Efficacy and Safety of Lumen-apposing Metal Stents in EUS Guided Transmural Drainage of Pancreatic Fluid Collections: A Systematic Review and Meta-analysis

Abstract

Introduction: Endoscopic ultrasound (EUS) guided transmural drainage has been increasingly utilized as a first line therapeutic modality for drainage of pancreatic fluid collections (PFC). Initially this procedure was performed using multiple plastic stents or fully covered biliary metallic stents. Recently, lumen apposing metal stents (LAMS) have been used for the two main types of PFC i.e. pancreatic pseudocysts (PP) and walled of necrosis (WON) with variable success. We conducted a systematic review and metaanalysis to evaluate the cumulative efficacy (technical success & clinical success) and safety (adverse events requiring intervention and /or hospitalization) of LAMS in the management of PFC. Methods: We searched Medline, Embase and Scopus from inception to June 1, 2016 to identify studies (with ≥10patients) reporting technical success, clinical success and adverse events (AE) of EUS guided transmural drainage of PFC using LAMS. Weighted pooled rates (WPR) were calculated for technical success, clinical success and AE for all the studies evaluating the performance of LAMS in PFC. Subgroup analysis was conducted based on clinical success in PP and WON. These were analyzed using random effects model. Proportionate difference was also calculated to compare the efficacy of LAMS in PP and WON. Publication bias was assessed using Egger's test and adjusted using trim and fill test. Quality assessment of studies was done using Newcastle Ottawa Scale (NOS). Results: A total of 9 studies with 489 patients were involved in the analysis. Two studies were of high quality while 7 were of moderate quality on NOS assessment. The WPR for technical success of LAMS in PFC management was 97% (95%, 98%), with no heterogeneity (I2=0%) and no publication bias (P=0.07). WPR for clinical success in PFC was 88% (84%,92%), with low heterogeneity (I2=32%), with publication bias (P=0.04). Adjusted WPR for clinical success was 86%(81%,90%). WPR for clinical success in PP was 88%(78%,94%), whereas WPR for clinical success in WON was 86%(83%,90%) with no heterogeneity. Proportionate difference for clinical success in WON as compared to PP was 1.5% (-6%, 9%) (P=0.72). WPR for direct endoscopic necrosectomy was 67.7% (50%,80%) with substantial heterogeneity (I2=80%). WPR for AE was 13% (8%, 18%), with moderate heterogeneity (I2=46%) and no publication bias (P=0.12). Most common AE requiring intervention was stent migration (3.8%), followed by infection (3.2%), bleeding (1.8%) and stent occlusion (1.6%). Conclusion: LAMS appears to be a very promising option for EUS guided transmural drainage of PFC. It has equal efficacy in PP and WON and has a relatively low adverse event rate (requiring further intervention/hospitalization).Figure 1Figure 2