Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2022 Apr 1.
Published in final edited form as: J Sleep Res. 2020 May 22;30(2):e13092. doi: 10.1111/jsr.13092

Interhemispheric sleep depth coherence predicts driving safety in sleep apnea

Ali Azarbarzin 1,2, Magdy Younes 3, Scott A Sands 1,2, Andrew Wellman 1,2, Susan Redline 1,2, Charles A Czeisler 1,2, Daniel J Gottlieb 1,2,4
PMCID: PMC7680278  NIHMSID: NIHMS1612246  PMID: 32441843

Summary

Obstructive sleep apnea is associated with increased risk of car crashes; however, conventional measures of sleep apnea severity do not clearly identify those individuals who are at greatest crash risk. Here we tested whether, among individuals with sleep apnea, those with reduced interhemispheric sleep depth coherence, measured by correlation between right and left hemisphere odds ratio product, are at greater risk. The sample was derived from the Sleep Heart Health Study, a prospective observational cohort study, and included 1378 adults with sleep apnea. The occurrence of car crash was ascertained by a questionnaire administered 2 years after the sleep study, which asked about the occurrence of crashes during the year prior to questionnaire administration. We computed the sleep depth coherence from electroencephalograms recorded during baseline sleep studies and after 5 years. The weighted kappa coefficient and Bangdiwala’s B were 0.34 and 0.59, respectively, indicating a fair to moderate stability over a 5-year interval. Multivariate logistic regression, adjusted for age, sex, race, body mass index, and miles driven per year, was used to assess the risk of car crash. Compared to the lowest quartile of sleep depth coherence (<0.86), individuals in the highest quartile (>0.93) had a 62% (95% CI: 22%−81%) lower risk of accident. Further adjustments for usual sleep duration and sleepiness did not meaningfully alter these findings. Higher interhemispheric sleep depth coherence is associated with significantly lower risk of motor vehicle crashes in individuals with sleep apnea. This suggests that high interhemispheric sleep depth coherence may be a marker of resistance to sleep apnea-related adverse neurocognitive outcomes.

Keywords: car crash, interhemispheric sleep depth, odds ratio product, apnea-hypopnea index, sleepiness, neurocognitive outcomes

INTRODUCTION

Obstructive Sleep Apnea (OSA) is a common disorder(Young et al., 1993) associated with adverse cardiometabolic(Nieto & Pickering, 2000; Redline et al., 2010) and neurocognitive(Engleman, Kingshott, Martin, & Douglas, 2000; Reynolds & Banks, 2010) consequences, with an estimated prevalence rate of 17% in women and 22% in men(Franklin & Lindberg, 2015), with the prevalence increasing in elderly populations(Heinzer et al., 2015; Peppard et al., 2013). OSA is associated with increased risk of motor vehicle crash (MVC)(Gottlieb, Ellenbogen, Bianchi, & Czeisler, 2018; Sassani et al., 2004; Tregear, Reston, Schoelles, & Phillips, 2009), even among those who do not report excessive daytime sleepiness(Gottlieb, et al., 2018). Additionally, OSA has been linked to other neurocognitive impairments, including poor attention, impaired memory, sleepiness and fatigue(Kielb, Ancoli-Israel, Rebok, & Spira, 2012). While the underlying neural mechanisms responsible for these impairments remain unclear, past research indicates that the repeated nocturnal hypoxemia(Azarbarzin et al., 2019) and arousals associated with obstructive respiratory events are likely to be the key factors affecting neural function in OSA(Colt, Hass, & Rich, 1991; Li & Veasey, 2012; Tsai, 2010). Failure of the apnea-hypopnea index to adequately describe these physiological processes may contribute to the observed heterogeneity in neurocognitive impairment for a given OSA severity.

There is growing evidence that OSA is associated with regional reductions in sleep electroencephalography (EEG) power(Jones et al., 2014), topography-specific alterations in spindle frequency(Schonwald, Carvalho, de Santa-Helena, Lemke, & Gerhardt, 2012), localized cortical thinning in brain(Joo, Jeon, Kim, Lee, & Hong, 2013), and reduced fMRI-measured brain connectivity(Canessa et al., 2018; Santarnecchi et al., 2013). Moreover, repetitive periods of obstructive breathing may be associated with reduced functional interhemispheric connectivity(Rial et al., 2013), unlike the normal pattern in humans, in whom sleep generally occurs bihemispherically(Rattenborg, Amlaner, & Lima, 2000). Younes and colleagues have developed and validated a measure of sleep depth, called the odds ratio product (ORP), which is a continuous index of sleep depth derived from the surface EEG with a range of 0 (deep sleep) to 2.5 (full wakefulness)(Younes & Hanly, 2016; Younes et al., 2015). This measure can be derived independently for each hemisphere. In a recent study, a substantially reduced interhemispheric sleep depth coherence (measured by the correlation between right and left ORP (R/L ORP)) was observed in mechanically ventilated patients who failed a spontaneous breathing trial(Dres et al., 2019).

