The enigma of severe obstructive sleep apnea without sleepiness

It is common for a physician to feel astonished in front of a severe obstructive sleep apnea (OSA) patient who does not complain about sleepiness despite his polysomnography showing hundreds of apneas and hypopneas, oxygen desaturations, and arousals fragmenting sleep all night long. How can such a patient not be sleepy during the day?

Part of the problem may be related to the methods used to measure sleepiness. The Epworth Sleepiness Scale score1 is a worldwide used questionnaire that asks to subjectively rate the propensity to fall asleep in specific daytime situations such as while reading, as a passenger in a car, or watching TV. The ESS correlates moderately well with objective measures of sleepiness such as the multiple sleep latency test (MSLT) and maintenance of wakefulness test (MWT) and correlates weakly with measures of OSA severity such as the apnea‐hypopnea index (AHI) or oxygen saturation.2 Notably, objective measurements of sleepiness do not correlate much better with the AHI or other polysomnographic indices than does the ESS.3, 4

According to a meta‐analysis that examined the psychometric properties of the ESS,2 the questionnaire is considered reliable for assessing subjective sleepiness and has a relatively high internal consistency, suggesting that it is appropriate to use for comparisons between groups. Moreover, the ESS is widely available and can be performed more easily and quickly compared to the time‐consuming and expensive objective measures of sleepiness. Although the ESS score tends to increase with the severity of OSA, according to the Sleep Health Heart Study, only 46% percent of subjects with moderate‐to‐severe sleep‐disordered breathing reported an abnormal daytime sleepiness (ESS score > 10).5 It is clear from these data that the ESS should not be used during patients screening for OSA as it lacks sensitivity or specificity for this; nonetheless, it is important to assess sleepiness to better understand the patients, in order to better characterize them and to make more informed therapeutic choices, as is has been shown that the presence of sleepiness can affect treatment adherence and outcomes.6

A second problem may be the definition of sleep apnea severity. In patients with OSA, the main factors that have been considered responsible for daytime sleepiness are the frequency of apneas and hypopneas (AHI), the degree of overnight oxygen desaturation, and the degree of sleep fragmentation as assessed by frequency of arousals from sleep and sleep architecture.4 In the clinical setting, only the AHI is used as the single parameter characterizing OSA severity, although it provides a simple count of obstructive episodes per hour of sleep, regardless of the duration and the depth of the ventilatory disturbance or blood gas changes. Thus, the reason for the mild or absent association between sleepiness symptoms and OSA severity may be related to the poor stratification of patients that relies only on the AHI. Indices like the newly introduced hypoxic burden that quantifies the ventilatory disturbance related to OSA may be more helpful in this regard.7

The dissociation between sleepiness and OSA severity has been reported in several observational studies, with a particular emphasis on older people8 and on people with cardiovascular diseases.9 Regarding the latter group, it has been shown that only a minority of individuals with cardiovascular diseases and OSA complains of excessive daytime sleepiness. Arzt et al10 compared subjective daytime sleepiness assessed by Epworth Sleepiness Scale (ESS) scores among 155 patients with HF and a community sample of 1,139 subjects without HF in the Wisconsin Sleep Cohort Study. The HF population had lower ESS scores for any degree of OSA severity, despite sleeping approximately 80 minutes less. Similar findings have been shown in patients with atrial fibrillation 11 and stroke.9

In the paper from Tam et al12 published in this issue of The Journal of Clinical Hypertension, the authors found that in an Asian cohort of 280 patients, after adjusting for anthropometric variables, age, and apnea‐hypopnea index (AHI), the patients with concomitant hypertension and moderate‐to‐severe OSA had an ESS score 3.7 points lower compared to normotensive subjects with OSA alone. The study suffers from some limitations, for example the relatively small sample size and the fact that the diagnosis of OSA was obtained with 2 different methods (either a complete in‐lab polysomnography or a home cardio‐respiratory monitoring) using different scoring criteria for the two systems. The diagnosis of hypertension and other diseases were extracted from the hospital database, and there is a lack of details regarding the hypertensive condition, for example how many of these patients had well‐controlled vs drug‐resistant hypertension. Nevertheless, the analysis is comprehensive of several confounding factors and the results of this retrospective study are consistent with a previous report 13 suggesting that patients with hypertension and moderate‐to‐severe OSA on average are less sleepy compared to OSA patients without hypertension.

Cross‐sectional studies consistently found an increased prevalence of hypertension in patients with OSA and, on the other hand, moderate or severe OSA can be detected in a third of patients with primary hypertension and in up to 80% of those with drug‐resistant hypertension.14 OSA patients are frequently “non‐dippers,” meaning that they generally do not experience the physiologic reduction in blood pressure that happens overnight during sleep. Although the suppression of OSA by continuous positive airway pressure (CPAP) immediately reduces nocturnal blood pressure, it is still debated if treating OSA can consistently improve blood pressure in a clinically significant way, as the CPAP showed only modest reduction in recent meta‐analysis.14

As suggested by Tam and coworkers, one possible explanation for the lack of sleepiness in patients with OSA and cardiovascular conditions, including hypertension, is that both disorders may contribute to the increase in central adrenergic alerting mechanisms: the sympathetic autonomic over‐activation may be responsible for a hyperaroused state15, 16 that may mask the sleep‐inducing effects of OSA.

Daytime sleepiness commonly brings patients to medical attention and it is generally the principal clinical indication to treat OSA; its absence has important implications. Most patients with OSA and cardiovascular diseases do not report excessive daytime sleepiness, and thus, the majority of the patients lack this usual indication for treatment, and it remains uncertain whether OSA patients who do not report excessive daytime sleepiness require therapy at all.9, 14

In summary, studies like this by Tam and coworkers highlight some important gaps in the knowledge of OSA and its symptoms: we possibly need a better way to measure sleepiness in the clinical context, surely a better definition of OSA severity and a better understanding of the mechanisms that link sleepiness (or lack of thereof) and cardiovascular diseases in OSA patients.

CONFLICT OF INTEREST

LTM has a financial interest in Apnimed Corp., a company developing pharmacologic therapies for sleep apnea. His interests are reviewed and managed by Brigham and Women's Hospital and Partners HealthCare in accordance with their conflict of interest policies. LTM also works as a consultant for Apnimed and received personal fees as a consultant for Novion Pharmaceuticals and Cambridge Sound Management outside the submitted work.