Abstract
Colorectal cancer is a clinical condition whose treatment often involves intestinal resection. Such treatment frequently results in two major gastrointestinal complications after surgery: anastomotic leakage and prolonged ileus. Anastomotic leakage is a serious complication which, more often than not, is diagnosed late; to date, C-reactive protein is the only available diagnostic marker. A monocentric, prospective, open case-control study was performed in patients (n = 117) undergoing colorectal surgery. Intestinal fatty acid binding protein (i-FABP), citrulline, D-lactate, exhaled hydrogen, Escherichia coli genomic DNA, and ischemia modified albumin (IMA) were determined preoperatively, postoperatively, and on the following four consecutive days. Bacterial DNA was not detected in any sample, and i-FABP and D-lactate lacked any distinct potential to detect postoperative bowel complications. Exhaled breath hydrogen content showed unacceptably low sensitivity. However, citrulline turned out to be a specific marker for prolonged ileus on postoperative days 3-4. Using a cut-off value of 20 μmol/L, a sensitivity and specificity of ~75% was achieved on postoperative day 4. IMA was found to be an efficient predictor of anastomosis leak by calculating the difference between preoperative and postoperative values. This test had 100% sensitivity and 80% specificity and 100% negative and 20% positive predictive value.