Validation in colorectal procedures of a useful novel approach for the use of C-reactive protein in postoperative infectious complications.

Abstract

Our aim was to validate a novel use of C-reactive protein (CRP) measurement to identify postoperative infectious complications in patients undergoing colorectal surgery, and to compare the predictive value in this setting against white blood cell (WBC) count and neutrophil-to-lymphocyte ratio (NLR). This was a retrospective study of CRP, NLR and WBC measurements in patients undergoing colorectal surgery. CRP, NLR and WBC were recorded on the second postoperative day and on the day of infectious complication (patients who developed infectious complications) or within 3days prior to discharge (subjects with no complications). The test for detecting infectious complications consisted of comparing the value of the inflammatory marker on the day on which a complication was suspected against the value recorded on the second postoperative day. The test was considered positive if a given value was higher than the registered peak at postoperative day 2. Factors influencing the postoperative peak CRP were also studied. A total of 254 patients were retrospectively studied. Patients whose CRP value was higher than on the second postoperative day had a diagnostic accuracy for infectious complications of up to 94.4% and sensitivity, specificity, positive predictive value and negative predictive value of up to 97.4%, 93.4%, 85.7% and 99.1%, respectively. Poorer results were observed when WBC count and NLR were used rather than CRP measurement. Multiple linear regression analysis showed that surgical procedure and approach, as well as additional resections, were independent factors for 48h peak CRP. C-reactive protein is a better parameter than WBC count and NLR for detecting infectious complications. Our proposed methodology presents good diagnostic accuracy and performance and could potentially be used for any surgical procedure.