Comparative Effectiveness of Interventions in Initial Management of Spontaneous Pneumothorax: A Systematic Review and a Bayesian Network Meta-analysis.

Mummadi, Srinivas R; de Longpre', Jennifer; Hahn, Peter Y

STUDY OBJECTIVE The best initial strategy for nontension symptomatic spontaneous pneumothorax is unclear. We performed a systematic review and meta-analysis to identify the most efficacious, safe, and efficient initial intervention in adults with nontension spontaneous pneumothorax. METHODS MEDLINE, Scopus, Web of Science, and ClinicalTrials.gov were searched from January 1950 through December 2019 (print and electronic publications). Randomized controlled trials evaluating needle aspiration, narrow-bore chest tube (<14 F) with or without Heimlich valve insertion, and large-bore chest tube (≥14 F) insertion in spontaneous pneumothorax were included. Network meta-analyses were performed with a Bayesian random-effects model. RESULTS Twelve studies were included in this review (n=781 patients). Analyses of efficacy (n=12 trials) revealed no significant differences between the interventions studied: narrow- versus large-bore chest tubes, odds ratio (OR) 1.05 (95% credible interval [CrI] 0.38 to 2.87); large-bore chest tube versus needle aspiration, OR 1.25 (95% CrI 0.65 to 2.62); and narrow-bore chest tube versus needle aspiration, OR 1.32 (95% CrI 0.54 to 3.42). Analyses of safety (n=10 trials) revealed a significant difference between needle aspiration and large-bore chest tube interventions: OR 0.10 (95% CrI 0.03 to 0.40). No differences were observed in needle aspiration versus narrow-bore chest tube (OR 0.29 [95% CrI 0.05 to 1.82]), and narrow- versus large-bore chest tube comparisons (OR 0.35 [95% CrI 0.07 to 1.67]). Analyses of efficiency were not pursued because of variation in reporting the length of stay (n=12 trials). Narrow-bore chest tube (<14 F) had the highest likelihood of top ranking in terms of immediate success (surface under the cumulative ranking curve=64%). Needle aspiration had the highest likelihood of top ranking in terms of safety (surface under the cumulative ranking curve=95.8%). CONCLUSION In the initial management of nontension spontaneous pneumothorax, the optimal strategy between the choices of a narrow-bore chest tube (<14 F, top ranked in efficacy) and needle aspiration (top ranked in safety) is unclear. Complications were more common in large-bore chest tube (≥14 F, including 14-F tube) insertions compared with needle aspiration.