OSA has the potential to contribute to or exacerbate a cerebrovascular accident

As cerebrovascular accidents (CVA) and sleep apnea are common, it is possible that the documented association between these two health problems represents only a chance occurrence. In fact, the only cause-and-effect relationship between these phenomena may be that seen when apnea occurs after there has been damage to the central mechanisms responsible for respiratory control. To date, there is no strong scientific evidence indicating that obstructive sleep apnea (OSA) can lead to cerebrovascular disease or, specifically, to CVA. Nevertheless, epidemiological and laboratory studies, anecdotal reports, polysomnographic analyses, and the circadian nature of a CVA suggest that, in some cases, OSA has the potential to contribute to or exacerbate a CVA.

In any given patient, CVA and sleep apnea undoubtedly can occur together by chance on occasion, as they share many risk factors. CVA is the third leading cause of death in the United States, whereas the prevalence of OSA ranges from 4% to 9% for women and 9% to 24% for men.1 A few of the risk factors that CVA and sleep apnea share are male gender, advanced age, diabetes, and tobacco use; both have also been associated with hypertension and heart disease.2

It has been proven through animal research and inferred by many human CVA studies that, in some instances, a CVA (especially when associated with a brain-stem lesion) can lead to apnea. It is frequently assumed that injury to the medulla oblongata of the brain stem predisposes a patient to only central sleep apnea or Cheyne-Stokes respiratory patterns, as it has been postulated that injury to the solitary nucleus can produce diaphragmatic dysfunction.3 Nevertheless, OSA can occur if the brain-stem injury involves the nucleus ambiguus or its associated upper motor neurons, which innervate the upper airways.4

Given the frequency of CVA and sleep apnea and the known relationship between central-nervous-system injury and apnea, it may seem a little presumptuous to hypothesize that sleep apnea can cause CVA. Nevertheless, epidemiological surveys, case reports, small retrospective and prospective (selected and nonselected) polysomnographic studies of patients with transient ischemic attacks (TIAs) and CVAs, laboratory studies, morbidity and mortality studies, and the circadian rhythmicity of CVA suggest that OSA can contribute to CVA, in some cases.

Significant snoring is one of the most frequently noted symptoms of classic OSA. Habitual snoring, in and of itself, has been associated with a greater relative risk of developing multi-infarct dementia,5 combined ischemic heart disease and CVA,6 and CVA.7 Nevertheless, many of the studies that have shown an association between snoring and a variety of health problems were done before the ready availability of polysomnography. The specific clinical context presented in many individual papers often implies an actual relationship between OSA (not snoring) and CVA.

For example, in 1989, Palomake et al8 studied 167 men with CVA, of whom 36% had CVAs in sleep. Snoring was the only potential CVA risk factor studied that appeared significantly related to experiencing a CVA in sleep. Further study by Palomake9 suggested that untreated OSA could cause CVA; it was determined that the odds ratio of snoring as a risk factor for CVA increased if the patient also had other classic signs and symptoms suggesting underlying OSA, such as obesity and excessive daytime sleepiness.

Polysomnographic Studies on TIAs
Polysomnographic studies performed on patients at risk for CVA have been very intriguing. Experiencing a TIA is considered a very strong risk factor for developing a CVA. In 1987, Rivest and Reiher10 were able to study a patient during a series of TIAs. They evaluated a 64-year-old man with a history of recurrent transient left hemiparesis, ophthalmoplegia, and Babinski signs precipitated by sleep. In the laboratory, they were able to document (polysomnographically) that all events were immediately preceded by significant obstructive apneic events. This temporal association suggested that the OSA was directly responsible for the TIAs and had the potential to be directly responsible for CVA.

In 1995, polysomnographic analysis of a patient with TIAs led to what might be considered a new CVA preventive.11 A 64-year-old woman reported waking with a transient motor aphasia. After an otherwise unremarkable, very thorough investigation for other CVA risk factors, overnight oximetry and polysomnography revealed loud snoring, with the lowest oxygen saturation level being below 50% and the apnea/hypopnea index showing 83.4 events per hour. After OSA was treated with continuous positive airway pressure (CPAP) therapy, there were dramatic improvements in the patient’s sleep quality and daytime alertness, and there was subsequently resolution of her TIAs. These findings suggested that, in some cases of nocturnal TIAs, CVA prevention could include instituting CPAP therapy if underlying OSA was present.

In 1996, a prospective polysomnographic study of unselected, consecutively encountered individuals with a history of TIAs showed a relatively high prevalence of OSA in the patient population, compared with a control group.12 OSA was diagnosed in 69% of 13 patients with previous TIAs and was seen in only 16% of the control population. The TIA patient group had a mean lowest oxygen saturation of 79% with a mean apnea/hypopnea index of 19 events per hour. The high prevalence of OSA in this population of TIA patients was used to hypothesize that apnea might have the potential to cause CVA.