Given the report that OSA is associated with altered interhemispheric functional connectivity(Rial, et al., 2013), we further explored the functional implications of interhemispheric sleep depth coherence by examining the association of R/L ORP with motor vehicle crash risk in individuals with OSA in a community-dwelling cohort of middle-aged and older adults.

METHODS

Study Sample

The primary sample included participants in the baseline (1995–1998) and 2-year follow-up examinations of the Sleep Heart Health Study (SHHS), a community-based, prospective cohort study designed to examine the cardiovascular outcomes of sleep-disordered breathing in middle-aged and older adults (Figure 1) (Quan et al., 1997; Redline et al., 1998). Baseline polysomnograms (PSG) were available from 5793 of the 6441 SHHS participants, through the National Sleep Research Resource website (www.sleepdata.org). A total of 2626 had an AHI≥5 events/hour. After excluding those with missing data on age, sex BMI, or race (n=23), missing driving status (n=365), non-drivers (n=224), and those missing miles driven per year (n=547) or motor vehicle accident history (n=17), and those in whom R/L ORP could not be measured (n=72), there were 1378 individuals with OSA included in the primary analysis (Figure 1). For assessing stability and reliability of R/L ORP, we also extracted this metric from PSGs performed as part of the SHHS 5-year follow-up examination, performed between 2001–2003. Of the 1378 individuals included in the primary sample, the R/L ORP data were available in 675 individuals who returned for the follow-up examination.

Figure 1:

Figure 1:

Open in a new tab

Study sample

Polysomnography

During both baseline and 5-year follow-up examinations, participants underwent unattended in-home polysomnography (PSG), as previously described(Quan, et al., 1997; Redline, et al., 1998). All PSGs were scored at a central sleep reading center. Respiratory events identified using thermistry and inductance plethysmography were annotated according to level of associated oxygen desaturation and presence or absence of EEG-based arousal. In this report, the AHI was calculated using apneas and hypopneas each associated with oxygen desaturations ≥4%.

Driving History

A follow-up questionnaire was administered approximately 2 years after the baseline sleep study. The question “Do you drive?” was asked to ascertain driving status and followed by a series of questions about driving frequency, including average time spent driving per day or miles driven per year. The occurrence of motor vehicle crashes was ascertained by the question: “How many accidents have you had in the last year while you were the driver?”

Interhemispheric Sleep Depth Coherence (R/L ORP)

The method of measuring ORP was described in detail elsewhere(Younes, et al., 2015). Briefly, fast Fourier transform is performed on a central EEG derivation in 3-second consecutive intervals. Power in each of four frequency ranges is calculated and assigned a rank (0–9) depending on its location within the entire range of values observed in the specific frequency range in a large sample of clinical PSG records. The four ranks form a 4-digit number, referred to as bin number, which describes the relative powers in the four frequency ranges. For example, a bin number of 9710 describes a 3-second epoch with very high power in the slowest frequency range, moderately high power in the next frequency range and very low power in the two highest frequency ranges. The probability of each of the possible 10,000 bins occurring during wake epochs or during arousals was determined during the development phase using a large number of clinical PSGs scored manually by three highly experienced scorers(Younes, et al., 2015). Probability (0 to 100%) is divided by 40 (% of epochs scored wake in development files), resulting in a range from 0 (never occurs during wakefulness or in arousals) to 2.5 (never occurs during epochs scored asleep).

For determining the R/L sleep depth coherence, ORP was measured in the right and left central electrodes (C3-M2, C4-M1). The 10 values in each 30-second epoch were averaged. Agreement between the 30-second averages of the right and left signals was determined by intra-class correlation. All epochs across the night were included in the correlation analysis after excluding epochs with technically inadequate signals.

Statistical Analysis

Distributions of covariates and sleep measures (mean (standard deviation), mean (95% Confidence Interval) or median [Q1, Q3]) were summarized and stratified by the quartiles of R/L ORP. Due to an observed threshold effect in a previous study(Dres, et al., 2019), the R/L ORP was categorized based on its quartiles with the first quartile (individuals with lowest interhemispheric coherence) as the reference group. To assess the stability of R/L ORP over the 5 years between PSGs, quartile assignment was compared using the weighted Cohen’s kappa statistic(Cohen, 1960) and Bangdiwala’s B statistic(Bangdiwala, Haedo, Natal, & Villaveces, 2008). The weighted kappa coefficient and Bangdiwala’s B were 0.34 and 0.59, respectively, indicating a fair to moderate stability(Bangdiwala & Shankar, 2013) over a 5-year interval.

