GASTROJEJUNAL ANASTOMOTIC STENOSIS: A MATCHED COHORT STUDY COMPARING EFFICACY AND SAFETY OF LUMEN-APPOSING METAL STENT PLACEMENT TO STANDARD BALLOON DILATION
Kelly Hathorn*1, Pedro Cortés2, Thomas R. McCarty1, Thomas J. Wang3, Ahmad Najdat Bazarbashi1, Marvin Ryou1, Walter W. Chan1, Christopher C. Thompson1
1Brigham & Women's Hospital, Boston, MA; 2Harvard Medical School, Boston, MA; 3Massachusetts General Hospital, Boston, MA
Background: Lumen apposing metal stents (LAMS) have been used with increased frequency for management of gastrojejunal anastomotic (GJA) stenosis following Roux-en-Y Gastric Bypass (RYGB). Less is known regarding the comparative outcomes of LAMS placement with balloon dilation. Thus, the aim of this study was to evaluate the efficacy and safety of LAMS placement compared with standard balloon dilation for treatment of benign GJA stenosis.
Methods: This is a retrospective matched cohort study. Adults with a history of RYGB whom underwent endoscopic treatment were included. Patients were matched in 1:1 fashion by sex, GJA size (± 2 mm), and the number of interventions prior to ‘index’ procedure, defined as either LAMS placement or the corresponding balloon dilation (i.e. Patient X, with LAMS after 2 failed dilations, was matched with Patient Y, with a 3rd balloon after 2 failed dilations.) Patient demographics, smoking status, and non-steroidal anti-inflammatory (NSAID) use were reviewed. Size of the gastric outlet (extracted from procedure note or estimated by two reviewers from endoscopy imaging), and presence of ulcer was recorded. GJA stenosis was defined as the inability to pass the standard upper endoscope. The primary outcome was number of interventions required following the index procedure for symptomatic GJ stenosis. Secondary outcomes were procedure-related complications.
Results: A total of 95 patients who underwent endoscopic therapy for GJA stenosis were identified. Ultimately, 37 patients who underwent balloon dilation and LAMS placement were matched in 1:1 fashion with patients who underwent balloon dilation only. Baseline characteristics are summarized in Table 1. Overall, a total of 288 interventional procedures were performed (209 balloon dilations, 55 LAMS placements) with a mean number of interventions per patient of 3.89 ±2.17. After the index procedure, 23 (62.16%) patients in balloon+LAMS compared to 34 (91.89%) patients in balloon only cohort required a repeat interventional procedure (p=0.0047); the number of interventional procedures required per patient was lower in the balloon+LAMS group (1.31 ±1.55 (range 0-7) vs. 2.35 ±1.87 (range 0-8) in the balloon only group (p=0.0112, Figure 1). 14 patients (37.84%) in the balloon only cohort ultimately underwent surgical revision, compared to 3 (8.11%) in the balloon+LAMS cohort (p=0.0047). On multivariable regression controlling for age, obesity, GJA size, smoker, and presence of ulcer, LAMS is an independent predictor of decreased risk for post-index interventions (OR 0.261, CI 0.102-0.668).
Conclusion: This study demonstrates that patients with GJA stenosis treated with balloon dilation and LAMS require fewer interventional procedures post-LAMS placement compared to conventional balloon dilation therapy. Furthermore, LAMS appears to reduce the need for surgical revision.
Table 1: Patient Characteristics 
Figure 1: Interventional Procedures Post-Index Procedure
Back to 2020 Abstracts