S29 Predictors of continuous positive airways pressure usage at six months in minimally symptomatic patients. Further data from the MOSAIC trial: Abstract S29 Table 1

Abstract

Introduction Continuous positive airways pressure (CPAP) is the standard treatment for obstructive sleep apnoea (OSA). However, compliance with CPAP is variable, and usually found to be lower in the relatively asymptomatic; in some studies only ODI has been predictive (1). We aimed to establish the significant predictors of subsequent compliance, in less symptomatic patients, by using compliance data from the Multicentre Obstructive Sleep Apnoea International Cardiovascular (MOSAIC) trial, which specifically recruited minimally symptomatic patients. Methods Patients were randomized to 6-months of CPAP or standard care(2) if they had an oxygen desaturation index (ODI) of > 7.5 hour due to OSA on a baseline sleep study, but had insufficient daytime OSA symptoms to warrant CPAP. All patients had medical history, Pocock cardiovascular risk score (calculated 5 year risk of fatal cardiovascular event), Epworth Sleepiness Score (ESS), Sleep Apnoea Quality of Life Index (SAQLI) and SF-36 completed. Repeat overnight pulse oximetry was performed after entry for uniformity of trial ODI. Usage data from those randomized to CPAP were downloaded, and the last week of the trial used to generate average hours/ night (h/n) usage. Data from the last week was thought to most accurately reflect continued CPAP usage beyond 6 months. Study withdrawals were assumed not to have used CPAP. Patients were also asked if they were willing to continue CPAP treatment beyond 6 months. Statistical analysis was performed using appropriate tests (SPSS software). Results Baseline characteristics from all 195 patients randomized to CPAP are summarized below. The median CPAP usage was 2:12 h/n (IQR 0:00 to 5:27). 138 (71%) patients were willing to continue CPAP beyond 6 months. BMI was significantly, but weakly, correlated to CPAP usage (BMI r=0.15 p=0.04). Males had significantly greater median usage at 2:48 h/n (IQR 0:00, 5:47) compared to 0:48 h/n in females (IQR 0:00, 3:06) (p=0.02). There were no other significant predictors of usage (age, gender, ODI, ESS, sleep symptoms, smoking status, cardiovascular risk, ethnicity, SAQLI, SF-36). Logistic regression did not reveal any variables that predicted willingness to continue CPAP beyond 6 months. Conclusions There were no strong predictors of either CPAP usage, or a willingness to continue, at six months. Both BMI and gender had weak correlations with CPAP usage. Thus a trial of CPAP is required to determine likely patient benefit in these minimally symptomatic patients, and this cannot be predicted in advance. (Table Presented). DBP - EM:201545