Effect of phrenic nerve stimulation on patients with central sleep apnea: A meta-analysis.

Patients with central sleep apnea (CSA) have a lower quality of life and higher morbidity and mortality. Phrenic nerve stimulation (PNS) is a novel treatment for CSA that has been shown to be safe. However, the effects of PNS on sleep changes are still under debate. This meta-analysis was performed to evaluate the efficacy of PNS in patients with CSA. PubMed, Scopus, EMBASE, Cochrane Central Register of Controlled Trials (CENTRAL) and Web of Science databases were searched for relevant studies published. We performed random-effects meta-analyses of the changes in apnea-hypopnea index (AHI), central apnea index (CAI), Arousal Index, percent of sleep with O2 saturation <90% (T90), Epworth Sleepiness Scale (ESS) and sleep efficiency. Ten studies with a total of 580 subjects were analyzed. Overall meta-analysis showed AHI [SMD: -2.24, 95% confidence interval (CI): was -3.11 to -1.36(p＜0.00001)], CAI [SMD: -2.32, 95% CI: -3.17 to -1.47 (p＜0.00001)] and Arousal Index (p = 0.0002, SMD (95% CI) -1.79 (-2.74 to -0.85)) significantly reduced after PNS. No significant changes were observed in T90, ESS and sleep efficiency (p > 0.05). Meta-analysis of observational studies demonstrated AHI, CAI and Arousal Index had a decreasing trend between before and after PNS (all, p＜0.05). However, ESS and T90 did not change significantly after PNS (p > 0.05). Meta-analysis of RCTs showed that CSA patients had trends of a lower AHI (I2 = 0%), CAI (I2 = 74%), Arousal Index (I2 = 0%), T90 (I2 = 0%) and ESS (I2 = 0%) after PNS (all, p＜0.05). The use of PNS appears to be safe and feasible in patients with CSA. However, larger, independent RCTs are required to investigate the efficacy and long-term effect of PNS and more attention should be paid to T90 and ESS.