The work of the few groups that have polysomnographically examined unselected, consecutively encountered CVA populations prospectively and compared them with controls has, in large part, been stimulated by anecdotal reports of polysomnographic studies performed in isolated CVA cases. At the Department of Neurology Sleep Disorders Center, University of Iowa College of Medicine, Iowa City, our group’s research interests were piqued by a 34-year-old morbidly obese man with severe OSA who awoke from sleep with clinical evidence of having suffered a first-time CVA.13 He had a history of snoring, excessive daytime sleepiness, and obesity prior to the development of his CVA. He had an otherwise unremarkable medical history, with the exception of failing a surgical procedure for weight loss. His sleep study showed an oxygen-saturation low of 60% with an apnea/hypopnea index of 36 events per hour. The patient’s history, in conjunction with his brain CT (which revealed a hemorrhage in the putamen and proximal posterior limb of the right internal capsule), suggested that he had experienced his CVA as the result of a nocturnal hypertensive event.

Autonomic Studies
We hypothesized, from these findings and in conjunction with previous studies that reported high blood pressures following nocturnal obstructive respiratory events,14 that hemorrhagic CVA could occur as a result of apnea-induced hypertension during sleep.13,15 This led to our collaborative laboratory experiments in which the autonomic nervous system was studied using microneurography. In essence, we found that simultaneously occurring excessive sympathetic and parasympathetic activation, with subsequent profound blood-pressure fluctuations and cardiac arrhythmias, could result from obstructive apneic events.16-19

During voluntary end-expiratory apneas in subjects studied in the laboratory setting and in patients with sleep-related OSA studied during nocturnal apneas, we demonstrated simultaneous parasympathetic activation (bradyarrhythmias with complete heart block and sinus pause) and sympathetic activation.17 Using fingertip blood-pressure–analysis techniques, periods of complete heart block were found to result in blood pressure drops from 180/100 mm Hg (recorded prior to the apneas) to less than 50 mm Hg (during the apneic events).

In healthy volunteers simulating OSA by performing the Mueller maneuver (inspiring against a closed glottis),18 we were able to document a mean arterial pressure that initially fell from 95 mm Hg to 81 mm Hg during transient reductions in sympathetic nerve activity. Immediately prior to release of the Mueller maneuver, sympathetic activity dramatically increased. Upon release of the Mueller maneuver, there was an upward surge in the mean arterial pressure to 104 mm Hg.

In a subsequent study,19 we then analyzed blood pressures using intra-arterial or fingertip blood-pressure–analysis methods while monitoring sympathetic nerve activity in patients with OSA. In the patients studied, the mean arterial pressure increased from 92 mm Hg in the waking state to 127 mm Hg in rapid–eye-movement (REM) sleep. Peak sympathetic activity increased 246% during REM sleep. Institution of CPAP treatment for OSA resulted in a decrease in blood pressure and sympathetic activity.

Research20 indicates that, during OSA events, elevated sympathetic activity is associated with a concomitant instability in blood pressure. This marked variability of blood pressure has been associated with significant reductions in intracerebral perfusion pressures. In addition, microneurographic studies show a sustained elevation of sympathetic tone in patients with OSA, even during the waking state.19 These findings suggest that OSA might be able to produce acute and chronic effects that could predispose a patient to CVA while asleep or awake.

Polysomnographic Studies on Patients’ CVAs
We performed a prospective study21 that compared the polysomnograms of 27 healthy age-matched and gender-matched controls without CVA to 24 consecutively studied nonselected inpatients with recent CVA. OSA was diagnosed in 19% of the controls and in 71% of the CVA patients, with 54% of the patient population having CVAs during sleep. The mean lowest oxygen saturation was 91% in the control subjects and 85% in the CVA group. Overall, the mean apnea/hypopnea index was four events per hour for controls and 26 events per hour for individuals with CVA.

These findings led to the speculation that hypoxemia and hypercarbia, along with the previously documented autonomic responses associated with OSA, might produce acute and chronic changes that could predispose patients to CVA.22 When cardiac disease and hypertension were selected out, OSA still had a relatively high prevalence in the CVA population. In this selected subgroup, there was no significant difference appreciated between the mean body mass indices of the non-CVA individuals and those of patients with CVA.

