Obstructive Sleep Apnea (OSA) in Primary Care: Evidence-based Practice

OSA—The Clinical Spectrum

Each of us spend a third of our lives asleep. Dysfunctions in this basic state lead to declines in quality of life, diminished waking performance, more frequent illness, and increases in both morbidity and mortality. In the medical care setting, sleep disorders are common. Although 30% of the general population report symptoms of sleep disruption, >50% of primary care patients have sleep complaints.³ Despite the high prevalence of sleep disorders in the population and primary care setting, sleep complains are often underaddressed by physicians.

OSA occurs secondary to the obstruction of the airway during sleep, resulting in continued breathing effort with diminished airflow. OSA occurs at a high frequency in the primary care clinic population. It is one of the most physiologic disruptive and dangerous of sleep-related diagnoses, affecting 1 of every 5 adults in some populations.⁴ As many as 18 million Americans suffer from sleep apnea. It is more common among men and those who snore, are overweight, have high blood pressure, or have physical abnormalities in their upper airways.^4,5 General population studies suggest that snoring and daytime sleepiness are present in at least 4.1% of people over 40 years of age, with associated apnea diagnosed and treated in only 0.6% of this population.⁶

A large multicenter trial called the Sleep Heart Health Study has been designed to study the prospective effects of OSA on a large population included in previously developed cardiovascular studies (6440 men and women over 40 years of age).⁵ Cross-section analysis data from this study has emphasized the significant contribution of OSA to pulmonary, cardiac, endocrine, and cognitive disease.^7–10 Prospective data from this study will address these factors as well as the morbidity, comorbidity, and mortality associated with OSA that have been demonstrated in multiple retrospective studies.

Of the sleep disorders, OSA is the best studied from a cost-effect, epidemiologic, and evidence-based perspective. The associated morbidity, mortality, comorbidities, and quality of life effects are well researched and described. Adult OSA has a long-term and clear association with obesity and daytime cognitive impairment that includes daytime sleepiness.^5,7,11 Daytime sleepiness has been shown to lead to an increase in motor vehicular accidents in untreated OSA patients. OSA severity correlates in part with the number of respiratory events per hour. This is generally reported as the Apnea-Hypopnea Index (AHI), which includes the total of the number of apneas (events of breathing cessation >10 seconds) and hypopneas (>30% decline in breathing effort coupled with >4% SaO₂ desaturation and/or arousal) per hour. Subjects with an AHI >10 have a 6.3-fold increased risk of having a traffic accident compared with 152 case-matched control with AHI <10.¹² Patients with moderate to severe OSA (AHI >34) have a 15-fold increased risk of having a motor vehicle accident.¹³ Recent epidemiologic studies that have cross-matched sleep apnea evaluation with long-term prospective cardiovascular risk studies have served to point out the consistent and strong association between OSA and essential hypertension. Odds of hypertension increases with increasing severity of apnea in a graded dose–response fashion with an odds ratio of 1.27 for hypertension in the group with AHI >30 versus the nonapnic group with an AHI of <5.^10,14,15 Research supports the association between OSA and increased mortality, congestive heart failure (both right and left sided), myocardial infarction, and cerebral vascular accidents.^16,17 Cross-sectional analysis data from the Sleep Heart Health Study demonstrated the following relative odds ratios for OSA patients: congestive heart failure (2.38), stroke (1.58), and coronary heart disease (1.27).⁸ Evidence is increasingly clear for the association of adult OSA with diabetes and metabolic syndrome.^9,18 Cardiac arrhythmias (bradycardia, atrial fibrillation, and ventricular tachycardia) are often seen in polysomnography (PSG) studies of OSA patients; however, the clinical significance and OSA association of these arrhythmias has yet to be fully studied¹⁹ (Table 3).

The pathophysiology and clinical presentation of pediatric OSA differs from that of adult OSA. In pediatric patients, OSA is most clearly associated with poor school performance. In first-graders performing at the bottom 10% percent of grade level, >20% have OSA. Furthermore, the grades of all children who had surgery (tonsillectomy and adenoidectomy) improved, whereas the grades of children who did not have surgery showed no change.²⁰ Studies also support the association of pediatric OSA with failure to thrive, enuresis, and learning disability.²² Studies have been contradictory in addressing the association of pediatric OSA with obesity and attention deficit/hyperactivity disorder, with strong associations occurring in specific patient populations and not in other clinically defined settings²¹ (Table 3).

OSA Cost and Health Care Use

The costs of untreated sleep apnea have been addressed in several studies. In 238 consecutive OSA patients studied in 1999, the mean annual medical cost was $2720 greater per patient before diagnosis compared with age-, body mass index-, and gender-matched controls.²³ Patients with OSA use health care resources at higher rates than control subjects for years before diagnosis.²⁴ For the 10 years before diagnosis, physician claims for patients eventually diagnosed with OSA are twice ($3972 per patient) that of age-matched controls ($1969 per patient).²⁵ Of all comorbid diagnoses, significantly increased health care use is found for cardiovascular disease and hypertension in OSA patients.¹⁹ There was a rise in health care costs each year before diagnosis, with initial data suggesting that after diagnosis, yearly claims were halved. By the time patients were finally diagnosed with sleep apnea, they had already been heavy users of health services for several years.²⁶ In Canada, hospital stays are 1.27 days per patient per year 1 year before OSA diagnosis and 0.53 days per patient per year 1 year after diagnosis. These differences were only seen in those patients adhering to treatment with no difference between patients and controls for nonadherers.²⁷ In pediatric OSA, there are also suggestions for increased health care use with a 226% increase in health care use 1 year before evaluation, more hospital days, more drug use, and more visits to ER, with the severity of OSA correlating directly to total annual cost independent of age.²⁸

