ADVERTISEMENT

Issue Navigator

Volume 10 No. 10
Earn CME
Accepted Papers
Classifieds





Scientific Investigations

The Oxford Sleep Resistance test (OSLER) and the Multiple Unprepared Reaction Time Test (MURT) Detect Vigilance Modifications in Sleep Apnea Patients

Anniina Alakuijala, M.D., Ph.D.1; Paula Maasilta, M.D., Ph.D.2; Adel Bachour, M.D., Ph.D.2
1Department of Clinical Neurophysiology, HUS Medical Imaging Center, Helsinki University Central Hospital, Finland, and Department of Neurological Sciences, University of Helsinki, Helsinki, Finland; 2Sleep Unit, Pulmonary Department, Helsinki University Central Hospital, Finland

ABSTRACT

Study Objectives:

The Oxford Sleep Resistance Test (OSLER) is a behavioral test that measures a subject's ability to maintain wakefulness and assesses daytime vigilance. The multiple unprepared reaction time (MURT) test measures a subject's reaction time in response to a series of visual or audible stimuli.

Methods:

We recruited 34 healthy controls in order to determine the normative data for MURT. Then we evaluated modifications in OSLER and MURT values in 192 patients who were referred for suspicion of sleep apnea. We performed OSLER (three 40-min sessions) and MURT (two 10-min sessions) tests at baseline. Of 173 treated OSA patients, 29 professional drivers were retested within six months of treatment.

Results:

MURT values above 250 ms can be considered abnormal. The OSLER error index (the number of all errors divided by the duration of the session in hours) correlated statistically significantly with sleep latency, MURT time, and ESS. Treatment improved OSLER sleep latency from 33 min 4 s to 36 min 48 s, OSLER error index from 66/h to 26/h, and MURT time from 278 ms to 224 ms; these differences were statistically significant.

Conclusions:

OSLER and MURT tests are practical and reliable tools for measuring improvement in vigilance due to sleep apnea therapy in professional drivers.

Citation:

Alakuijala A, Maasilta P, Bachour A. The Oxford Sleep Resistance Test (OSLER) and the multiple unprepared reaction time test (MURT) detect vigilance modifications in sleep apnea patients. J Clin Sleep Med 2014;10(10):1075-1082.


Patients with obstructive sleep apnea (OSA) usually demonstrate cognitive dysfunction both subjectively and objectively. Deficits in memory and in short and sustained attention have been attributed to daytime sleepiness due to fragmented sleep, while deficits in verbal functioning, problem-solving, and executive functions have been attributed to nocturnal hypoxemia.1,2 Vigilance and alertness are considered vital not only in traffic, but in many high-risk occupations as well. Furthermore, vigilance is fundamental to all other aspects of cognition.3

Excessive daytime sleepiness (EDS) is a subjective feeling of having difficulty remaining awake and fully alert. The most widely used subjective assessment of sleepiness is the Epworth Sleepiness Scale (ESS).4 However, ESS scores correlate rather poorly with objective measures, and expectations of benefit from treatment can clearly affect the resultant score.5 The most commonly used objective tools are the multiple sleep latency test (MSLT), which measures the subject's tendency to fall asleep if given the opportunity, and the maintenance of wakefulness test (MWT), which measures the subject's ability to stay awake in a soporific situation.6 Both tests are considered burdensome and expensive.

BRIEF SUMMARY

Current Knowledge/Study Rationale: Patients with obstructive sleep apnea usually demonstrate deficits in memory and attention both subjectively and objectively. The Oxford Sleep Resistance Test (OSLER) is a behavioral test that measures a subject's ability to maintain wakefulness and assesses daytime vigilance, and the multiple unprepared reaction time (MURT) test measures a subject's reaction time in response to a series of visual or auditory stimuli.

Study Impact: This study demonstrates that OSLER and MURT tests are practical and reliable tools for measuring improvement in vigilance due to CPAP therapy in professional drivers suffering from OSA. In the future, more attention ought to focus on safety in traffic and alertness-sensitive occupations, and screening for EDS in a simple and inexpensive way could be one of the keys to reaching this goal.