Multivariable logistic regression analysis was used to assess the associations between R/L ORP quartiles and the occurrence of any motor vehicle crash during the year preceding the follow-up visit in individuals with OSA. This model was adjusted for covariates(Gottlieb, et al., 2018), including age, sex, race, BMI, and miles driven per year (Model 1). A likelihood ratio test was used to compare Model 1 with a baseline model including only covariates. Statistical significance was accepted at a p-value of <0.05. In a sensitivity analysis, Model 1 was further adjusted for usual sleep duration, Epworth sleepiness score, and AHI (Model 2) and the adjusted odds ratios (aOR) were calculated and compared. All the statistical analyses were conducted using the R statistical package (http://www.r-project.org).

RESULTS

The characteristics of participants in the analytic sample, grouped by quartiles of R/L ORP are shown in Table 1. Compared to lowest quartile of R/L ORP (lowest sleep depth coherence), those in the top quartile were generally older, more likely to be women, less overweight, less sleepy, and had lower AHI (Table 1). The unadjusted occurrence of MVC over the two years following PSG was significantly lower in those in the highest quartile of sleep depth coherence than in those in the lowest quartile of sleep depth coherence (3.6% vs. 9.4%, p<0.01).

Table 1:

Summary characteristics by quartiles of interhemispheric sleep depth coherence

R/L ORP Quartiles
Q1
(0.12–0.86)		 Q2
(0.87–0.90)		 Q3
(0.91–0.93)		 Q4
(0.94–0.98)
Age (year) 62.4 (10.2) 63.9 (9.68) 64.7 (9.50) 67.5 (8.84)
Sex
 Male 264 (68.9%) 226 (71.5%) 257 (68.4%) 185 (61.1%)
 Female 119 (31.1%) 90 (28.5%) 119 (31.6%) 118 (38.9%)
BMI (kg/m2) 30.8 (5.76) 30.0 (4.81) 29.3 (4.78) 29.5 (5.08)
Race:
 White 346 (90.3%) 286 (90.5%) 347 (92.3%) 282 (93.1%)
 Black 22 (5.74%) 20 (6.33%) 16 (4.26%) 16 (5.28%)
 Other 15 (3.92%) 10 (3.16%) 13 (3.46%) 5 (1.65%)
Apnea Hypopnea Index (events/hour) 17.4 (14.3) 16.5 (13.9) 17.0 (13.1) 14.7 (11.6)
Epworth Sleepiness Scale 8.26 (4.68) 8.61 (4.48) 8.32 (4.39) 7.31 (3.86)
Usual Sleep duration (hour) 7.08 (1.11) 7.13 (1.14) 7.03 (1.12) 7.17 (1.16)
Miles driven per year (thousands) 11.0 (7.96) 11.4 (12.7) 10.6 (9.60) 8.96 (7.26)
Open in a new tab

R/L ORP, interhemispheric sleep depth coherence quantifying correlation between odds ratio product (ORP) obtained from right and left central electroencephalograms; BMI, body mass index.

A likelihood ratio test revealed a statistically significant improvement in model fit when R/L ORP was added to a multivariable logistic regression model (with MVC as outcome) that included age, sex, race, BMI, and miles driven per year (p<0.05). In this model (Model 1), the odds of having a MVC were 62% (95% CI: 22%−81%) lower for the highest quartile than for the lowest quartile of R/L ORP (aOR 0.38 (95% CI: 0.19, 0.78), Figure 2), while the odds ratios of MVC were not significantly different from the lowest quartile for the second (aOR 0.81 for Q2, Figure 2) and third (aOR 0.91 for Q3, Figure 2) quartiles. Further adjustments for usual sleep duration, Epworth Sleepiness Scale score, and AHI did not materially change these findings (Figure 2).

Figure 2:

Figure 2:

Open in a new tab

The association between motor vehicle crash (MVC) and R/L ORP (quartiles) in individuals with sleep apnea after adjusting for age, sex, race, BMI, and miles driven per year in Model 1. Model 2 was further adjusted for usual sleep duration, Epworth Sleepiness Scale score, and apnea-hypopnea index. Note that the lowest quartile was chosen as the reference. AHI, apnea-hypopnea index (apneas and hypopneas with ≥4% desaturations).

