrative it was assumed

A B Fig. 2. In 2A is seen as the patient reads in a semi-recumbent position, the modified form of the story "One day, maybe one night" by Arnold Stadler (Fischer paperback 2003). In the face of the electrodes are glued EOG, EEG and the EMG and its right to recognize a

The text was justified, typed in the font "Times New Roman" and the size 12. The pages were not numbered and included 36 lines with 11 cm length. A lamp (40 watts) was used for lighting, placed at a distance of one meter above the patient's head. At the beginning of the *Reading test*, the patient was informed by a verbal instruction, to read the text as possible in the normal reading speed and without interruptions. Patients were asked to keep the book at a distance of 40 cm. Lack of vision and of reading ability has been excluded by

polysomnography. In **2B**, the patient is asleep and the book has resigned.

Table 2. Percentile of the normal reference range for ln PUI and PUI

that the degree of tonic activation would remain constant.

**2.7 Reading test (Fig. 2)** 

chair or on the bed and will be asked to refrain movements (e.g., grimacing, shaking), which may prevent falling asleep to refrain (Hartse et al. 1982, Mitler et al. 1982). Three to four times a day electrophysiological recordings (C3/A2, C4/A1, EOG and EMG) are recorded in 2-hour intervals of 20 minutes. The earliest start of the first test procedure should be scheduled two hours after waking. As with the MSLT test history is filmed with a video camera. Evaluated will be the sleep latency from the moment "light off" until the onset of the first two epochs of sleep stage 1 or 2.

In various standardization studies, inconsistent cut-off values were found from 13.5 to 18 minutes (Banks et al. 2004, Rühle 2005). Reasons for the different standard values according to Shreter et al. (2006) are that the test exercises have a significant influence on occasion staying awake in the test situation. So they provided proof that the sleep latency on the MWT was deliberately suppressed because the OSAS patients were afraid to get the license revoked. In considering the relationship between the MWT and ESS were calculated a satisfactory correlation of r = .48 (p < .001), with the common variance of the two devices was only 23% (Sangal et al. 1997b).

## **2.6 Pupillography (Fig. 1)**

The Pupillograph Sleepiness Test (PST) from Amtech (Weinheim) reflects the fatigue waves of the pupil described by Löwenstein. Normally, the pupil size will be constant in normal central nervous system activation in the dark for a long time. However, occur with increased daytime sleepiness after a few minutes spontaneous fluctuations (oscillations) on the pupil, which are recorded with infrared videography. Cause of fluctuations in pupil size is a mechanism of the autonomic nervous system. With reduced central nervous system activating two divisions acting simultaneously, which inhibit the Edinger-Westphal nucleus. This leads to instability of the central sympathetic activation and consequently fluctuating in an inhibition of parasympathetic activity and the Edinger-Westphal nucleus (Löwenstein et al. 1963, Yoss et al. 1970).

Fig. 1. Experimental setup for the pupillography. The patient wears an infrared protective goggle, has propped his chin on a device and looks toward the infrared camera.

#### **Evaluation**

The average Pupil Unrest Index (PUI) is the average pupil size fluctuations in millimetres per second over a period of 11 minutes. Higher PUI values indicate a clinically significant daytime sleepiness in (Table 2). In a normal population (n = 349) between 20 and 60 years, was found a mean value for ln PUI of 1.50 ± 0.39 mm/min. Thus, abnormal values are obtained from ln PUI > 1.89 and pathological values from ln PUI > 2.28. The cut-off value of > 6.64 was found for 84.1% of a healthy sample (Wilhelm et al 2001), which was established that this is independent of gender and age (r = .85 to .94). The PUI correlated low, but significantly with the subjective estimates of daytime sleepiness in SSS (r = .29, p < .010). The implementation objectivity and evaluation objectivity seem to be sufficiently given, because the change in pupil size can be deliberately manipulated. The reliability was tested in healthy control subjects and is satisfactory (r = .64, p < .001) (Weeß et al. 2000).


Table 2. Percentile of the normal reference range for ln PUI and PUI
