Algorithmic approach to utilization of CT scans for detection of internal hernia in the gastric bypass patient

Abstract Background While surgical exploration remains the gold standard for diagnosing internal hernia (IH) after certain bariatric surgeries, decisions for operative intervention are often based on computed tomography (CT) findings. Objectives The aim of this study is to review our institutional experience and create an algorithm to approach patients presenting with abdominal pain and/or emesis after certain bariatric procedures. Settings University Hospital Methods Following institutional review board approval, a retrospective chart review of all patients presenting with obstruction symptoms after laparoscopic Roux-en-Y gastric bypass (LRYGB) was performed at 2 institutions from 2008 to 2013. Patients without CT scans or with incidental hernia defect findings were excluded. CT and intraoperative findings were compared via univariate statistical analysis. Results Fifty-two patients who underwent an operation for a suspected IH were identified. Of the 50 patients, 25 (50%) had IH at operation. Twenty-nine patients (58%) had positive CT scans read for IH and/or obstruction. Of these 29, 19 (66%) were found to have IH at operation and 10 (34%) underwent negative diagnostic laparoscopy. Of the 21 patients with negative CT scans, 6 (29%) had IH at operation versus 15 (71%) who were negative. The sensitivity of CT scan to detect an internal hernia is 76% with 95% confidence interval (CI) [53% to 90%] and specificity is 60% with 95% CI [39% to 78%]. Sensitivity increased to 96% with 95% CI [78% to 99.8%] when combining CT scans with neutrophilia findings. Conclusion Positive CT scans are sensitive for IH but not specific. CT scans will not detect IH in 1:4 patients; despite negative findings, surgical exploration should remain the gold standard for patients with acute abdominal pain after LRYGB or biliopancreatic diversion when IH is a consideration.