C-reactive protein is a useful negative predictor of anastomotic leak in oesophago-gastric resection.

Abstract

Post-operative C-reactive protein (CRP) has been investigated as a predictor of anastomotic leak (AL) following colorectal surgery, but its role in oesophago-gastric surgery is not yet established. Clinical data and post-operative CRP values of patients who underwent elective oesophago-gastric resection between January 2004 and July 2014 were analysed retrospectively. Only patients with an oesophageal anastomosis were included. AL was defined as leakage of contrast material seen on imaging or a leak confirmed intraoperatively on return to theatre. One hundred and forty-five patients were identified. Of the 145 patients, 13 (9%) developed AL. The CRP on post-operative days (POD) 2, 3 and 6 had the greatest diagnostic accuracy, with an area under the curve of 0.82, 0.80 and 0.91 respectively. Using a cut-off value of 209 mg/L on POD 2, the sensitivity was 100%, the specificity was 61%, the positive predictive value (PPV) was 21% and the negative predictive value (NPV) was 100%. Using a cut-off value of 190 mg/L on POD 3, the sensitivity was 100%, the specificity was 59%, the PPV was 21% and the NPV was 100%. Using a cut-off value of 154 mg/L on POD 6, the sensitivity was 100%, the specificity was 78%, the PPV was 29% and the NPV was 100%. Post-operative CRP is an accurate negative predictive test for the development of AL following oesophago-gastric surgery. It may help to discriminate between patients with a high risk of leak and those in which AL is unlikely to occur.