Back to Annual Meeting Program

Introduction: Esophagectomy is currently the treatment modality of choice in patients with esophageal carcinoma. Post-operatively, routine fluoroscopic imaging with barium sulfate is employed in order to detect occult anastamotic leaks (AL) prior to resumption of enteral feeding. This modality is plagued by a low sensitivity, and its routine use has been called into question. Accordingly, we sought to demonstrate the clinical impact of routine barium esophagography (BE) in the post-operative management of patients undergoing esophagectomy for malignant disease.
Methods: All patients (pts) undergoing esophagectomy from 2005-11 for malignant disease at a North American university hospital were identified from a prospectively collected database. All patients were subject to BE within the first week postoperatively. Patients were dichotomized according to whether they had an AL or not and the sensitivity and specificity of barium swallow was determined. In pts who had an AL, the relationship between barium swallow results and time to AL, hospital length of stay, and start of enteral feeding was determined. Furthermore, the effect of BE results on postoperative management, defined as cessation of enteral feeding, additional interventions, or delay in discharge was recorded. Data are expressed as median (range). Mann Whitney U and Fischer’s exact test determined significance (*p<0.05).
Results: Two-hundred and twenty-seven pts underwent esophagectomy over the study period. Twenty-nine pts (12.8%) developed an AL, of which 12 (41.4%) had a positive BE, 11 (37.9%)had a negative BE and the remaining 6 (20.7%) were not subject to BE and were diagnosed either clinically (1/6), by CT (4/6), or endoscopically (1/6). AL in pts with a negative BE was confirmed either clinically (4/11 pts), by CT (5/11 pts), endoscopically (1/11 pts), or at reoperation (1/11 pts). In pts who had an AL, those with a positive BE leaked earlier than those with a negative BE (POD 7(2-8) vs POD 10.5(6-22)* respectively). The sensitivity and specificity of barium BE in this series was 36.3% and 99.7% respectively. Result of BE in pts with an AL did not correlate with hospital length of stay or date of commencement of enteral feeding. Overall, BE altered postoperative management in 6/227 (2.6%) pts with 4/227 (1.7%) pts undergoing further testing which went on to confirm a leak. Conversely 2/227 (0.9%) pts demonstrated clinically insignificant AL, having their discharge delayed without additional intervention.
Conclusions: Barium esophagram has a poor sensitivity in the detection of anstamotic leak and has minimal impact in the postoperative management of patients undergoing esophagectomy for malignant disease. The routine use of contrast esophagograms after esophageal resection should be abandoned.