The (Still) Elusive Definition of Hypopnea

摘要

Once upon a time there was a 30-apnea rule that was used by Medicare to determine candidacy for positive airway pressure (PAP) treatment. The origin of this rule is somewhat obscure but was probably based on the initial studies of obstructive sleep apnea (OSA) in the early 1970s.1 To characterize manifestations of OSA in comparison to a control group, the inclusion criterion was established to be at least 5 apneas/h during a 6-hour polysomnographythus, 30 apneas. This rule did not recognize the existence of hypopneas, and it complicated the practical definition of sleep-disordered breathing as it rapidly became clear that many patients expressed polysomnographic patterns that were characterized by hypopneaswhatever they wererather than apneas.2 While performing split-night studies, sleep technicians would literally count apneas until the threshold was reached at which point titration of PAP could commence. Often there was insufficient time remaining to successfully identify effective PAP treatment. There were also occasions when hypopneas were scored as apneas to facilitate needed treatment with the hope that records would not be audited. Most importantly, using the 30-apnea rule, many patients did not qualify for the payment of needed treatment through their insurance provider.The first comprehensive effort to define hypopneas was published in 1999.3 This report concluded that it was not necessary to distinguish obstructive hypopneas from apneas because both types of events have similar pathophysiology. This rather vague and perhaps overly flexible definition did not resolve the controversy related to the 30-apnea rule. In 2001, the Clinical Practice Review Committee of the American Academy of Sleep Medicine (AASM) published a position paper that recommended specific criteria for the definition of hypopnea.4 The definition advanced was that used in the Sleep Heart Health Study: 30% or greater reduction in airflow or chest wall movement accompanied by a decrease in oxyhemoglobin desaturation of 4%. While this definition was not rigorously evidence-based by current standards, it was felt that using these criteria would result in high interobserver agreement, and it seemed reasonable based on early findings from the Sleep Heart Health Study. This definition was ultimately recognized by the Centers for Medicare and Medicaid Services (CMS) for reimbursement resulting in the welcome passing of the 30-apnea rule.In 2007, the AASM published its comprehensive manual for the scoring of sleep and associated events.5 The manual included a recommended hypopnea scoring rule matching the defintion of the Clinical Practice Review Committee, and proposed a new alternative hypopnea scoring rule based on 50% drop in flow associated with a 3% decrease in saturation or an arousal. In 2012, an update to the scoring rules was published.6 The alternative rule was elevated to a recommended rule, and the previously recommended rule was retained for those patients that require CMS reimbursment. There have been several subsequent studies that demonstrated significant differences in the resulting apnea-hypopnea index (AHI) based on which scoring criteria are used.79 This confusing vacillating definition has created a rather bizarre, and perhaps unsettling, situation wherein the severity of the diagnosis of sleep-disordered breathing, and perhaps its presence or absence, is determined by the patient's insurance coverage.The paper by Won et al.10 in this issue of the Journal of Clinical Sleep Medicine provides a useful perspective on the issue of hypopnea definition by including outcome data based on the two different scoring criteria. The study was a retrospective observational cohort study from three Veterans Affairs medical centers. A total of 1,400 patients were included in the analysis. As expected, using hypopnea criteria based on 3% desaturation or an arousal identified an additional 175 patients diagnosed with sleep-disordered breathing among 468 previously negative sleep studies. It was noted that using the 3% criteria alone (excluding arousals) identified only 36.6% of these patients. This observation implies that many patients would not be diagnosed by type III in-home testing. The study demonstrated that using the criteria that requires 3% desaturation or an arousal identified a group of patients who experience excessive daytime sleepiness but do not appear to be at increased risk of cardiovascular disease. The authors also suggest that other polysomnographic features that are not captured by a summary AHI may be important for risk stratification. A recent paper that identifies the importance of respiratory event duration may be more predictive of mortality than the AHI.11 Perhaps counterintuitively, it appears that the shorter the event duration the greater the all-cause mortality.The study by Won et al. is the first paper that suggests a rationale for using both criteria in the evaluation of polysomnography results to bet