Predictive value of preoperative neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio in surgical site infection following radical resection for rectal cancer

Objective 　To investigate the predictive value of preoperative neutrophil-to-lymphocyte ratio and platelet-to-lymphocyte ratio in surgical site infection (SSI) following radical resection for rectal cancer. Methods 　Retrospective analysis of clinical data was performed about 298 patients undergoing radical resection of rectal cancer in the Chinese PLA General Hospital from January 2015 to February 2018. According to whether SSI occurred 30 days after surgery, patients were divided into SSI group (n=20) and control group (n=278). Gender, age, preoperative neoadjuvant chemoradiation, surgical procedure, T stage, and preoperative neutrophil count, lymphocyte count, platelet count, neutrophil-lymphocyte ratio (NLR), platelet-lymphocyte ratio (PLR), albumin, and hemoglobin level were compared between the two groups. Subgroup analysis was performed in the SSI group according to laparotomy and minimally invasive surgery. Gender, age, preoperative neoadjuvant chemoradiation (CRT), T stage, and preoperative neutrophil count, lymphocyte count, platelet count, NLR, PLR, albumin, and hemoglobin level were compared between two the subgroups. The predictive value of preoperative NLR and preoperative PLR in surgical site infection following radical resection for rectal cancer were analyzed by receiver operating characteristic (ROC) curve. Results 　The ratio of male to female in the SSI group is higher than control group (19/1 vs. 178/100, P=0.005). There was no significant difference in other demographic and clinical data such as age, non-restorative operation ratio, minimally invasive surgery ratio, neoadjuvant CRT ratio, T stage, preoperative albumin and preoperative hemoglobin between the two groups (P>0.05). There was no significant difference in preoperative neutrophils count [(3.96±1.03)×109/L vs. (3.62±1.28)×109/L, P=0.245], lymphocytes count [(1.47±0.45)×109/L vs. (1.71±0.64)×109/L, P=0.103] and platelets count [(249.10±57.42)×109/L vs. (230.21±68.53)×109/L, P=0.231] between the two groups. Preoperative NLR in the SSI group was significantly higher than that in the control group (2.77±0.52 vs. 2.39±1.23, P=0.010). Preoperative PLR in the SSI group was higher than that in the control group (184.46±69.54 vs. 152.93±73.82, P=0.065), but the difference was not statistically significant. In the SSI group, the age of the laparotomy subgroup was significantly lower than that of the minimally invasive surgery subgroup (49.20±5.54 vs. 61.87±10.24, P=0.018). There was no significant difference in gender, non-restorative operation ratio, neoadjuvant CRT ratio, T stage, preoperative neutrophil count, lymphocyte count, platelet count, NLR, PLR, albumin, and hemoglobin level between two the subgroups (P>0.05). ROC curve analysis showed that AUC of preoperative NLR in prediction of SSI following radical resection for rectal cancer was 0.711(95%CI 0.643-0.779). When the optimal cut off point was 2.13, its sensitivity and specificity was 95.0% and 51.4%, respectively. AUC of preoperative PLR in prediction of SSI following radical resection for rectal cancer was 0.665(95% CI 0.553-0.777). When the optimal cut off point was 150.69, its sensitivity and specificity was 75.0% and 59.7%. Conclusion 　Preoperative NLR and PLR have predictive value for SSI following radical resection for rectal cancer. DOI: 10.11855/j.issn.0577-7402.2019.03.10