DISCUSSION

The presence of high interhemispheric sleep depth coherence (R/L ORP ≥ 0.93) among individuals with OSA was significantly associated with reduced MVC risk in this community-dwelling cohort of middle-aged and older adults. While caution is needed in drawing causal inferences from observational data, these findings suggest that high interhemispheric sleep depth coherence may be a marker of reduced susceptibility to OSA-related adverse neurocognitive outcomes. In a previous small study, OSA subjects (n=15) were found to have a lower central interhemispheric agreement in sleep depth than controls(Saunamaki, Jehkonen, Huupponen, Polo, & Himanen, 2009). Similarly, several studies concerning OSA and its neurocognitive outcomes reported a reduction in interhemispheric coherence, quantified by both EEG (power(Jones, et al., 2014), spindle frequency(Schonwald, et al., 2012), and sleep depth(Saunamaki, et al., 2009)) and fMRI characteristics(Canessa, et al., 2018; Santarnecchi, et al., 2013).To our knowledge this is the first time that the agreement in interhemispheric sleep depth has been examined in relation to driving safety. The findings may reflect the sensitivity of the R/L ORP to the OSA-related stressors that result in neuronal dysfunction and attentional lapses that associated with MVCs.

R/L ORP Threshold and Reproducibility

The protective effect noted in this study was only observed in the highest quartile of R/L ORP, with a very high average correlation of 0.93, suggesting that even minor reductions in R/L ORP may be associated with increased risk of car crashes. By contrast, in critically ill patients, most patients with R/L ORP >0.7 passed the weaning trial whereas most patients with a lower correlation failed. (Dres, et al., 2019). This discrepancy could be due in part to the difference in patient populations (community-dwelling adults vs. ICU patients) or to the very different outcomes of interest, with differences in driving vigilance being sensitive to a much milder impairment of sleep depth coherence than is needed to affect respiratory function. Sleep depth coherence may also vary from day to day, reflecting changes in sleep duration or the well-recognized night-to-night variability in OSA severity(Sforza, Roche, Chapelle, & Pichot, 2019). In the current study, where R/L ORP was determined 1–2 years before the car crash, R/L ORP may have been less reflective of the value in the night preceding the car crash than in the ICU study where R/L ORP was determined immediately before the weaning trial. In this regard, the findings on reproducibility of R/L ORP over a 5-year interval are noteworthy, showing that the higher an individual’s R/L ORP at the first visit, the less likely that individual was to move to the lowest quartile in the second visit. It is possible that greater day-to-day variability in sleep depth coherence of the individuals in the lower quartiles results in a greater number of days with severe impairment, on which they may be at increased MVC risk.

Strengths and Limitations

There are several notable strengths of this study, including the use of an objective, automated process for generation of the R/L ORP measure. As EEG signals are readily available from polysomnography performed in clinical sleep laboratories, this would facilitate clinical implementation of interhemispheric sleep depth coherence measurements, should the functional importance of these measures be validated. Another strength of this study is that the analytical sample was obtained from a large community-dwelling cohort, which is more generalizable to the public than a cohort of clinically-referred individuals. In addition, baseline sleep apnea assessment was performed prospectively and not biased by the motor vehicle crash history.

However, the study had several limitations, including the under-representation of minority populations and young individuals. In addition, residual confounding cannot be excluded due to the observational nature of the study. For this reason, any mechanistic insights into the relation of interhemispheric differences in individuals with OSA to crash risk is speculative. However, unihemispheric sleep is widely seen in aquatic species and birds, while operating under the conditions when bihemisperic sleep may be dangerous (Rattenborg, et al., 2000). Although speculative, it is possible that this mechanism is activated in OSA subjects under conditions (e.g. hypoxemia) in which natural sleep is deemed unsafe(Dres, et al., 2019). It is also possible that reduced interhemispheric coherence is a result of interhemispheric differences in the autonomic(Takase et al., 2004) or EEG arousal(Amatoury et al., 2016; Azarbarzin, Ostrowski, Hanly, & Younes, 2014; Azarbarzin et al., 2015) responses to respiratory events. However, it is likely that in individuals with OSA, low interhemispheric sleep depth coherence is a result of sleep deprivation due to frequent respiratory events. Unpublished observations by one of the authors (MY) demonstrate a deterioration of R/L ORP in healthy individuals who were 1) sleep deprived for 36 hours, 2) restricted to 5 hours of sleep for 4 consecutive nights, or 3) exposed to frequent application of noise transients during sleep. Regardless of mechanism, individuals with OSA who maintain high levels of interhemispheric sleep depth coherence despite the presence of OSA may be less vulnerable to neurocognitive effects of sleep disturbance.

