Abstract
BACKGROUND: Little is known about patient outcomes following treatment of malignant pleural effusions (MPE) in the real-world setting. RESEARCH QUESTION: We aimed to compare post-procedure all-cause mortality between individuals who received indwelling pleural catheter (IPC) insertion versus chemical pleurodesis for managing MPEs. STUDY DESIGN AND METHODS: We performed a retrospective population-based study using provincial health administrative data (Ontario, Canada) of adults with a MPE who underwent IPC insertion or chemical pleurodesis between 2015 and 2019. Individuals were followed until death or March 31, 2021. Difference in post-procedure mortality was calculated using inverse probability of treatment weighting (IPTW)-adjusted Cox proportional hazard regression analysis to balance potential confounders at baseline. RESULTS: We identified 4,790 (77.3%) individuals who received an IPC and 1,407 (22.7%) who had chemical pleurodesis for MPE. IPC insertions are increasing and chemical pleurodesis procedures are decreasing. The majority of IPCs were inserted in outpatients (61%), by pulmonologists (64.2%) and at sites with higher annual IPC volume, while chemical pleurodesis procedures were generally done by thoracic surgeons (74%) and at sites with higher annual pleurodesis volumes. In unadjusted comparison median time from initial cancer diagnosis to intervention was significantly longer in the IPC group (244 days, interquartile range [IQR]:33-903) compared to pleurodesis group (81 days, IQR:10-737; p < 0.0001). Unadjusted median time from index procedure to death was significantly longer in the pleurodesis group (165[IQR:48-457] days vs. 81[IQR:29-256] days, p < 0.0001), however the difference between groups became insignificant after the IPTW was applied (HR 1.27, 95%CI 0.95-1.69). 35% of IPCs were removed prior to death or end of follow-up. INTERPRETATION: After adjusting for differences in baseline characteristics there was no difference in post-procedure mortality between IPC and chemical pleurodesis groups. In the real world, there are significant differences in the characteristics of patients who receive these two procedures and notable regional practice variation between procedure use. Future research should evaluate these variations in care and their effect on patient outcomes.