The Oxford Sleep Resistance Test (OSLER) was developed as a behavioral variant measuring a subject's ability to maintain wakefulness.7 The circumstances closely follow those of the MWT. A subject is instructed to remain awake and to respond by hitting a button on a portable device each time a dim light flashes at 3-second intervals during a 40-min period. Each session is terminated if the subject fails to respond to 7 subsequent illuminations, at which moment the sleep latency (SL) is set. If the subject does not make 7 successive errors during the test session, SL will be 40 min. In addition to SL, analysis of the error profile during the test, including 1 to 6 consecutive missed hits, was shown to reveal abnormal fluctuations in vigilance.8,13 The original test consisted of four 40-min sessions,7 but other researchers have since attempted shorter versions with 1, 2, or 3 sessions, or with 20-min sessions812; still other groups have followed the longer protocol.1315

The OSLER-2, an advanced version of the device, includes the opportunity to perform multiple unprepared reaction time (MURT) tests. The MURT test measures a subject's reaction time in response to a series of visual or auditory stimuli presented at random intervals between 1 and 10 second over a period of ten minutes. Normal values for the MURT are as yet unavailable.

In this study, we first report the OSLER-2 MURT data from a group of healthy subjects as a pilot study and then report the results of the OSLER and MURT tests of patients referred for suspicion of OSA before and after therapy.

METHODS

Healthy Subjects Study

We recruited 34 healthy adults (23 females) mainly among health care professionals and their family members. We excluded subjects who considered themselves sleepy, did shift work, used any kind of medication or substances that affect the central nervous system, had signs of any sleep disorder, or who reported sleeping poorly the night before the test. All the healthy subjects gave informed consent before they participated in the study. This group was small, and its gender distribution differed from our routine OSA patients. We aimed to determine the feasibility of the protocol and gain a rough idea of a cutoff value for normal MURT time, as it is as yet unpublished.

The subject was seated in an isolated normally lighted and video-monitored room and instructed to press a button with their dominant thumb when they heard or saw the stimulus on the screen of the portable device as quickly as possible while still maintaining accuracy. If they pressed the button without a stimulus, the hit was considered invalid. After they pressed the button, a millisecond counter appeared on the screen, providing instantaneous feedback. The stimuli appeared at random intervals between 1 and 10 sec over a period of 10 minutes. A total of about 100 stimuli were emitted for each MURT session. The accommodation test began between 09:00 and 11:00, followed by another test a few minutes later.

Patient Group Studies

We studied all patients referred to our Sleep Unit for suspicion of OSA who underwent an OSLER test during the study period (October 2009–December 2012), altogether 192 patients. Most of the patients were snorers and suffered from a variable degree of daytime somnolence. They were predominantly professional drivers (161 of them) or worked in other professions where dozing or falling asleep constitutes a significant risk for the subject and the environment. Our sleep unit is considered a tertiary referential center for sleep apnea and serves as a part of the University Hospital. Our ethics committee approved the study; because the study was based on documents completed during normally scheduled outpatient visits, no written informed consent was required.

Overnight Polygraphy

Patients used an Embletta Gold device (Embla, Denver, CO, USA) to perform the whole-night cardiorespiratory polygraphy at home before the OSLER and MURT tests. Respiratory parameters were scored manually according to AASM criteria,16 using the alternative rule for scoring hypopneas (i.e., a drop ≥ 50% in the nasal pressure signal amplitude together with a desaturation ≥ 3% from the pre-event baseline).

The Oxford Sleep Resistance (OSLER) Test

The subject was seated in a quiet, darkened, temperature-controlled room and was asked to respond to the flash of a dim light by hitting a button. The light-emitting diode flashed regularly for 1 sec every 3 seconds. The subject was instructed to remain awake for the maximum testing period of 40 minutes. When the subject failed to respond for 21 sec (i.e., 7 consecutive illuminations), the test ended with the assumption that he or she had fallen asleep. The first session began at 09:00 followed by 2 sessions at 2-h intervals, with lunch being served between the second and third sessions. The OSLER test was supervised with video monitoring from the adjacent room.

