Efficacy of Postoperative Upper Gastrointestinal Series (UGI) and Computed Tomography (CT) Scan in Bariatric Surgery: a Meta-analysis on 7516 Patients

BACKGROUND: To demonstrate the lack of utility and efficacy of routine early postoperative upper gastrointestinal study (UGI) in obese patients undergoing bariatric surgery and to show the higher efficacy of CT scan in cases of clinical suspicion of a leakage, a meta-analysis was performed. MATERIALS AND METHODS: A literature search including articles published in last 18 years was performed. For both UGI and CT scan, sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) were calculated. A first analysis considered overall patients, and a second analysis considered only symptomatic patients. RESULTS: Starting from 1233 eligible citations, 18 articles, including 7516 patients, were left. The pooled sensitivity was 54% for UGI (95% CI 34-74) with a high heterogeneity (I2 = 99.8%, p < .001), whereas CT scan showed a pooled sensitivity of 91% (95% CI 89-93) significantly higher than sensitivity of UGI series (p < 0.01), with a high heterogeneity (I2 = 98.9%, p < .001). In symptomatic patients the pooled sensitivity of UGI series was significantly lower than sensitivity of CT scan [49% (95% CI 31-68) vs 94% (95% CI 92-96), p < 0.01]. PPV showed a significant difference between UGI series and CT scan (54 vs 100%, p < 0.01). Specificity for UGI series was 98.6%, and specificity for CT scan was 99.7% (p = ns); the mean NPV was 96 and 98% for UGI series and CT scan (p = ns). CONCLUSIONS: According to our results, a CT scan triggered by clinical suspicion must be considered the first-line procedure to detect a postoperative leak following primary sleeve gastrectomy or Roux-en-Y gastric bypass.