Test Characteristics of Abdominal Computed Tomography for the Diagnosis of Gastro-gastric Fistula in Patients with Roux-en-Y Gastric Bypass.

Abstract

Gastrogastric fistulae (GGF) occur in 1-6% of Roux-en-Y gastric bypass (RYGB) patients. Many patients undergo abdominal computed tomography (CT) as an initial test owing to its wide availability; however, CT diagnostic accuracy for GGF is unclear. Our aim was to evaluate test characteristics of abdominal CT compared to upper gastrointestinal series (UGI) and esophagogastroduodenoscopy (EGD) for diagnosing GGF using surgery as a gold standard. Retrospective review of RYGB patients who underwent abdominal CT with oral contrast within 1 year. Demographics, weight parameters, and symptoms were collected. Surgery within 1 year of the diagnostic tests was included as the gold standard comparison. Primary outcomes included CT sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and diagnostic accuracy (DA) for GGF. One hundred thirty-seven patients were included, where 42 (30.1%) had positive CT and 95 (69.3%) had negative CT for GGF. Compared to surgical confirmation, CT abdomen with PO contrast had sensitivity of 73.1% (59-84.4), specificity of 95.2% (88.3-98.7), PPV 90.5% (77.4-97.3), NPV of 85.1% (76.3-91.2), and DA 89.7%. UGI series had sensitivity of 58.5% (42.1-73.7), specificity of 98.8% (93.5-99.9), PPV of 96% (79.7-99.9), NPV of 82.8% (73.9-89.7), and diagnostic accuracy (DA) of 85.4%. EGD had sensitivity of 78.3% (63.6-89.1), specificity of 98.8% (93.5-99.9), PPV 97.3 (85.8-99.9), and DA 91.5%. There were no significant differences in diagnostic test characteristics among modalities. Abdominal CT with oral contrast has similar diagnostic test characteristics to UGI and EGD at detecting GGF when using surgical diagnosis as a gold standard.