In addition to SL, we calculated the total number of missed hits (errors) for each session. Each missed hit was counted as one error. We obtained the OSLER error index by dividing the number of errors made in one session by the duration of the session (time spent awake) in hours.

The Multiple Unprepared Reaction Time (MURT) Test

The circumstances and instructions of the MURT test were identical to those for the healthy subjects described above. The first MURT session began at around 08:40 and served as an accommodation test. The second MURT test followed the first OSLER test session. The subjects were asked to leave the room and walk around for ten minutes to restore alertness. The median value of each test and the shortest median reaction time from both tests (“selected MURT”) were reported in milliseconds.

After analyzing the normal MURT values, we accepted the MURT values of the patients when the number of invalid hits per test did not exceed ten.

Other Data

The background information collected from each patient included gender, age, body mass index (BMI), Epworth Sleepiness Scale (ESS), and the age-adjusted Charlson comorbidity index (CCI).17,18 Parameters related to sleep disordered breathing included the apnea-hypopnea index (AHI), oxygen desaturation index of 4% (ODI4), mean and minimum values of oxygen saturation measured with an oximeter at a fingertip during the night (SpO2), and percentage of time at oxygen saturation below 90% (T < 90).

Continuous positive airway pressure therapy (CPAP) was proposed for all patients with OSA and daytime symptoms—a total of 173 patients of the initial cohort of 192 tested. CPAP initiation was performed with a sleep study control as described previously.19 Of those treated OSA patients, OSLER and MURT tests were repeated on 29 patients within 6 months of treatment, regardless of the mode of therapy. These patients were retested because they were professional drivers and had complained about considerable daytime somnolence at baseline; thus it was important to know if they still had daytime somnolence hazardous to their work environment. The new respiratory parameters were collected either from the Autoset-CPAP device or with the help of overnight polygraphy. Figure 1 shows the study profile.

A flowchart of the study.

MURT, multiple unprepared reaction time; OSA, obstructive sleep apnea; OSLER, Oxford Sleep Resistance Test.

jcsm.10.10.1075a.jpg

jcsm.10.10.1075a.jpg
Figure 1

A flowchart of the study. MURT, multiple unprepared reaction time; OSA, obstructive sleep apnea; OSLER, Oxford Sleep Resistance Test.

(more ...)

Statistics

Results were generated with a computerized statistical package (IBM SPSS Statistics 19.0, Armonk, NY, USA). We used the Student's t test for continuous variables, Fisher's exact test for categorical variables, and Spearman rank correlation for correlation analyses. All p values were two-sided, and the significance level was set at 0.05 throughout. For descriptive purposes, we report values as means and standard deviation (SD), and also as minimal and maximal values in selected cases.

RESULTS

Healthy Subjects

A total of 34 healthy adults (23 females) were recruited for the MURT normal group (Table 1). The mean of their median MURT was 204 ms in the first test and 190 ms in the second test; SDs were 44 ms and 30 ms, respectively. We noticed a statistically significant difference (p < 0.02) between the first and second MURT values.

Healthy subjects for the MURT validation study.

jcsm.10.10.1075.t01.jpg

table icon
Table 1

Healthy subjects for the MURT validation study.

(more ...)

The mean number of invalid hits was 2.6 (SD 2.5 and 3.9) for both tests. We therefore considered the test normal in the patients' group when the number of invalid hits did not exceed ten.

