Gastric peroral endoscopic pyloromyotomy for delayed gastric conduit emptying: outcomes and predictors of treatment response.

Abstract

Background: Delayed gastric conduit emptying (DGCE) is common in patients following esophagectomy, and its treatment remains challenging. This study evaluated efficacy and safety of gastric peroral endoscopic pyloromyotomy (GPOEM) and predictors of treatment response.

Methods: We prospectively included patients with symptoms of DGCE following esophagectomy scheduled for GPOEM in two tertiary referral centers. EndoFLIP measurements were conducted before and 3 months after GPOEM. Gastric emptying scintigraphy (GES) was performed prior to, 3 and 12 months after GPOEM. Efficacy was evaluated by clinical success, defined as a score of 1 or 2 on a six-point Likert scale 3 months after GPOEM. Predictive factors for clinical success, including pyloric distensibility, GES and response to prior pylorus-directed therapy (botulinum injection and pyloric dilatation), were evaluated by exploratory analyses.

Results: The technical success rate of GPOEM was 95% (19/20; 95% CI 75-100). At 3-, and 12-months post-procedure, the clinical success rates were 68% (13/19; 95% CI 37-84) and 44% (8/18; 95% CI 22-67). Distensibility prior to GPOEM did not differ significantly between patients with and without clinical success (3.7 vs. 4.0 mm²/mmHg, p = 0.70). Pre-GPOEM 4-h gastric retention was greater in patients with clinical success than those without, but this was not significant (33 vs 16%, p = 0.12). Although there was no significant association between prior response to pylorus-directed therapy and GPOEM success (p = 0.10), the positive and negative predictive value were 100% (9/9; 95% CI 66-100) and 40% (2/5; 95% CI 5-85).

Conclusion: GPOEM is feasible and safe in patients with DGCE following esophagectomy, demonstrating initial efficacy in two-thirds of patients, although this declined to less than half at 1-year follow-up. In exploratory analyses, no predictors of treatment response were identified. Larger studies are needed to investigate if response to pylorus-directed therapy may predict an increased likelihood of treatment response.