Effectiveness of CPAP Therapy for OSA

A meta-analysis of applicable studies demonstrated that consistent evidence exists showing that treatment of OSA with continuous positive airway pressure (CPAP) therapy leads to significant improvement in daytime sleepiness and quality of life measures as well as reduced diastolic and systolic blood pressure.²⁹ In OSA patients, there is reduced hospitalization with cardiovascular and pulmonary disease in OSA patients on nasal CPAP treatment.^27,29 CPAP treatment reduces the need for acute hospital admission due to cardiovascular and pulmonary disease in patients with OSA. For the 2 years before and 2 years after CPAP use in CPAP users, 413 hospital days were used before treatment and 54 hospital days were used after treatment. In OSA CPAP nonusers, 137 hospital days were used before treatment and 188 hospital days were used after treatment. This reduction of concomitant health care consumption should be taken into consideration when assessing the cost-benefit evaluation of CPAP therapy.

Hypersomnia

In our modern, fast-paced world, an adequate level of alertness is required for well-being and performance. In a year 2000 United States national survey, 32% of respondents reported daytime sleepiness despite adequate sleep the preceding night.³⁰ This diagnostic category includes a group of diagnoses sharing the primary characteristic of inducing significant daytime sleepiness. These diagnoses have significant affects on waking performance and therefore morbidity and mortality. The National Health and Safety Administration in 1999 estimated that 1.5% of 100,000 police-reported crashes and 4% of all traffic crash fatalities involved drowsiness and fatigue as principal causes. Beyond the personal and social loss associated with these accidents, the National Health and Safety Administration in 1994 estimated cost at $83,000 per fatality, resulting in a total cost of $12.5 billion, with 85% of this cost being from workplace loss and loss of productivity.³¹

Diagnostic Evaluation of OSA

To diagnosis and manage OSA, sleep physicians routinely use diagnostic tests that require the sleep laboratory for evaluation of the patient. The sleep-related breathing disorders generally require PSG for evaluation. PSG is the recording of multiple physiologic signals during sleep. The standard PSG recording montage includes channels of electroencephalography, electrooculography, and chin electromyography that are required for sleep staging as well as recordings of respiratory effort, airflow, pulse oximetry, snoring, sleep position, electrocardiogram, leg electromyography, and video monitoring. Additional channels are sometimes used including end-tidal or transcutaneous CO₂ and additional electroencephalography channels if potential nocturnal seizure disorders are being evaluated. In evaluating the sleep-related breathing disorders, a split-night protocol is often used in which a therapeutic treatment or “titration” portion of the PSG is added after at least 120 minutes of diagnostic sleep time. During the titration, CPAP, Bi-PAP, and oxygen are used in an attempt to eliminate or reduce respiratory events and restore normal sleep. The PSG report is scored by a sleep technologist and interpreted by a sleep medicine physician. The PSG interpretation that the patient receives should include data regarding sleep architecture, respiratory parameters, periodic limb movements, a description of any parasomnia or seizure activity, electrocardiogram abnormalities, and the results of and appropriate setting of any titration attempted during the night of study.

Daytime sleepiness is generally evaluated via multiple sleep latency testing (MSLT) that includes 4 to 5 opportunities to nap in the sleep laboratory after a full night PSG under standard conditions with electroencephalography, electrooculography, and electromyography monitored so that sleep and rapid eye movement sleep onset can be determined. MSLT reports should include average or mean latency to sleep and the number of sleep onset rapid-eye-movement sleep periods recorded (a diagnostic criterion for narcolepsy). The maintenance of wakefulness test (MWT) is similar to the MSLT; however, for this procedure, the patient attempts to maintain wakefulness while monitored for appropriate testing periods to assess the patients ability to maintain wakefulness during the day.

Sleep laboratory testing can be expensive, and alternative approaches have been attempted. At this point, however, in-laboratory, full PSG with respiratory titration as required is the most cost-effective approach to evaluation of sleep disorders when required (Table 4). Limited PSGs include fewer recording channels and cannot determine whether the patient is actually asleep during the recording. Full home PSGs are a potential alternative, however, incomplete recordings are obtained in 20% of studies and titration cannot be attempted in the night of study.^35,36 Auto-titrating PAP systems have minimal diagnostic capacity and can report inappropriate settings for misdiagnosed patients, for patients with central apneas, and for those with nasal congestion or mouth leaks on CPAP therapy.³⁷ Polysomnographic testing provides a wealth of useful information for the primary care physician involved in the treatment of the patient's sleep disorder. The primary care physician able to understand the data and interpretation from a high-quality PSG will find much information useful in patient care.

SUMMARY

The evidence basis is excellent in supporting the importance the diagnosis and treatment of OSA in adults. Although data exists documenting appropriate approaches to pediatric apnea and hypersomnolence, a greater number of larger and better-designed studies are needed to support both diagnostic and therapeutic approaches (Table 5). The diagnosis and treatment of OSA should be considered as part of the management of diabetes, hypertension, and congestive heart failure—core aspects of primary care medicine. Sleep laboratory testing can be used as an objective insight into the patient's pulmonary, cardiac, neurological, endocrine, cognitive, and psychiatric status. The family physician with training in sleep and an understanding of appropriately used testing procedures can use current evidence based knowledge in the field to provide high-quality sleep medicine in a primary care setting.