Central sleep apnea treatment in patients with heart failure with reduced ejection fraction: a network meta-analysis

Adaptive servo-ventilation (ASV) is contraindicated for the treatment of central sleep apnea (CSA) in patients with heart failure with reduced ejection fraction (HFrEF), limiting treatment options. Though continuous positive airway pressure (CPAP), bi-level PAP with back-up rate (BPAP-BUR), and transvenous phrenic nerve stimulation (TPNS) are alternatives, not much is known about their comparative efficacies, which formed the basis of conducting this network meta-analysis. We sought to analyze their comparative effectiveness in reducing apnea hypopnea index (AHI). Additionally, we also studied their comparative effectiveness on subjective daytime sleepiness as assessed by Epworth sleepiness score (ESS).

Randomized controlled trials (RCTs) from PubMed were analyzed in a network meta-analysis and relative superiority was computed based on P-score ranking and Hasse diagrams.

Network meta-analysis based on 8 RCTs showed that when compared to guideline-directed medical therapy (GDMT-used as a common comparator across trials), reduction in AHI by ASV (- 26.05 [- 38.80; - 13.31]), TPNS (- 24.90 [- 42.88; - 6.92]), BPAP-BUR (- 20.36 [- 36.47; - 4.25]), and CPAP (- 16.01 [- 25.42; - 6.60]) were statistically significant but not between the interventions. Based on 6 RCTs of all the interventions, only TPNS showed a statistically significant decrease in ESS (- 3.70 (- 5.58; - 1.82)) when compared to GDMT, while also showing significant differences when compared with ASV (- 3.20 (- 5.86; - 0.54)), BPAP-BUR (- 4.00 (- 7.33; - 0.68)), and CPAP (- 4.45 (- 7.75; - 1.14)). Ranking of treatments based on Hasse diagram, accounting for both AHI and ESS as outcomes for relative hierarchy showed relative superiority of both ASV and TPNS over BPAP-BUR and CPAP.

Results indicated relative superiority of TPNS and ASV to BPAP-BUR and CPAP in their effects on AHI and ESS.