A Postoperative Systemic Inflammation Score Predicts Short- and Long-Term Outcomes in Patients Undergoing Surgery for Colorectal Cancer.

BACKGROUND: Following surgery, a significant proportion of patients develop postoperative complications that are associated with poorer long-term survival. Stereotypical markers of the systemic inflammatory response (SIR) have been shown to identify patients at increased risk of developing such complications. The aim of the present study was to examine the prognostic value of a postoperative systemic inflammation-based score in patients undergoing potentially curative surgery for colorectal cancer.
METHODS: Patients with histologically proven colorectal cancer undergoing resection between 1999 and 2013 (n = 813) were grouped into two cohorts-a retrospective test cohort (n = 402) and a prospective validation cohort (n = 411). Patients were assessed for postoperative complications and had routine blood samples taken daily. The relationship between markers of the postoperative SIR and survival was examined using Cox regression analysis.
RESULTS: In the test cohort, 87 patients developed an infective complication, while in the validation cohort, 106 patients developed an infective complication. In both cohorts, the postoperative SIR (C-reactive protein and albumin thresholds of >150 mg/L and <25 g/L, respectively) were associated with the development of infective complications (all p < 0.01). Using these thresholds, a scoring system [postoperative Glasgow prognostic score (poGPS)] was created, and on days 3 and 4 was associated with an incremental increase in the infective complication rate (all p < 0.001) and complication severity (p < 0.001). In the overall cohort, there were 175 cancer and 139 non-cancer deaths. The poGPS was also significantly associated with overall survival (p < 0.05).
CONCLUSIONS: The postoperative SIR, evidenced by the poGPS, was associated with increased complication rates and severity and a reduction in survival.