Effect of CPAP therapy on blood pressure in patients with obstructive sleep apnoea: a worldwide individual patient data meta-analysis
Martino F. Pengo, Esther I. Schwarz, Ferrán Barbé, Peter A. Cistulli, Luciano F. Drager, Cristiano Fava, Flávio Danni Fuchs, Msm Ip, Kelly A. Loffler, Macy Mei‐Sze Lui, Miguel Ángel Martínez-García, R. Doug McEvoy, Yüksel Peker, Craig L. Phillips, Timothy Quinnell, Davide Soranna, Joerg Steier, John Stradling, Antonella Zambon, Gianfranco Parati
Abstract
Background Obstructive sleep apnoea (OSA) is associated with hypertension, and OSA treatment can reduce systolic blood pressure (SBP) and diastolic blood pressure (DBP), but with a modest mean effect size and vast heterogeneity among studies. The aim of this individual patient data (IPD) meta-analysis was to understand which OSA phenotypes could benefit the most in terms of BP reduction. Methods A systematic review of randomised controlled trials that compared continuous positive airway pressure (CPAP) with either passive or active treatment was conducted. Studies were eligible if they included adult patients with OSA diagnosed by full polysomnography or cardiorespiratory polygraphy (defined as apnoea–hypopnoea index >5 events·h −1 ) and if BP was measured both before and after CPAP treatment. Results In total, 36 parallel studies (n=9434 patients) were included. CPAP treatment was associated with BP reduction in patients with uncontrolled office SBP only, while BP was not reduced by CPAP in patients with controlled BP (SBP −2.6 versus 0 mmHg; p<0.0001; DBP −1.7 versus −1 mmHg; p=0.091). Differences were seen also when BP changes were compared between patients aged ≤60 versus >60 years after multiple imputation only (p=0.0127 for SBP and p=0.017 for DBP). No differences were seen in terms of BP reduction when comparing patients with/without severe nocturnal hypoxia. Conclusions This IPD meta-analysis of the BP effects of OSA treatment with CPAP shows that OSA patients with uncontrolled BP at baseline benefit the most from CPAP therapy in terms of BP reduction. These results have important implications for the decision on how to best manage arterial hypertension associated with OSA.