Background: Aqueous contrast swallow study is recommended as a screening procedure for the evaluation of esophageal anastomotic integrity following esophagectomy. The aim of this study was to assess the accuracy of water- soluble contrast swallow screening as a predictor of clinically significant anastomotic leak in patients with esophagectomy. Patients and Methods: The records of 505 consecutive patients undergoing esophagectomy in Mayo Clinic from January 1991 through December 1995 were retrospectively reviewed. 464 ( 92%) patients had water- soluble contrast swallows performed in the early postoperative period ( median postoperative day 7, range 4 - 11 days). Results: A total of 39 radiological leaks were obtained but only 17 of these had clinical signs of anastomotic leakage. Furthermore, 25 patients who had normal swallow study developed a clinical anastomotic leak. There were therefore 22 ( 4.7%) false positive and 25 ( 5.4%) false negative results giving values for the specificity, sensitivity and false negative error rate of the radiological examination of 94.7, 40.4, and 59.5% respectively. Aspiration of the contrast agent was noted on fluoroscopy in 30 ( 6.5%) patients. Only 2 ( 0.4%) patients developed aqueous contrast agent- caused aspiration pneumonia. There was no procedure- related mortality. Conclusion: While radiological assessment of esophageal anastomoses in the early postoperative period using aqueous contrast agents appears to be a relatively safe procedure, the poor sensitivity and high false negative error rate of this technique, when performed on postoperative day 7 and in a series with clinical anastomotic leak rate of 9%, is insufficient for it to be worthwhile as a screening procedure. Copyright (C) 2005 S. Karger AG, Basel.