The literature suggests that untreated OSA can contribute to CVA. Partinen et al23,24 studied 198 patients with OSA. These individuals were either treated using tracheostomy or placed into a weight-loss group. A 7-year reassessment showed that, in the tracheostomy group, only 1% experienced CVA, while 3% died (1% from a vascular etiology). In the weight-loss group 9% experienced a CVA, while 17.3% died (11% from a vascular etiology). As these results were independent of age and obesity, it was suggested that OSA could lead to a higher risk of CVA and death if not adequately treated. In the 4-year reassessment of 24 subjects with CVA in our 1996 study,21 all of the patients who subsequently died had significant OSA, and 80% experienced their CVAs during sleep. Only one individual in the patient group, who died of urosepsis, used CPAP. The mean apnea/hypopnea index from the 1996 study for patients surviving the 4-year follow-up period was 22.1 events per hour. The mean apnea/hypopnea index for patients who died within the follow-up period was 41.3 events per hour. The findings suggested that the diagnosis and severity of OSA in CVA might be associated with greater mortality.

Circadian Rhythm
CVA (especially ischemic CVA) appears to have a characteristic circadian rhythmicity. Sleep and the early hours after awakening are associated with a relatively high frequency of CVA.21,25 If CVAs had an equal probability of occurring at any time, 33% should occur during the generally desirable 8-hour sleep period for an adult. In our study21 of acute CVAs, a higher than expected percentage (54%) of individuals suffered CVAs in sleep (P=.0304).
The circadian rhythmicity of CVA and the periodicity of the REM sleep stage permit a great deal of speculation regarding the relationship between CVA and OSA. In general, the longest REM period occurs during the early morning hours. During REM sleep, adults normally have a relative instability of the autonomic nervous system evidenced as an increase in sympathetic activation26 and relatively high blood pressures.27 Normal muscle twitches that occur in REM sleep result in additional blood-pressure surges. The oropharyngeal paresis normally associated with REM sleep may exacerbate obstructive apneic events, as well as the associated autonomic and cardiovascular instability previously documented through microneurographic studies.

In addition, during REM sleep, it has been shown that cerebral blood flow increases throughout the brain.28 Obstructive apneic events have been documented as increasing intracranial pressure while reducing cerebral perfusion pressure.20 This combination of effects may add to other factors that could predispose people with OSA to ischemic CVA at a time when there normally is an increased demand by the brain for oxygen.

A study29 has shown that people generally experience their lowest fibrinolytic activity and their highest catecholamine, blood-viscosity, and platelet-activity levels during the early hours of the morning. It has been hypothesized that the elevated catecholamine levels that have been associated with OSA30 may increase the risk of developing thrombosis and subsequent CVA in an otherwise relatively hostile (but normal) morning hematological environment. The combination of OSA and circadian, autonomic, metabolic, and hematological concomitants may increase the risk of developing CVA during sleep.

Although it is a fact that apnea can result from direct damage to respiratory centers in the central nervous system, whether OSA can lead to CVA is not clear. Ethical and logistic constraints may not allow for large, prospective, double-blind, controlled experiments in which polysomnography could be performed prior to and after CVA, comparing treated and untreated patients with OSA. Nevertheless, the literature has documented a high prevalence of OSA in the CVA population and suggests that, in some cases, it can contribute to CVA. The potential negative effects of OSA’s concomitant metabolic, autonomic, and hematologic irregularities on the ischemic penumbra of the central nervous system should, in my opinion, encourage all health care professionals to consider the possibility of underlying OSA when addressing the patient population with CVA.

Mark Eric Dyken, MD, is director and associate professor of neurology, Department of Neurology Sleep Disorders Center, University of Iowa College of Medicine, Iowa City.

