Diagnostic accuracy of C-reactive protein and white blood cell counts in the early detection of inflammatory complications after open resection of colorectal cancer: a retrospective study of 1,187 patients.

ABSTRACT

PURPOSE: Although widely used, there is a lack of evidence concerning the diagnostic accuracy of C-reactive protein (CRP) and white blood cell counts (WBCs) in the postoperative period. The aim of this study was to evaluate the diagnostic accuracy of CRP and WBCs in predicting postoperative inflammatory complications after open resection of colorectal cancer. METHODS: In this retrospective study, clinical data and the CRP and WBCs, routinely measured until postoperative day 5 (POD 5), were available for 1,187 patients who underwent colorectal cancer surgery between 1997 and 2009. Using the receiver-operating characteristic (ROC) methodology, the diagnostic accuracy was evaluated according to the area under the curve (AUC). RESULTS: Three hundred forty-seven patients (29.2%; 95% CI, 26.7-31.9%) developed various inflammatory complications. Anastomotic leakage occurred in 8.0% (95% CI, 6.1-9.1%) of patients. The CRP level on POD 4 (AUC 0.76; 95% CI, 0.71-0.81) had the highest diagnostic accuracy for the early detection of inflammatory complications. With a cutoff of 123 mg/l, the sensitivity was 0.66 (95% CI, 0.56-0.74), and the specificity was 0.77 (95% CI, 0.71-0.82). The diagnostic accuracy of the WBC was significantly lower compared to CRP. CONCLUSION: Measurement of CRP on POD 4 is recommended to screen for inflammatory complications. CRP values above 123 mg/l on POD 4 should raise suspicion of inflammatory complications, although the discriminatory performance was insufficient to provide a single threshold that could be used to correctly predict inflammatory complications in clinical practice. WBC measurement contributes little to the early detection of inflammatory complications.