CPAP vs NIV for long term treatment of obesity hypoventilation syndrome: The results of the Pickwick randomized controlled trial

Rationale: Noninvasive ventilation (NIV) and continuous positive airway pressure (CPAP) are the most commonly prescribed treatment modalities for patients with obesity hypoventilation syndrome (OHS). Despite differences in cost and complexity between NIV and CPAP, there are no long-term effectiveness studies comparing the two treatment modalities.Methods: We performed a large multicenter randomized open-label controlled trial (Pickwick study) to determine the comparative long-term effectiveness of NIV and CPAP using hospitalization days as the primary outcome measure.Sequentially screened OHS patients with severe obstructive sleep apnea (OSA) were randomized to NIV or CPAP and were followed for at least three years. Hospital resource utilization, mortality, incident cardiovascular events, dropouts, compliance, and side effects were quantified. Statistical analysis was performed using intention-to-treat procedure evaluating the incidence density ratio and survival analysis. Adjusted analysis was performed by Poisson, negative binomial and Cox regressions.Results: In total, 363 patients were screened, 215 were randomized and 202 were available for primary analysis. The median (IQR) follow-up was 5.42 (2.17) years. Hospitalization days per 100 person-years (95% CI) was 158.5 days (148.3-169.3) for CPAP and 156.2 days (155.5-167.5) for NIV [adjusted hazard ratio 0.68 (0.14- 3.35); p= 0.638]. Other hospital resource utilization including hospitalizations, emergency department visits and ICU days had similar comparative results. Incident cardiovascular events per 100 person-years (95% CI) was 5.1 (2.95-8.1) for CPAP and 7.46 (4.62-11.4) for NIV [adjusted hazard ratio 1.4 (0.73-2.71); p= 0.315]. Mortality rate was 14.7% for CPAP and 11.3% for NIV [adjusted hazard ratio 0.73 (0.33-1.62); p= 0.439] (see Figure). Dropouts, adherence to positive airway pressure (PAP) therapy and secondary effects were similar between the two groups. Subgroups of PAP treatment adherence analysis (first tertile vs second and third tertiles) showed similar results between NIV and CPAP. Intra-group comparisons, however, revealed significantly improved hospital resource utilization and survival in both NIV and CPAP subgroups that were adherent to therapy (see Figure: Cumulative survival for NIV and CPAP arms (upper). Cumulative survival between both adherence subgroups (first tertile 1 vs second and third tertiles) in CPAP (left lower) and NIV (right lower) arms. Note that the first tertile corresponds with a treatment use 4 hours/night and second + third tertiles with >4 hours/night). Conclusion: NIV and CPAP have similar long-term effectiveness. Given that CPAP therapy is less costly and easier to implement, it should be the preferred treatment modality for patients with OHS and concomitant severe OSA (Figure Presented).