All Patients

During the study period, we carried out OSLER and MURT tests on 192 patients (18 females). Their characteristics appear in Table 2. Nineteen patients had an AHI < 5.0/h. Of 173 subjects with OSA, 37 had mild OSA (AHI 5.0-14.9/h), 56 had moderate OSA (AHI 15.0-29.9/h), and 80 had severe OSA (AHI ≥ 30.0/h). The mean OSLER SL among all untreated patients was 36 min 41 s, and the mean error index was 36.9/h. The mean value of the median MURT in the first test was 260 ms, and in the second test, 258 ms. The better median MURT time for each subject was on average 243 ms. One patient made > 10 invalid hits in both MURT test sessions, so his test results had to be rejected.

Baseline characteristics of all patients.

jcsm.10.10.1075.t02.jpg

table icon
Table 2

Baseline characteristics of all patients.

(more ...)

We found a strong positive and statistically significant correlation (r = 0.826, p < 0.01) between a high error index and a short SL. There was also a positive and significant correlation (r = 0.512, p < 0.01) between a high error index and a long MURT time, and between a short SL and a long MURT time (r = 0.436, p < 0.01). Further, the correlation between a high error index and high ESS scores was also positive and significant (r = 0.322, p < 0.01). We found no significant correlation between an error index and AHI, ODI4, or T < 90.

Gender Differences

Table 3 shows the differences between male and female patients. Women's AHI and ODI4 values were lower, but their ESS scores were higher than men's. Strikingly, OSLER SL, the error index, and median reaction time in the MURT test differed statistically significantly between women and men. Age, comorbities, and BMI did not differ between the groups.

All patients classified by gender.

jcsm.10.10.1075.t03.jpg

table icon
Table 3

All patients classified by gender.

(more ...)

Normal vs. Short Sleep Latency Groups

When all the patients were classified according to their inability to stay awake 3 times for 40 min, the error index was significantly higher, the median MURT time significantly longer, and the ESS scores significantly higher in the short SL subgroup than in the normal SL subgroup (Table 4). In addition, female and older subjects, as well as subjects with some comorbid illnesses had shorter OSLER sleep latencies. Values of AHI, ODI4, and T < 90 showed no statistically significant difference between the 2 subgroups.

All patients classified by OSLER sleep latency result.

jcsm.10.10.1075.t04.jpg

table icon
Table 4

All patients classified by OSLER sleep latency result.

(more ...)

Low vs. High Error Index

In order to examine the error profiles more closely, we then divided the patients by their error indexes and chose an arbitrary cutoff point of 10 errors per hour (Table 5). The subgroup with a higher error index had a shorter SL, longer MURT time, higher ESS scores, higher BMI, higher CCI, and higher ODI4 than did patients with an error index < 10/hour. All these differences were statistically significant. Here, the gender difference was even more prominent than between the normal and short SL groups, i.e., the proportion of female subjects in the high error index subgroup was much higher than their proportion in the low error index subgroup.

All patients classified by OSLER error index result.

jcsm.10.10.1075.t05.jpg

table icon
Table 5

All patients classified by OSLER error index result.

(more ...)

Treated Sleep Apnea Patients

All the OSA patients were offered a CPAP treatment. Among treated OSA patients, OSLER and MURT were re-tested in 29 patients within 6 months after initiation of CPAP therapy. These patients worked as professional drivers and they had had daytime somnolence before the treatment. By the time of the retest, 25 patients were using their CPAP device ≥ 4 h per night, but 4 subjects had already abandoned CPAP therapy. Two of these had received mandibular advancement devices, one was switched to bilevel ventilation therapy, and one patient managed to lose considerable weight. Neither the mode of therapy nor the adherence to CPAP had any clear effect to the results in the retest; thus we decided to look at all those retested patients as one group of treated sleep apnea patients.

The results appear in Table 6 and Figures 2 and 3. During the follow-up period, we noticed significant improvement in all the sleep study parameters and ESS scores with no significant change in BMI or medication. Moreover, the OSLER SL increased significantly from 33 min 4 s to 36 min 48 s, the error index decreased significantly from 66 to 26 errors/h, and the MURT time decreased significantly from 278 to 224 ms.

Treated patients.

jcsm.10.10.1075.t06.jpg

table icon
Table 6

Treated patients.