Another potential limitation is that self-report measures of driving and crash history were used. While the gold standard is using police reports, as discussed in our previous study(Gottlieb, et al., 2018), the self-reported rate of motor vehicle crashes in this study is similar to the rate reported by the AAA Foundation for Traffic Safety for U.S. adults(Gottlieb, et al., 2018; Tefft, 2012). Moreover, past studies have shown that self-reporting accurately reflects the accidents occurring within the prior year. Indeed, in elderly drivers (≥70 years), the agreement between self-report and state-recorded motor vehicle crashes was substantial (kappa = 0.64) (Singletary et al., 2017). Among younger drivers, the agreement between self- and police-reported crashes was 85% (Boufous et al., 2010).

Although the SHHS used a conservative definition of AHI, requiring both apneas and hypopneas be associated with at least 4% desaturation, a sensitivity analysis in which all respiratory events associated with either a 3% or greater desaturation or an arousal(Kapur et al., 2017), as recommended for hypopneas by the American Academy of Sleep Medicine, did not meaningfully alter the findings (a similar threshold effect was observed for the highest quartile). Finally, more research is needed to determine whether the trait-like behavior of R/L ORP, evidenced by its association with MVC occurring 1 to 2 years later and by the observed stability of the measure over a 5-year period, reflects endogenous (possibly genetic) differences among individuals or rather the influence of a stable environmental exposures such as OSA.

Conclusions

High interhemispheric sleep depth coherence, quantified by R/L ORP, is associated with a markedly lower risk of motor vehicle crashes in individuals with OSA. This suggests that individuals with OSA who maintain high levels of interhemispheric sleep depth coherence may be less susceptible to OSA-related adverse neurocognitive outcomes. Future investigations are warranted to replicate these findings and to explore the underlying mechanisms, in order to understand the potential clinical utility of the R/L ORP measure.

Footnotes

Conflict of Interests

The Sleep Heart Health Study (SHHS) was supported by the National Heart, Lung, and Blood Institute through the following cooperative agreements: U01-HL53940 (University of Washington), U01-HL53941 (Boston University), U01-HL63463 (Case Western Reserve University), U01-HL53937 (Johns Hopkins University), U01-HL53938 (University of Arizona), U01-HL53916 (University of California, Davis), U01-HL53934 (University of Minnesota), U01-HL63429 (Missouri Breaks Research), and U01-HL53931 (New York University).

This work was also supported by the National Institutes of Health (R01HL102321, R01HL128658, P01HL095491, UL1RR025758). Dr. Azarbarzin was supported by the American Heart Association (19CDA34660137) and the American Academy of Sleep Medicine Foundation (188-SR-17). Drs. Redline, Sands, Wellman and Azarbarzin were partially supported by NHLBI R35HL135818.

AA serves as consultant for Somnifix and Apnimed Corp. MY reports personal fees and other from Cerebra Health, Winnipeg, Manitoba, Canada, outside the submitted work. MY has also a patent on the methodology of generating the odds ratio product (ORP) with royalties paid to him from Cerebra Health. SAS serves as consultant for Cambridge Sound Management.

AW works as a consultant for Somnifix, Cambridge Sound Management, Nox, Bayer, Philips, Apnimed, Inspire, and Galvani, and has received grants from Somnifix and Sanofi. AW also has a financial interest in Apnimed Corp., a company developing pharmacologic therapies for sleep apnea. His interests were reviewed and are managed by Brigham and Women’s Hospital and Partners HealthCare in accordance with their conflict of interest policies.

SR received grants from Jazz Pharma, grants from Beckman Coulter, outside the submitted work.

CAC has received consulting fees from or served as a paid member of scientific advisory boards for Bose, Boston Red Sox, Columbia River Bar Pilots, Institute of Digital Media and Child Development, Purdue Pharma, Samsung, and Vanda Pharmaceuticals. He owns an equity interest in Vanda Pharmaceuticals. He has served as an expert witness on various legal cases related to sleep and circadian rhythms and he has received research support from Optum, San Francisco Bar Pilots, Schneider, Sysco, Philips Respironics, Vanda Pharmaceuticals, and the State of Washington Board of Pilotage Commissioners. The Harvard Medical School Sleep and Health Education Program and Brigham Sleep Health (CAC) have received funding for educational activities from Cephalon, Jazz Pharma, ResMed, Takeda Pharmaceuticals, Sanofi-Aventis, Sepracor, Simmons, and Mary Ann & Stanley Snider via Combined Jewish Philanthropies. CAC is the incumbent of an endowed professorship provided to Harvard University by Cephalon, and holds several process patents in the specialty of sleep and circadian rhythms (e.g., photic resetting of the human circadian pacemaker). CAC has received royalties from Houghton Miflin Harcourt, and from Koninklijke Philips Electronics/Philips Respironics for the Actiwatch-2 and Actiwatch Spectrum devices. CAC’s interests were reviewed and are managed by Brigham Health and Partners HealthCare in accordance with their conflict of interest policies.

