Effect of CPAP therapy on blood pressure in patients with OSA: a worldwide individual patient data meta-analysis

Abstract Background Obstructive sleep apnoea (OSA) is a well-known cardiovascular risk factor and is frequently associated with hypertension. The treatment of OSA with continuous positive airway pressure (CPAP) can resolve apnoeas, hypoxia, and sympathetic hyperactivation. Meta-analyses of randomised controlled trials (RCTs) show that CPAP treatment modestly reduces blood pressure (BP) with a pronounced heterogeneity among different OSA patient subgroups. Purpose To study the effects of CPAP treatment on BP changes, and to identify predictors of BP response to CPAP through analysis of individual patient data from published RCTs. Methods Study groups who had published RCTs on the effect of CPAP therapy on BP in OSA were invited to share the individual patient data (i.e. BP values, history of hypertension and antihypertensive drugs use). The outcomes of interest were the group-difference (BP at follow-up – BP at baseline) in office and out-of-office BP. Data analysis was performed with a one-step approach using a linear regression model, with treatment as covariate, including a random effect. An additional analysis was performed by stratifying patients in three subgroups: normotension, controlled hypertension and uncontrolled hypertension, based on BP at baseline, history of hypertension or treatment with antihypertensive drugs. Lastly, we evaluated the determinants of CPAP associated BP changes. Results Individual patient data from 34 parallel-group RCTs (n=7,456, 72% males, body mass index, BMI 31.9±14.8 kg/m2, age 58.6±10.9 years, apnoea-hypopnoea index, AHI 35.5±20.6 /hour) were analysed. The overall CPAP effect was larger for 24-BP compared to office BP measurements (Table 1), the highest reductions being observed for nocturnal BP. The stratified analysis showed a consistent BP lowering effect in the group of patients with uncontrolled hypertension whereas there was no significant BP change in those with controlled hypertension or in normotensives (figure 1). Higher BP levels at baseline and history of hypertension were independent predictors of BP reduction by CPAP for office and 24-BP measurements. Younger age and more severe nocturnal oxygen desaturations predicted a larger treatment effect on office BP only. Conclusions BP response to CPAP in OSA patients varies in different patient groups as a function of baseline BP, the greatest reduction being observed in younger patients with uncontrolled BP. Thus, our individual patient data meta-analysis indicates that younger patients with uncontrolled hypertension and more severe nocturnal oxygen desaturations measured by minimum SpO2 are likely to benefit the most from CPAP treatment in terms of BP reduction. Funding Acknowledgement Type of funding sources: None.