(more ...)

(A) Box-plot showing the OSLER sleep latency at baseline and within six months of OSA treatment. The number of sleep apnea patients at baseline and on treatment is 29. Each box shows the interquartile range, and the horizontal line within the box represents the median. The whiskers extend from the 10th to the 90th centiles. OSLER, Oxford Sleep Resistance Test. (B) Box-plot showing the OSLER error index at baseline and within six months of OSA treatment. OSLER, Oxford Sleep Resistance Test. (C) Box-plot showing the best selected MURT at baseline and within six months of OSA treatment. MURT, multiple unprepared reaction time.

jcsm.10.10.1075b.jpg

jcsm.10.10.1075b.jpg
Figure 2

(A) Box-plot showing the OSLER sleep latency at baseline and within six months of OSA treatment. The number of sleep apnea patients at baseline and on treatment is 29. Each box shows the interquartile range, and the horizontal line within the box represents the median. The whiskers extend from the...

(more ...)

Kaplan-Meier plot showing the percentage of patients staying awake during the 40-min OSLER test (mean of all 3 sessions) at baseline and within six months of OSA treatment.

jcsm.10.10.1075c.jpg

jcsm.10.10.1075c.jpg
Figure 3

Kaplan-Meier plot showing the percentage of patients staying awake during the 40-min OSLER test (mean of all 3 sessions) at baseline and within six months of OSA treatment.

(more ...)

DISCUSSION

To the best of our knowledge, this is the largest published sample of untreated OSA patients measured with both OSLER and MURT tests thus far. Different parameters of the disease severity and confounding factors were analyzed for the first time in a larger group.

According to the results of our pilot study for normal values, reaction times of less than 250 ms and fewer than ten invalid hits per test in the MURT with the OSLER-2 device are considered normal. Comparing the MURT results among the healthy subjects revealed that all of them had shorter reaction times in the latter test. Thus, they could benefit from the accommodation; for this reason, we recommend routinely carrying out two MURT sessions.

In contrast, both untreated and treated OSA patients failed to improve their reaction time from the first MURT session to the second MURT test. The patients may still have been sleepy after the soporific OSLER test circumstances, even though we tried to avoid any drowsiness by providing a ten-minute break between the tests for them to stand up and walk a little in the lab. It has been reported that sleep deprivation enhances the time-on-task effect,3 and that the subjective sleepiness of sleep deprived subjects is higher in the end of a test than in the beginning.20 Thus, untreated and even treated OSA patients may tire more easily than healthy subjects from the sustained effort of concentration required for the OSLER test. The alertness of OSA patients may fluctuate more during the day.21 Regardless of our finding, however, the evidence suggests that two MURT sessions are needed to obtain reliable results.

MURT is quite similar to the psychomotor vigilance test (PVT), the most established assay of vigilant attention in sleep medicine.22 There, visual stimuli occur randomly at intervals ranging from 2 to 10 s, and the standard test lasts 10 min. Response speed and the number of lapses proved to be the most sensitive PVT metrics for distinguishing sleepy from alert subjects.23 The advantages of MURT are that the stimuli are both visual and auditory and that it comes with OSLER-2.

The other main finding in our study was that treatment for OSA improved OSLER and MURT test values: SL, the error index, and reaction time. In other words, successful therapy not only ameliorated controlled sleep study parameters, but also resulted in objectively improved reaction times and sustained attention as measured with the MURT and OSLER tests, together with reduced subjective sleepiness as measured with ESS. Our finding is consistent with that of Craig et al.,12 who showed that CPAP reduced both subjective and objective daytime sleepiness as measured with ESS and OSLER, respectively. Previously, two other groups also reported similar improvements with CPAP therapy.10,24

Studies of cognitive changes pre-/post-treatment have not consistently demonstrated the superiority of CPAP over oral placebo, sham CPAP, and behavioral treatments.25 Vigilance and executive functioning, however, proved to be more affected than verbal and intellectual functioning in untreated OSA patients; thus testing for vigilance may be the most sensitive way to measure the effects of treatment.26,27

