P119 CPAP and adverse cardiovascular events in OSA: Are participants of randomized trials representative of sleep clinic patients?

INTRODUCTION: Randomized controlled trials (RCTs) have shown no reduction in adverse cardiovascular (CV) events in patients randomized to continuous positive airway pressure (CPAP) therapy for obstructive sleep apnea (OSA). This study examined whether randomized study populations were representative of OSA patients attending a sleep clinic. METHODS: Sleep clinic patients were 3,965 consecutive adults diagnosed with OSA by in-laboratory polysomnography from 2006–2010 at a tertiary hospital sleep clinic. Characteristics of these sleep clinic OSA patients were compared with participants of 5 well-known RCTs examining the effect of CPAP on adverse CV events in OSA. We determined the percentage of patients with severe (apnea hypopnea index, [AHI]≥30/h) or any OSA (AHI≥5/h) who met the selection criteria of each RCT, as well as identified those criteria that excluded the most patients. RESULTS: Compared to RCT participants, sleep clinic OSA patients were younger, sleepier, more likely to be female and less likely to have established CV disease. The percentage of patients with severe or any OSA who met the RCT selection criteria ranged from 1.2% to 20.2% and 0.8% to 21.1%, respectively. The selection criteria that excluded most patients were pre-existing CV disease, symptoms of excessive sleepiness, nocturnal hypoxemia and co-morbidities. DISCUSSION: A minority of sleep clinic patients diagnosed with OSA meet the selection criteria of RCTs of CPAP on adverse CV events in OSA. OSA populations in RCTs differ considerably from typical sleep clinic OSA patients. This suggests that the findings of RCTs may not be generalisable to most sleep clinic OSA patients.