Background: Anastomotic leakage (AL) after gastroesophageal resection for cancer is a serious complication. The aim was to evaluate mediastinal microdialysis in the detection of AL before clinical symptoms.
Methods: Sixty patients were included. Samples were collected every 4 hours in the 1st 8 postoperative days and analyzed for several metabolites.
Results: Forty-four patients had an uncomplicated postoperative recovery, 7 developed anastomotic-related complications, and 5 developed major nonanastomotic-related complications. Six patients were excluded (early catheter malfunction and reoperation). Logistic regression model on several metabolites demonstrated a 100% sensitivity, specificity, and positive and negative predictive values regarding the diagnosis of anastomotic complications within postoperative day 7. However, as independent markers, none of the measured metabolites were able to predict AL.
Conclusion: The diagnosis of anastomotic-related complications before clinical symptoms seemed possible by mediastinal microdialysis, but the diagnosis should be based on an interpretation of several metabolic events.
Keywords: Anastomotic leakage; Esophageal cancer; Lactate/glucose ratio; Lactate/pyruvate ratio; Mediastinal microdialysis.
Copyright © 2014 Elsevier Inc. All rights reserved.