Another finding was that women in our study had lower AHI, but were clearly sleepier than men, not only subjectively, but also objectively, according to the OSLER and MURT results. Women have long been known to have slightly slower reaction times to visually applied stimuli than men,28 but differences in OSLER test results have not previously been published. Women have been shown repeatedly to suffer more from EDS than do men with similar AHI.29 AHI is not the most sensitive indicator of the severity of sleep disordered breathing,30 especially when using polygraphy instead of polysomnography,31 as we did in our study.

Among our untreated OSA patients, a short OSLER SL, a high error index and a long MURT time correlated significantly, thus highlighting the high reliability and validity of the OSLER and MURT tests. These objective measurements also significantly correlated with subjective assessments (ESS), which is consistent with the results of Krieger et al.,9 but inconsistent with those of Ferreira et al.14 No strong correlation between MWT and ESS has been found previously.32,33 In contrast to ESS, AHI and the other measures related to sleep disordered breathing showed no correlation with the OSLER or MURT results in our study. Many untreated OSA patients can cope well with fragmented sleep and do not suffer from EDS.34

The main criticism of the OSLER test is its dependence on patient cooperation.9 By intentionally closing one's eyes and omitting the hits, one can pretend to be unable to stay awake. Meanwhile, cheating is also possible in MWT, especially without preceding polysomnography. All our patients wanted to keep their driver's license and tried their best to stay awake during the tests, so we did not find cheating to be a problem.

Limitations

Our study has certain limitations. We diagnosed OSA based on unattended polygraphy at home rather than in-lab polysomnography. Yet, recent research has demonstrated the essential equivalency between these approaches.35 We did not perform MWT or PVT tests to strengthen our results, as the study was carried out as routine clinical work. In addition, the number of OSA patients retested after the initiation of the treatment was small, but we could still notice statistically significant improvements in the OSLER and MURT test results. All our patients were of Caucasian origin (as is the vast majority of the population in our country). As for the MURT normative data study, our control group was rather small and young, and comprised predominantly women, unlike our routine OSA patients. According to our results among patients, female gender and higher age associate with slower MURT reaction time. Thus, this part of the study must be considered preliminary and further research is needed in order to provide population-based predicted values for MURT.

CONCLUSIONS

We conclude that the OSLER and MURT tests are practical and reliable tools for measuring improvement in vigilance due to CPAP therapy in professional drivers suffering from OSA. In the future, more attention ought to focus on safety in traffic and alertness-sensitive occupations, and screening for EDS in a simple and inexpensive way could be one of the keys to reaching this goal.

DISCLOSURE STATEMENT

This was not an industry supported study. The authors have indicated no financial conflicts of interest. All the authors declare that they engaged in no off-label or investigational use of any medical products.

ABBREVIATIONS

AASM

American Academy of Sleep Medicine

AHI

apnea-hypopnea index

BMI

body mass index

CCI

Charlson comorbidity index

CI

confidence interval

CPAP

continuous positive airway pressure therapy

EDS

excessive daytime sleepiness

ESS

Epworth Sleepiness Scale

MSLT

multiple sleep latency test

MURT

multiple unprepared reaction time

MWT

maintenance of wakefulness test

ODI4

oxygen desaturation index of four percentage units

OSA

obstructive sleep apnea

OSLER

Oxford Sleep Resistance Test

PVT

psychomotor vigilance test

SL

sleep latency

SpO2

oxygen saturation measured with an oximeter at a fingertip

T < 90

percentage of time at an oxygen saturation below 90%

ACKNOWLEDGMENTS

The authors dedicate this work to the late Tarja Harpala. We thank Leena Petman for valuable help during the study.

