Su1287 TIMING OF DIRECT ENDOSCOPIC NECROSECTOMY THROUGH LUMEN-APPOSING METAL STENTS FOR WALLED-OFF PANCREATIC NECROSIS IMPACTS TIME TO RESOLUTION

Abstract

Lumen-apposing metal stents (LAMS) are effective in resolving walled-off pancreatic necrosis (WON) and provide a conduit for direct endoscopic necrosectomy (DEN). However, the decision to perform, or the timing of DEN to facilitate resolution, is not well-defined. We aimed to determine if early DEN was associated with shorter time to resolve WON. A retrospective review of our endoscopy database identified WON patients who underwent LAMS. Patient demographics, WON characteristics, number of DEN, and outcomes were compiled. Subgroups by timing of DEN included concomitant DEN (CDEN; index DEN at time of LAMS placement) or delayed DEN (DDEN; index DEN at a later date). Patient demographics, procedure details, and outcomes were compared between groups using the Wilcoxon rank-sum test. Linear regression was performed to identify predictors for time to WON resolution. Variables significant at p <0.2 on univariate analysis were included in the multivariate model. Cox proportional hazards regression was performed to identify risk factors for resolution. Variables significant at p <0.2 on univariate analysis were included in the multivariate model. Nelson-Aalen cumulative hazards curves were constructed to depict time to resolution stratified by timing of index DEN. The primary outcome was time to resolution of WON, determined radiographically or endoscopically. Secondary outcomes included adverse events (see table) and total number of DEN sessions required. Between 1/11 and 6/19, 175 patients underwent LAMS for peripancreatic fluid collections. 155 (88.6%) were WON of which 50/155 (32.3%; 80% male, 52.5 years old) required DEN and comprised the study cohort. The size of WON and use of double pigtail stents through the LAMS was similar between groups. Overall, the mean time to resolution of WON (see figure) was 90.6 days with a mean of 44.5 days in CDEN vs. 100.7 days in DDEN (p=0.02) despite similar number of DEN between groups (p= 0.64). All patients in both groups had radiographic or endoscopic WON resolution. Mean length of stay was also significantly different between groups at 0.9 vs 14.7 days (p=0.03). Total number of adverse events were 5 in the CDEN vs 45 in the DDEN (p=0.45). In multivariate analysis, higher number of double pigtail stent usage through LAMS was associated with a longer time to WON resolution (Multivariate β coefficient (95% CI): 51.4 (22.8-80.1), p=0.005) suggesting greater complexity. 1) In our cohort of WON patients treated with LAMS, only one-third required DEN. 2) Despite a similar number of total DEN, concomitant LAMS and DEN performance during the index endoscopic session resulted in a significantly shorter time to resolution and did not require percutaneous drainage compared with DEN performed at a subsequent session. 3) The adverse event rates between groups were similar.FigureKaplan-Meier Curves of Resolution of Walled-off Necrosis by Timing of Necrosectomy (top pane) and Cumulative Probability of Resolution of Walled-off Necrosis (bottom pane)View Large Image Figure ViewerDownload Hi-res image Download (PPT)