Predictors of stent dysfunction after self-expandable metal stent placement for malignant gastric outlet obstruction: tumor ingrowth in uncovered stents and migration of covered stents.

Background: Endoscopic metallic stenting is widely accepted as a palliation therapy for malignant gastric outlet obstruction (GOO). However, the predictors of stent dysfunction have not been clarified. We aimed to evaluate the predictors, especially tumor ingrowth in uncovered self-expandable metallic stents (U-SEMS) and migration of covered self-expandable metallic stents (C-SEMS), which are the main causes related to the stent characteristics.

Methods: In this multicenter retrospective study, we compared patients with U-SEMS and C-SEMS in terms of clinical outcomes, and predictors of stent dysfunction.

Results: In total, 252 patients (126 with U-SEMS and 126 with C-SEMS) were enrolled. There were no significant differences in technical success, clinical success, GOO score, or time to stent dysfunction. Tumor ingrowth was significantly more frequent in U-SEMS (U-SEMS, 11.90% vs. C-SEMS, 0.79%; p = 0.002), and stent migration was significantly more frequent for C-SEMS (C-SEMS, 8.73% vs. U-SEMS, 0.79%; p = 0.005). Karnofsky performance status (p = 0.04), no presence of ascites (p = 0.02), and insufficient (<30%) stent expansion (p = 0.003) were significantly associated with tumor ingrowth in U-SEMS. Meanwhile, a shorter stent length (p = 0.05) and chemotherapy (p = 0.03) were predictors of C-SEMS migration.

Conclusions: Both U-SEMS and C-SEMS are effective with comparable patencies. Tumor ingrowth and stent migration are the main causes of stent dysfunction for U-SEMS and C-SEMS, respectively. With regard to stent dysfunction, U-SEMS might be a good option for patients receiving chemotherapy, while C-SEMS with longer stents for patients in good condition. (Clinical trial registration number: UMIN000024059).