REFERENCES

1 

Décary A, Rouleau I, Montplaisir J, authors. Cognitive deficits associated with sleep apnea syndrome: a proposed neuropsychological test battery. Sleep. 2000;23:369–81. [PubMed]

2 

Ayalon L, Ancoli-Israel S, Aka AA, et al., authors. Relationship between obstructive sleep apnea severity and brain activation during a sustained attention task. Sleep. 2009;32:373–81. [PubMed Central][PubMed]

3 

Lim J, Dinges DF, authors. Sleep deprivation and vigilant attention. Ann N Y Acad Sci. 2008;1129:305–22. [PubMed]

4 

Johns MW, author. A new method for measuring daytime sleepiness: the Epworth sleepiness scale. Sleep. 1991;14:540–5. [PubMed]

5 

Crawford MR, Bartlett DJ, Coughlin SR, et al., authors. The effect of continuous positive airway pressure usage on sleepiness in obstructive sleep apnoea: real effects or expectation of benefit? Thorax. 2012;67:920–4. [PubMed]

6 

Littner MR, Kushida C, Wise M, et al., authors. Practice parameters for clinical use of the multiple sleep latency test and the maintenance of wakefulness test. An American Academy of Sleep Medicine report. Sleep. 2005;28:113–21. [PubMed]

7 

Bennett LS, Stradling JR, Davies JO, authors. A behavioral test to assess daytime sleepiness in obstructive sleep apnoea. J Sleep Res. 1997;6:142–5. [PubMed]

8 

Mazza S, Pépin JL, Deschaux C, et al., authors. Analysis of error profiles occurring during the OSLER test: a sensitive mean of detecting fluctuations in vigilance in patients with obstructive sleep apnea syndrome. Am J Respir Crit Care Med. 2002;166:474–8. [PubMed]

9 

Krieger AC, Ayappa I, Norman RG, et al., authors. Comparison of the maintenance of wakefulness test (MWT) to a modified behavioral test (OSLER) in the evaluation of daytime sleepiness. J Sleep Res. 2004;13:407–11. [PubMed]

10 

Nussbaumer Y, Bloch KE, Genser T, et al., authors. Equivalence of autoadjusted and constant continuous positive airway pressure in home treatment of sleep apnea. Chest. 2006;129:638–43. [PubMed]

11 

Vennelle M, White S, Riha RL, et al., authors. Randomized controlled trial of variable-pressure versus fixed-pressure continuous positive airway pressure (CPAP) treatment for patients with obstructive sleep apnea/hypopnea syndrome (OSAHS). Sleep. 2010;33:267–71. [PubMed Central][PubMed]

12 

Craig SE, Kohler M, Nicoll D, et al., authors. Continuous positive airway pressure improves sleepiness but not calculated vascular risk in patients with minimally symptomatic obstructive sleep apnoea: the MOSAIC randomized controlled trial. Thorax. 2012;67:1090–6. [PubMed]

13 

Priest B, Brichard C, Aubert G, et al., authors. Microsleep during a simplified maintenance of wakefulness test: A validation study of the OSLER test. Am J Respir Crit Care Med. 2001;163:1619–25. [PubMed]

14 

Ferreira JJ, Neutel D, Mestre TA, et al., authors. Usefulness of OSLER test in Parkinson's disease. Sleep Med. 2013;14:297–8. [PubMed]

15 

Idris I, Abdulla H, Tilbrook S, et al., authors. Exenatide improves excessive daytime sleepiness and wakefulness in obese patients with type 2 diabetes without obstructive sleep apnoea. J Sleep Res. 2013;22:70–5. [PubMed]

16 

Iber C, Ancoli-Israel S, Chesson AL, Quan S, authors. The AASM manual for the scoring of sleep and associated events, rules, terminology, and technical specifications. 1st ed. Westchester, IL: American Academy of Sleep Medicine, 2007.