1. Young T, Palta M, Dempsey J, Skatrud J, Weber S, Safwan B. The occurrence of sleep-disordered breathing among middle-aged adults. N Engl J Med. 1993;328:1230-1235.
2. Dyken ME, Somers V, Yamada T, Yeh M, Ren Z, Zimmerman MB. The effect hypertension, cardiac disease and stroke has on clinically suspected obstructive sleep apnea [abstract]. Sleep Research. 1995;24:388.
3. Guilleminault C, Dement W, eds. Sleep Apnea Syndromes. New York: Liss; 1978.
4. Askenasy JJM, Goldhammer I. Sleep apnea as a feature of bulbar stroke. Stroke. 1988;19:637-639.
5. Erkinjuntti T, Partinen M, Sulkava R, Palomaki H, Tilvis RS. Snoring and dementia. Age Ageing. 1987;16:305-310.
6. Koskenvuo M, Kaprio J, Telakivi T, Partinen M, Heikkila K, Sarna S. Snoring as a risk factor for ischaemic heart disease and stroke in men. BMJ. 1987;294:16-19.
7. Palomaki H, Partinen M, Erkinjuntti T, Kaste M. Snoring, sleep apnea syndrome, and stroke. Neurology. 1992;42:S75-S81.
8. Palomake H, Partinen M, Juvela S, Kaste M. Snoring as a risk factor for sleep-related brain infarction. Stroke. 1989;10:1311-1315.
9. Palomake H. Snoring and the risk of ischemic brain infarction. Stroke. 1991;22:1021-1025.
10. Rivest J, Reiher J. Transient ischemic attacks triggered by symptomatic sleep apneas [abstract]. Stroke. 1987;18:293.
11. Pressman MR, Schetman WR, Figueroa WG, Van Uitert B, Caplan HJ, Peterson DD. Transient ischemic attacks and minor stroke during sleep. Stroke. 1995;26:2361-2365.
12. Bassetti C, Aldrich MS, Chervin RD, Quint D. Sleep apnea in patients with transient ischemic attack and stroke: a prospective study of 59 patients. Neurology. 1996;47:1167-1173.
13. Dyken ME, Somers VK, Yamada T. Hemorrhagic stroke; part of the natural history of severe obstructive sleep apnea? [abstract]. Sleep Research. 1991;20:371.
14. Tilkian AG, Guilleminault C, Schroeder JS, Lehrman KL, Simmons FB, Dement WC. Hemodynamics in sleep-induced apnea: studies during wakefulness and sleep. Ann Intern Med. 1976;85:714-719.
15. Dyken ME, Somers VK, Yamada T. Stroke, sleep apnea and autonomic instability. In: Togawa K, Katayama S, Hishikawa Y, Ohta Y, Horie T, eds. Sleep Apnea and Rhonchopathy. Basel, Switzerland: Karger; 1993:166-168.
16. Somers VK, Dyken ME, Mark AL, Abboud FM. Autonomic and hemodynamic responses during sleep in normal and sleep apneic humans [abstract]. Clin Res. 1991;39:735.
17. Somers VK, Dyken ME, Mark AL, Abboud FM. Parasympathetic hyperresponsiveness and bradyarrhythmias during apnea in hypertension. Clin Auton Res. 1992;2:171-176.
18. Somers VK, Dyken ME, Skinner JL. Autonomic and hemodynamic responses and interactions during the Mueller maneuver in humans. J Auton Nerv Syst. 1993;44:253-259.
19. Somers VK, Dyken ME, Clary MP, Abboud FM. Sympathetic neural mechanisms in obstructive sleep apnea. J Clin Invest. 1995;96:1897-1904.
20. Jennum P, Borgesen SE. Intracranial pressure and obstructive sleep apnea. Chest. 1989;95:279-283.
21. Dyken ME, Somers VK, Yamada T, Ren Z-Y, Zimmerman MB. Investigating the relationship between stroke and obstructive sleep apnea. Stroke. 1996;27:401-407.
22. Dyken ME, Somers VK, Yamada T, Adams HP, Zimmerman MB. Investigating the relationship between sleep apnea and stroke [abstract]. Sleep Research. 1992;21:30.
23. Partinen M, Jamieson A, Guilleminault C. Long-term outcome for obstructive sleep apnea syndrome patients: mortality. Chest. 1988;94:1200-1204.
24. Partinen M, Guilleminault C. Daytime sleepiness and vascular morbidity at seven-year follow-up in obstructive sleep apnea patients. Chest. 1990;97:27-32.
25. Marsh E, Biller J, Adams H, et al. Circadian variation in onset of acute ischemic stroke. Arch Neurol. 1990;47:1178-1180.
26. Hornyak M, Cejnar M, Elam M, Matousek M, Wallin G. Sympathetic muscle nerve activity during sleep in man. Brain. 1991;114:1281-1295.
27. Somers VK, Dyken ME, Mark AL, Abboud FM. Sympathetic nerve activity during sleep in normal humans. N Engl J Med. 1993;328:303-307.
28. Klingelhofer J, Hajak G, Sander D, Schulz-Varszegi M, Ruther E, Conrad B. Assessment of intracranial hemodynamics in sleep apnea syndrome. Stroke. 1992;23:1427-1433.
29. Andreotti F, Davies GJ, Hackett DR, et al. Major circadian fluctuations in fibrinolytic factors and possible relevance to time of onset of myocardial infarction, sudden cardiac death and stroke. Am J Cardiol. 1988;62:635-637.
30. Fletcher EC, Miller J, Schaaf JW, Fletcher JG. Urinary catecholamines before and after tracheostomy in patients with obstructive sleep apnea and hypertension. Sleep. 1987;10:35-44.