DJG has received research grants from ResMed Corporation outside the submitted work and has received consulting fees or served as a paid member of scientific advisory boards for ResMed Corporation, VIVUS, Inc., Advance Medical, Inc., and T. Leland Seeger & Associates, Inc.

REFERENCES

  1. Amatoury J, Azarbarzin A, Younes M, Jordan AS, Wellman A, & Eckert DJ (2016). Arousal Intensity is a Distinct Pathophysiological Trait in Obstructive Sleep Apnea. Sleep, 39(12), 2091–2100. doi: 10.5665/sleep.6304 [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Azarbarzin A, Ostrowski M, Hanly P, & Younes M (2014). Relationship between arousal intensity and heart rate response to arousal. Sleep, 37(4), 645–653. doi: 10.5665/sleep.3560 [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Azarbarzin A, Ostrowski M, Younes M, Keenan BT, Pack AI, Staley B, & Kuna ST (2015). Arousal Responses during Overnight Polysomnography and their Reproducibility in Healthy Young Adults. Sleep, 38(8), 1313–1321. doi: 10.5665/sleep.4916 [DOI] [PMC free article] [PubMed] [Google Scholar]
  4. Azarbarzin A, Sands SA, Stone KL, Taranto-Montemurro L, Messineo L, Terrill PI, … Wellman A (2019). The hypoxic burden of sleep apnoea predicts cardiovascular disease-related mortality: the Osteoporotic Fractures in Men Study and the Sleep Heart Health Study. Eur Heart J, 40(14), 1149–1157. doi: 10.1093/eurheartj/ehy624 [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bangdiwala SI, Haedo AS, Natal ML, & Villaveces A (2008). The agreement chart as an alternative to the receiver-operating characteristic curve for diagnostic tests. J Clin Epidemiol, 61(9), 866–874. doi: 10.1016/j.jclinepi.2008.04.002 [DOI] [PubMed] [Google Scholar]
  6. Bangdiwala SI, & Shankar V (2013). The agreement chart. BMC Med Res Methodol, 13, 97. doi: 10.1186/1471-2288-13-97 [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Boufous S, Ivers R, Senserrick T, Stevenson M, Norton R, & Williamson A (2010). Accuracy of self-report of on-road crashes and traffic offences in a cohort of young drivers: the DRIVE study. Inj Prev, 16(4), 275–277. doi: 10.1136/ip.2009.024877 [DOI] [PubMed] [Google Scholar]
  8. Canessa N, Castronovo V, Cappa SF, Marelli S, Iadanza A, Falini A, & Ferini-Strambi L (2018). Sleep apnea: Altered brain connectivity underlying a working-memory challenge. Neuroimage Clin, 19, 56–65. doi: S2213-1582(18)30106-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  9. Cohen J (1960). A Coefficient of Agreement for Nominal Scales. Educational and Psychological Measurement, 20(1), 37–46. doi: 10.1177/001316446002000104 [DOI] [Google Scholar]
  10. Colt HG, Hass H, & Rich GB (1991). Hypoxemia vs sleep fragmentation as cause of excessive daytime sleepiness in obstructive sleep apnea. CHEST Journal, 100(6), 1542–1548. doi: 10.1378/chest.100.6.1542 [DOI] [PubMed] [Google Scholar]
  11. Dres M, Younes M, Rittayamai N, Kendzerska T, Telias I, Grieco DL, … Brochard L (2019). Sleep and Pathological Wakefulness at the Time of Liberation from Mechanical Ventilation (SLEEWE). A Prospective Multicenter Physiological Study. Am J Respir Crit Care Med, 199(9), 1106–1115. doi: 10.1164/rccm.201811-2119OC [DOI] [PubMed] [Google Scholar]
  12. Engleman HM, Kingshott RN, Martin SE, & Douglas NJ (2000). Cognitive function in the sleep apnea/hypopnea syndrome (SAHS). Sleep, 23 Suppl 4(Abstract), S102–108. [PubMed] [Google Scholar]
  13. Franklin KA, & Lindberg E (2015). Obstructive sleep apnea is a common disorder in the population-a review on the epidemiology of sleep apnea. J Thorac Dis, 7(8), 1311–1322. doi: 10.3978/j.issn.2072-1439.2015.06.11 [DOI] [PMC free article] [PubMed] [Google Scholar]
