Intrapleural thrombolytics for parapneumonic effusion: A network meta-analysis

Abstract

Objectives: Intrapleural thrombolytics have been trialed for facilitating pleural fluid drainage in patients with complicated parapneumonic effusion. The present study is a network meta-analysis of randomized clinical trials (RCTs) that have evaluated these thrombolytics. Methods: Electronic databases (Medline, Cochrane CENTRAL, and Google Scholar) were searched for appropriate RCTs evaluating the therapeutic effect of thrombolytics in patients with complicated parapneumonic effusion. Mortality, proportion of patients referred for surgical intervention, and serious adverse events were the outcome measures. Random-effects model was used for generating direct and mixed treatment comparison pooled estimates. Grading of the evidence for key comparisons was carried out. Odds ratio with 95% confidence intervals was used for representing the pooled estimates. Results: Seventy-six studies were retrieved with the search strategy of which 16 were included. No significant differences were observed for mortality. Compared to normal saline, significantly less proportion of patients was referred for surgical intervention with streptokinase (0.4, 0.2 to 0.8), urokinase (0.4, 0.2 to 0.8), alteplase (0.3, 0.1 to 0.7), and alteplase + DNase (0.2, 0.1 to 0.7). DNase alone increased the risk of referral to surgical intervention (3.4, 1.5 to 7.6). Only streptokinase was observed with increased risk of serious adverse events compared to normal saline (2.8, 1.1 to 7.1) and alteplase (6.7, 1.1 to 39.9). Moderate quality of evidence was observed for streptokinase with normal saline for the proportion of patients referred for surgical intervention while either low or very low quality strength was observed for all other comparisons. Conclusion: Streptokinase, urokinase, alteplase, and alteplase + DNase was observed with reduced proportions of patients referred for surgical interventions when used intrapleural in patients with parapneumonic effusion. Alteplase + DNase is likely to outperform others as it was observed with the least risk of patients referred for surgical interventions. Until additional data emerges that changes the pooled estimates, thrombolytics other than streptokinase are preferred due to increased risk of serious adverse events.