Comparing Initial Surgery versus Fibrinolytics for Pleural Space Infections: A Retrospective Multicenter Cohort Study

Rationale: When drainage of complicated pleural space infections alone fails, there exists two strategies in surgery and dual agent-intrapleural fibrinolytic therapy; however, studies comparing these two management strategies are limited. Objectives: To determine the outcomes of surgery versus fibrinolytic therapy as the primary management for complicated pleural space infections (CPSI). Methods: A retrospective review of adults with a CPSI managed with surgery or fibrinolytics between 1/2015 and 3/2018 within a multicenter, multistate hospital system was performed. Fibrinolytics was defined as any dose of dual-agent fibrinolytic therapy and standard fibrinolytics as 5-6 doses twice daily. Treatment failure was defined as persistent infection with a pleural collection requiring intervention. Crossover was defined by any fibrinolytics after surgery or surgery after fibrinolytics. Logistic regression with inverse probability of treatment weighting (IPTW) were employed to account for selection bias effect of management strategies in treatment failure and crossover. Results: We identified 566 patients. Surgery was the initial strategy in 55% (311/566). The surgery group had less additional treatments (surgery: 10% [32/311] versus fibrinolytics: 39% [100/255], P < 0.001), treatment failures (surgery: 7% [22/311] versus fibrinolytics: 29% [74/255], P < 0.001), and crossovers (surgery: 6% [20/311] versus fibrinolytics: 19% [49/255], P < 0.001). Logistic regression analysis with IPTW demonstrated a lower odds of treatment failure with surgery compared with any fibrinolytics (odds ratio [OR], 0.20; 95% confidence interval [CI], 0.10-0.30; P < 0.001); and compared with standard fibrinolytics (OR, 0.20; 95% CI, 0.11-0.35; P < 0.001). Conclusions: Although there is a lack of consensus as to the optimal management strategy for patients with a CPSI, in surgical candidates, operative management may offer more benefits and could be considered early in the management course. However, our study is retrospective and nonrandomized; thus, prospective trials are needed to explore this further.