S217 Self-Expanding Metal Stents for Proximal Malignant Colon Obstruction—A Meta-Analysis of Endoscopic Outcomes Relative to Distal Colon Obstruction

Introduction: Studies on the utility of self-expanding metal stent insertion for the treatment of proximal malignant colon obstruction (SEMS-PCO) are limited. Methods: A comprehensive search of the literature was performed of Medline and Embase to identify reports of endoscopic outcomes after SEMS-PCO (defined as obstruction proximal to the splenic flexure). Studies comparing SEMS-PCO to SEMS for distal colonic obstruction (DCO) were also included in this analysis. A meta-analysis of proportions and a comparative meta-analysis to compute odds ratios (OR) with 95% confidence intervals (CIs) was performed. Results: A total of 15 non-randomized studies enrolling 452 PCO patients were identified, 7 of which reported endoscopic outcomes of SEMS-DCO in 501 patients. In the PCO arm, 46% (30-63%) were male. Stage 4 cancer was reported in 32% (24-39%) of patients. Most obstructions were in the ascending colon (38%; 32-43%). The technical success (TS) of SEMS-PCO was 94% (91-96%) versus 96% (94-98%) for SEMS-DCO; Although the TS of SEMS-PCO was superior to SEMS-DCO on comparative analysis (OR 0.43; 0.20-0.90; I2 0%), this signal diminished upon subgroup analysis of SEMS-PCO as a bridge to surgery (BTS; OR 0.43; 0.13-1.41; I2 0%) and studies with a mixed (BTS + Palliative SEMS) cohort (OR 0.43;0.16-1.20; I2 7.94%). The clinical success (CS) was 88% for SEMS-PCO (83-93%) and SEMS-DCO (84-93%), with no significant difference in the two groups (Table 1). Adverse events (AEs) rate was 16% (9-23%) with SEMS-PCO and 20% (11-29%) with SEMS-DCO, with no significant difference in the two groups. SEMS-PCO migration rate was 24% (9-40%) versus 19% (7-31%) with SEMS-DCO. Tumor ingrowth and overgrowth rate was 40% (23-56%) and 37% (19-59%) with SEMS-PCO versus 5% (1-25%) and 67% (48-81%) with SEMS-DCO, respectively. Luminal perforation rate was 18% (7-28%) with SEMS-PCO versus 12% (3-21%) with SEMS-DCO. Endoscopic reintervention and surgical intervention due to SEMS-associated AEs were required in 47% (27-68%) and 27% (13-40%) of patients in the SEMS-PCO arm, with no significant difference upon comparison with SEMS-DCO. Conclusion: SEMS are efficacious in relieving proximal malignant colon obstruction. The adverse event profile of proximal vs distal colonic obstruction is similar despite additive technical challenges common to proximal obstructions. Randomized trials assessing stent insertion on an intention to treat basis in proximal malignant colon obstructions should validate our findings.Table 1.: Comparative Meta-Analysis of Outcomes of SEMS Insertion for The Treatment of Proximal versus Distal Malignant Colon Obstruction.