LAUSR

BACKGROUND Endoscopy has evolved to become first-line therapy for the treatment of post-bariatric leaks; however, many sessions are often required with variable success rates. Due to these limitations, the use of the cardiac septal defect occluder (CSDO) has recently been reported in this population. METHODS The study population was a multicenter retrospective series of patients with post-bariatric surgical leaks who underwent treatment with CSDO placement. Data on the type of surgery, previous treatment details, fistula dimensions, success rate, and adverse events were collected. Leaks were grouped according to the International Sleeve Gastrectomy Expert Panel Consensus. Outcomes included technical and clinical success and safety of the CSDO. Regression analysis was performed to determine the predictors of response. RESULTS Forty-three patients with leaks were included (31 sleeve gastrectomy and 12 Roux-en-Y gastric bypass). They were divided into acute (n = 3), early (n = 5), late (n = 23), and chronic (n = 12). Forty patients had failed previous endoscopic treatment and 3 patients had CSDO as the primary treatment. Median follow-up was 34 weeks. Technical success was achieved in all patients and clinical success in 39 patients (90.7%). All chronic, late, and early leaks were successfully closed, except one undrained late leak. The 5 patients with early leaks had an initial satisfactory response, but within 30 days, drainage recurred. The CSDOs were removed and replaced with larger-diameter devices leading to permanent defect closure. Acute leaks were not successfully closed in all 3 patients. Regression analysis showed that chronicity and previous treatment were associated with fistula closure; success rates for late/chronic leaks versus acute/early leaks were 97.1% and 62.5%, respectively (P = .0023). CONCLUSION This observational study found that the CSDO had a high efficacy rate in patients with non-acute leaks, with no adverse events. All early, late, and chronic leaks were successfully closed, except for one undrained late leak. However, early leaks required a second placement of a larger CSDO in all cases. These results suggest that the CSDO should be considered for non-acute fistula and that traditional closure methods are likely preferred in the acute and early settings.