17 

Charlson ME, Pompei P, Ales KL, et al., authors. A new method of classifying prognostic comorbidity in longitudinal studies: development and validation. J Chronic Dis. 1987;40:373–83. [PubMed]

18 

Charlson M, Szatrowski TP, Peterson J, et al., authors. Validation of a combined comorbidity index. J Clin Epidemiol. 1994;47:1245–51. [PubMed]

19 

Bachour A, Virkkala JT, Maasilta PK, authors. AutoCPAP initiation at home: optimal trial duration and cost-effectiveness. Sleep Med. 2007;8:704–10. [PubMed]

20 

Sallinen M, Onninen J, Tirkkonen K, et al., authors. Effects of cumulative sleep restriction on self-perceptions while multitasking. J Sleep Res. 2013;22:273–81. [PubMed]

21 

Manly T, Robertson IH, Galloway M, et al., authors. The absent mind: further investigations of sustained attention to response. Neuropsychologia. 1999;37:661–70. [PubMed]

22 

Wilkinson RT, Houghton D, authors. Field test of arousal: a portable reaction timer with data storage. Hum Factors. 1982;24:487–93. [PubMed]

23 

Basner M, Dinges DF, authors. Maximizing sensitivity of the psychomotor vigilance test (PVT) to sleep loss. Sleep. 2011;34:581–91. [PubMed Central][PubMed]

24 

West SD, Kohler M, Nicoll DJ, et al., authors. The effect of continuous positive airway pressure treatment on physical activity in patients with obstructive sleep apnoea: A randomised controlled trial. Sleep Med. 2009;10:1056–8. [PubMed]

25 

Sánchez AI, Martínez P, Miró E, et al., authors. CPAP and behavioral therapies in patients with obstructive sleep apnea: effects on daytime sleepiness, mood, and cognitive function. Sleep Med Rev. 2009;13:223–33. [PubMed]

26 

Beebe DW, Groesz L, Wells C, et al., authors. The neuropsychological effects of obstructive sleep apnea: a meta-analysis of norm-referenced and case-controlled data. Sleep. 2003;26:298–307. [PubMed]

27 

Lim W, Bardwell WA, Loredo J, et al., authors. Neuropsychological effects of 2-week continuous positive airway pressure treatment and supplemental oxygen in patients with obstructive sleep apnea: a randomized placebo-controlled study. J Clin Sleep Med. 2007;3:380–6. [PubMed Central][PubMed]

28 

Bellis CJ, author. Reaction time and chronological age. Proc Soc Exp Biol Med. 1933;30:801.

29 

Ye L, Pien GW, Weaver TE, authors. Gender differences in the clinical manifestations of obstructive sleep apnea. Sleep Med. 2009;10:1075–84. [PubMed]

30 

Anttalainen U, Polo O, Vahlberg T, et al., authors. Women with partial upper airway obstruction are not less sleepy than those with obstructive sleep apnea. Sleep Breath. 2013;17:873–6. [PubMed]

31 

Berry RB, Budhiraja R, Gottlieb DJ, et al., authors. Rules for scoring respiratory events in sleep: Update of the 2007 AASM manual for the scoring of sleep and associated events. J Clin Sleep Med. 2012;8:597–619. [PubMed Central][PubMed]

32 

Sangal RB, Sangal JM, Belisle C, authors. Subjective and objective indices of sleepiness (ESS and MWT) are not equally useful in patients with sleep apnea. Clin Electroencephalogr. 1999;30:73–5. [PubMed]

33 

Sauter C, Asenbaum S, Popovic R, et al., authors. Excessive daytime sleepiness in patients suffering from different levels of obstructive sleep apnoea syndrome. J Sleep Res. 2000;9:293–301. [PubMed]

34 

Kapur VK, Baldwin CM, Resnick HE, et al., authors. Sleepiness in patients with moderate to severe sleep-disordered breathing. Sleep. 2005;28:472–7. [PubMed]

35 

Rosen CL, Auckley D, Benca R, et al., authors. A multisite randomized trial of portable sleep studies and positive airway pressure autotitration versus laboratory-based polysomnography for the diagnosis and treatment of obstructive sleep apnea: the HomePAP Study. Sleep. 2012;35:757–67. [PubMed Central][PubMed]