Significance of white blood cells count, C-reactive protein and procalcitonin for the detection of infectious complications after cytoreductive surgery and hyperthermic intraperitoneal chemotherapy
10.3760/cma.j.cn115396-20230406-00087
- VernacularTitle:白细胞计数、降钙素原和C反应蛋白在预测细胞减灭术联合腹腔热灌注术后感染中的意义
- Author:
Haodong SHI
1
;
Qi ZHANG
;
Xiaoyan LIANG
;
Bo HUANG
Author Information
1. 山西医科大学第五临床医学院胃肠胰腺外科，太原 030001
- Keywords:
Cytoreduction surgical;
Hyperthermic intraperitoneal chemotherapy;
Risk factors;
White blood cell count;
C-reactive protein;
Procalcitonin
- From:
International Journal of Surgery
2023;50(9):596-604,F4
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To investigate the risk factors for infection after cytoreductive surgery combined with hyperthermic intraperitoneal chemotherapy (CRS+ HIPEC), and observe the infection in patients who underwent this combined procedure, to explore the predictive value of postoperative white blood cell count (WBC), C-reactive protein (CRP), procalcitonin (PCT) and systemic immune inflammation index (SII) for postoperative infection.Methods:Clinical data of 106 patients who underwent CRS + HIPEC in the Fifth Medical Clinical College of Shanxi Medical University between July 2019 and July 2022 were retrospectively analyzed. These patients, including 61 males and 45 females, (58.93±10.65) years old, were divided into the infection group ( n=19) and the non-infection group ( n=87) according to the presence of postoperative infection. Risk factors were analyzed for patients in the infection group, and postoperative WBC, CRP, PCT and SII were determined for patients in both groups to determine their prognostic values. Risk factors for postoperative secondary infection in patients with CRS+ HIPEC were analyzed using univariate and multivariate logistic regression. Predictive values of WBC, PCT, CRP and SII for postoperative infection were evaluated by receiver operating characteristic (ROC) curve, and the optimal cutoff values of these variables were determined by Youden index and evaluated using sensitivity, specificity, positive predictive value, and negative predictive value as evaluation indexes. Nomogram prediction model was constructed using R software and samples were included in this model to calculate the total score of these patients. ROC curve analysis and calibration curve verification were then performed. Results:Univariate analysis showed significant differences in age, body mass index (BMI), postoperative fistulization, preoperative serum albumin, combined multiple organ resection, and operation duration between the infection and non-infection groups ( P<0.05). WBC, CRP, PCT and SII were compared between the infection group and the non-infection group on were compared on postoperative days 1, 3, 5 and 7, and the ROC curves were plotted accordingly. The area under the ROC curve (AUC) of the WBC, CRP, PCT and SII on postoperative Days 1, 3, 5 and 7, and their 95% CI, sensitivities and specificities were compared. It was found that the predictive values of the 5th postoperative day WBC and PCT, and the 7th postoperative day CRP were superior to those determined on other postoperative days, and SII was not significant in predicting postoperative secondary infection in patients who underwent CRS+ HIPEC. The cut-off values of the 5th postoperative day WBC and PCT and the 7th postoperative day CRP were 7.7×10 9/L, 2.068 ng/mL and 76.43 mg/L, respectively, and AUCs and their 95% CI were 0.754 and (0.625, 0.883), 0.830 and (0.717, 0.943), 0.715 and (0.584, 0.846), respectively, with sensitivities of 78.9%, 68.4% and 63.2%, respectively, and specificities of 70.1%, 96.5% and 75.9%, respectively. The predictive values of PCT on postoperative days 1, 3, 5 and 7 were superior to those of WBC, CRP and SII determined on each corresponding day. The cut-off values of the 5th postoperative day WBC and PCT and the 7th postoperative day CRP were used as the classification thresholds, and the results after classification as well as significant variables in univariate analysis, including age, BMI≥25 kg/m 2, postoperative fistulization, preoperative serum albumin≥35 g/L, number of organs resected and operation duration were included in multivariate logistic regression analysis. The results showed that BMI≥25 kg/m 2, combined multiple organ resection, WBC and PCT on the 5th postoperative day and, CRP on the 7th postoperative day were independent risk factors for secondary infection ( P<0.05). A Nomogram prediction model was then constructed. Points indicated the scores for each variable, and the corresponding scores were 70 when BMI was ≥25 kg/m 2, 80 when multiple organ resection was combined, 100 when the 5th postoperative day WBC was ≥7.7×10 9/L, and 79 when the 7th postoperative day CRP was ≥76.3. The sum of the scores for all variable was calculated and used as total score for the patient. The total score obtained from the Nomogram prediction model was used for ROC curve analysis and calibration curve verification. The ROC curve analysis showed that the AUC was 0.966, with a sensitivity of 0.895 and specificity of 0.966, indicating an excellent discriminative power of the model. The significance level of the calibration curve was 0.05, and the absolute error between the predicted and actual incidences of postoperative infection after CRS+ HIPEC was 0.038. Conclusions:The incidence of secondary infection after CRS+ HIPEC is related to factors such as BMI and combined multiple organ resection. Inflammation markers in peripheral blood, including PCT, CRP and WBC, can serve as predictors for postoperative secondary infection in patients with CRS+ HIPEC, and the fifth postoperative day WBC and PCT and the seventh postoperative day CRP among others have the highest diagnostic values for postoperative infections. In addition, the predictive value of combined diagnosis is superior to that of individual testing.