  14. Gottlieb DJ, Ellenbogen JM, Bianchi MT, & Czeisler CA (2018). Sleep deficiency and motor vehicle crash risk in the general population: a prospective cohort study. BMC Med, 16(1), 44. doi: 10.1186/s12916-018-1025-7 [DOI] [PMC free article] [PubMed] [Google Scholar]
  15. Heinzer R, Vat S, Marques-Vidal P, Marti-Soler H, Andries D, Tobback N, … Haba-Rubio J (2015). Prevalence of sleep-disordered breathing in the general population: the HypnoLaus study. Lancet Respir Med, 3(4), 310–318. doi: 10.1016/S2213-2600(15)00043-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Jones SG, Riedner BA, Smith RF, Ferrarelli F, Tononi G, Davidson RJ, & Benca RM (2014). Regional reductions in sleep electroencephalography power in obstructive sleep apnea: a high-density EEG study. Sleep, 37(2), 399–407. doi: 10.5665/sleep.3424 [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Joo EY, Jeon S, Kim ST, Lee JM, & Hong SB (2013). Localized cortical thinning in patients with obstructive sleep apnea syndrome. Sleep, 36(8), 1153–1162. doi: 10.5665/sleep.2876 [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Kapur VK, Auckley DH, Chowdhuri S, Kuhlmann DC, Mehra R, Ramar K, & Harrod CG (2017). Clinical Practice Guideline for Diagnostic Testing for Adult Obstructive Sleep Apnea: An American Academy of Sleep Medicine Clinical Practice Guideline. J Clin Sleep Med, 13(3), 479–504. doi: 10.5664/jcsm.6506 [DOI] [PMC free article] [PubMed] [Google Scholar]
  19. Kielb SA, Ancoli-Israel S, Rebok GW, & Spira AP (2012). Cognition in obstructive sleep apnea-hypopnea syndrome (OSAS): current clinical knowledge and the impact of treatment. Neuromolecular Med, 14(3), 180–193. doi: 10.1007/s12017-012-8182-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  20. Li Y, & Veasey SC (2012). Neurobiology and neuropathophysiology of obstructive sleep apnea. Neuromolecular Med, 14(3), 168–179. doi: 10.1007/s12017-011-8165-7 [DOI] [PubMed] [Google Scholar]
  21. Nieto FJ, Young TB, Lind BK, Shahar E, Samet JM, Redline S, D’Agostino RB, Newman AB, Lebowitz MD, & Pickering TG (2000). Association of sleep-disordered breathing, sleep apnea, and hypertension in a large community-based study. Sleep Heart Health Study. JAMA, 283(14), 1829–1836. [DOI] [PubMed] [Google Scholar]
  22. Peppard PE, Young T, Barnet JH, Palta M, Hagen EW, & Hla KM (2013). Increased prevalence of sleep-disordered breathing in adults. Am J Epidemiol, 177(9), 1006–1014. doi: 10.1093/aje/kws342 [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Quan SF, Howard BV, Iber C, Kiley JP, Nieto FJ, O’Connor GT, … Wahl PW (1997). The Sleep Heart Health Study: design, rationale, and methods. Sleep, 20(12), 1077–1085. [PubMed] [Google Scholar]
  24. Rattenborg NC, Amlaner CJ, & Lima SL (2000). Behavioral, neurophysiological and evolutionary perspectives on unihemispheric sleep. Neurosci Biobehav Rev, 24(8), 817–842. doi: S0149763400000397 [pii] [DOI] [PubMed] [Google Scholar]
  25. Redline S, Sanders MH, Lind BK, Quan SF, Iber C, Gottlieb DJ, … Kiley JP (1998). Methods for obtaining and analyzing unattended polysomnography data for a multicenter study. Sleep Heart Health Research Group. Sleep, 21(7), 759–767. [PubMed] [Google Scholar]
  26. Redline S, Yenokyan G, Gottlieb DJ, Shahar E, O’Connor GT, Resnick HE, … Punjabi NM (2010). Obstructive sleep apnea-hypopnea and incident stroke: the sleep heart health study. Am J Respir Crit Care Med, 182(2), 269–277. doi: 10.1164/rccm.200911-1746OC [DOI] [PMC free article] [PubMed] [Google Scholar]
  27. Reynolds AC, & Banks S (2010). Total sleep deprivation, chronic sleep restriction and sleep disruption. Progress in brain research, 185(Journal Article), 91–103. [DOI] [PubMed] [Google Scholar]
