PHILADELPHIA—Some experts have suggested that polysomnography performed in a sleep laboratory is no longer necessary to detect sleep-disordered breathing because simpler home sleep studies are sufficient. However, Nancy A. Collop, MD, believes that polysomnography is still superior, a position she defended in a pro–con debate at the 67th Annual International Scientific Assembly of the American College of Chest Physicians in Philadelphia.[1]

“Polysomnography is considered the gold standard,” stressed Dr. Collop, who is a Professor of Medicine at the University of Mississippi Medical Center as well as Director of the Sleep Disorders Center at the Jackson Veterans Administration Medical Center. The role of home monitoring in the diagnosis of sleep-disordered breathing is still undefined, Dr. Collop posited.

“Unattended home monitoring is acceptable as a screening tool for sleep-disordered breathing in an appropriate patient population with appropriate safeguards,” asserted her opponent, David W. Hudgel, MD.[2] Dr. Hudgel is Director of the Sleep Disorders Medicine Program at Henry Ford Hospital in Detroit.

WHY POLYSOMNOGRAPHY IS BEST

Polysomnography can accurately detect a variety of sleep disturbances—not only sleep-disordered breathing but also seizures, bruxism, and parasomnias, noted Dr. Collop. It can also provide data on sleep architecture in the presence of abnormal brain waves.

Because polysomnography is closely monitored, technical problems can be corrected while the study is performed, thereby preventing the need to repeat it. Furthermore, it can be stopped early to permit immediate treatment if sleep-disordered breathing becomes quickly apparent. In addition, some research suggests that patients prefer polysomnography to home monitoring, presumably because they have no responsibility for equipment.

Home sleep studies appear to perform comparably to polysomnography when the two are tested in a controlled laboratory setting. But when the home sleep studies are performed in patients’ homes, technical failure rates range as high as 20%, said Dr. Collop.

Accuracy is also a concern with these studies, she noted, because they rely heavily on the apnea-hypopnea index (AHI), which is a highly variable measurement given that no standard definition for hypopnea exists. Dr. Collop added that sleep studies can underestimate the AHI if they do not include the equipment necessary for an optimal measurement, such as monitors to determine body position and total sleep time.

Also, the AHI may be insensitive to subtler but equally serious forms of sleep–disordered breathing, such as upper airway resistance syndrome (UARS). Patients with UARS display just as much flow limitation, nighttime arousal, and daytime sleepiness as do those with sleep apnea but do not have an elevated AHI, Dr. Collop explained. Yet, some home sleep studies may not be able to differentiate patients with UARS from healthy controls.

Diagnosing an elusive sleep disorder like UARS requires the use of more complex technology, such as electroencephalography, respiratory inductance plethysmography, nasocannula pressure transduction, or an esophageal pressure transduction. “That would be difficult to do in the home,” commented Dr. Collop.

Other disadvantages of home sleep studies are the inability to immediately resolve technical problems; an increased role for patients that limits the complexity of the study; and a lack of data on their cost, safety, and compliance. It is also unclear whether the results of home monitoring must be interpreted by a board-certified sleep specialist or whether other physicians with less sleep medicine expertise can accurately interpret the results.

Home sleep studies can also present logistical problems: They may require substantial travel if technicians have to go to patients’ homes to connect the leads. If the leads are applied while patients are at the sleep center, there is a risk that the leads will disconnect before the study is conducted in the home, Dr. Collop said.

HOME MONITORING IS FEASIBLE

Although Dr. Hudgel shares many of Dr. Collop’s concerns about home monitoring, he believes it can play an important role in screening patients for sleep-disordered breathing. He noted that many studies have simultaneously compared home monitoring to polysomnography for the diagnosis of this disorder. Home monitoring provides the correct diagnosis in about 70% to 80% of cases, he said. It also has only a 10% to 12% technical failure rate under optimal conditions. Dr. Hudgel acknowledged, however, that the accuracy of home monitoring can vary because a number of different devices can be used for home sleep studies.

One device detects obstructive sleep apnea (OSA) by measuring nasal pressure through nasal prongs. In seven studies, the sensitivity of this device was 66% to 100% at an AHI of 15 to 20; its specificity was 77% to 93%. The device did tend to overestimate the occurrence of apneic episodes, though, particularly at higher AHIs, Dr. Hudgel admitted.

Another home monitoring device assesses oxygen saturation, heart rate, snoring, and body position. In four studies, its sensitivity for OSA was 97% to 100%. Its specificity was relatively poor, however. These studies also suggested that scoring home sleep tests by hand is more reliable than automated scoring, Dr. Hudgel noted.

One unique device designed for home use indirectly monitors sleep, rapid-eye-movement sleep, and wakefulness by evaluating eye movement and body motion. Both its sensitivity and specificity for OSA are good—91% and 70%, respectively. Furthermore, data have suggested that a large proportion of patients prefer this device to polysomnography performed in a sleep laboratory.

Home sleep monitors that rely on pulse oximetry are not very useful for diagnosing OSA, some studies have indicated. However, their accuracy markedly increases when results are combined with clinical findings, Dr. Hudgel stated. It is not surprising that arterial oxygen saturation alone often has low specificity for OSA, he remarked, because younger patients with sleep-disordered breathing often have good lung function and thus may not desaturate during apneic episodes.

COMBINED APPROACH CAN PRODUCE GOOD RESULTS

Perhaps the best use of home sleep monitoring is in conjunction with clinical prediction rules that triage patients based on home monitoring results to determine which ones require polysomnography.[3] For example, said Dr. Hudgel, polysomnography would be unnecessary in patients with a low clinical probability of OSA and a low respiratory disturbance index during a home sleep study.

This type of triaging could reduce the use of polysomnography by about 40%. The end result would be better access to care for sleep-disordered breathing and, presumably, some cost savings, Dr. Hudgel said. “[Home sleep monitoring] is a tool we surely can use with care,” he concluded.

Dr. Collop concurred that such a triaging system could help establish a role for home sleep studies. However, she emphasized that the population of patients who might be candidates for home monitoring has not yet been clearly defined; even more important is that there is still no consensus on exactly what constitutes sleep-disordered breathing or when treatment is appropriate. Until such a consensus is reached, the role of home monitoring will remain undefined, she concluded.

—Timothy Begany