  28. Rial R, Gonzalez J, Gene L, Akaarir M, Esteban S, Gamundi A, … Nicolau C (2013). Asymmetric sleep in apneic human patients. Am J Physiol Regul Integr Comp Physiol, 304(3), R232–237. doi: 10.1152/ajpregu.00302.2011 [DOI] [PubMed] [Google Scholar]
  29. Santarnecchi E, Sicilia I, Richiardi J, Vatti G, Polizzotto NR, Marino D, … Rossi A (2013). Altered cortical and subcortical local coherence in obstructive sleep apnea: a functional magnetic resonance imaging study. J Sleep Res, 22(3), 337–347. doi: 10.1111/jsr.12006 [DOI] [PubMed] [Google Scholar]
  30. Sassani A, Findley LJ, Kryger M, Goldlust E, George C, & Davidson TM (2004). Reducing motor-vehicle collisions, costs, and fatalities by treating obstructive sleep apnea syndrome. Sleep, 27(3), 453–458. doi: 10.1093/sleep/27.3.453 [DOI] [PubMed] [Google Scholar]
  31. Saunamaki T, Jehkonen M, Huupponen E, Polo O, & Himanen SL (2009). Visual dysfunction and computational sleep depth changes in obstructive sleep apnea syndrome. Clin EEG Neurosci, 40(3), 162–167. doi: 10.1177/155005940904000308 [DOI] [PubMed] [Google Scholar]
  32. Schonwald SV, Carvalho DZ, de Santa-Helena EL, Lemke N, & Gerhardt GJ (2012). Topography-specific spindle frequency changes in obstructive sleep apnea. BMC Neurosci, 13, 89. doi: 10.1186/1471-2202-13-89 [DOI] [PMC free article] [PubMed] [Google Scholar]
  33. Sforza E, Roche F, Chapelle C, & Pichot V (2019). Internight Variability of Apnea-Hypopnea Index in Obstructive Sleep Apnea Using Ambulatory Polysomnography. Front Physiol, 10, 849. doi: 10.3389/fphys.2019.00849 [DOI] [PMC free article] [PubMed] [Google Scholar]
  34. Singletary BA, Do AN, Donnelly JP, Huisingh C, Mefford MT, Modi R, … McGwin G (2017). Self-reported vs state-recorded motor vehicle collisions among older community dwelling individuals. Accid Anal Prev, 101, 22–27. doi: S0001-4575(17)30048-9 [pii] [DOI] [PMC free article] [PubMed] [Google Scholar]
  35. Takase B, Akima T, Satomura K, Ohsuzu F, Mastui T, Ishihara M, & Kurita A (2004). Effects of chronic sleep deprivation on autonomic activity by examining heart rate variability, plasma catecholamine, and intracellular magnesium levels. Biomed Pharmacother, 58 Suppl 1, S35–39. doi: S0753-3322(04)80007-6 [pii] [DOI] [PubMed] [Google Scholar]
  36. Tefft BC (2012). Motor vehicle crashes, injuries, and deaths in relation to driver age: United States, 1995–2010. . Washington, D.C.: AAA Foundation for Traffic Safety. [Google Scholar]
  37. Tregear S, Reston J, Schoelles K, & Phillips B (2009). Obstructive sleep apnea and risk of motor vehicle crash: systematic review and meta-analysis. J Clin Sleep Med, 5(6), 573–581. [PMC free article] [PubMed] [Google Scholar]
  38. Tsai JC (2010). Neurological and neurobehavioral sequelae of obstructive sleep apnea. NeuroRehabilitation, 26(1), 85–94. doi: 10.3233/NRE-2010-0538 [DOI] [PubMed] [Google Scholar]
  39. Younes M, & Hanly PJ (2016). Immediate postarousal sleep dynamics: an important determinant of sleep stability in obstructive sleep apnea. J Appl Physiol (1985), 120(7), 801–808. doi: 10.1152/japplphysiol.00880.2015 [DOI] [PubMed] [Google Scholar]
  40. Younes M, Ostrowski M, Soiferman M, Younes H, Raneri J, & Hanly P (2015). Odds ratio product of sleep EEG as a continuous measure of sleep state. Sleep, 38(4), 641–654. doi: 10.5665/sleep.4588 [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Young T, Palta M, Dempsey J, Skatrud J, Weber S, & Badr S (1993). The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med, 328(17), 1230–1235. doi: 10.1056/NEJM199304293281704 [DOI] [PubMed] [Google Scholar]

RESOURCES