Kong Xiangshu, Liu Kuiran
Department of Obstetrics and Gynecology, Shengjing Hospital of China Medical University, Shenyang, People's Republic of China.
Infect Drug Resist. 2023 Mar 17;16:1521-1536. doi: 10.2147/IDR.S399666. eCollection 2023.
To study the early predictive value of WBC, CRP and PCT on infectious complications after epithelial ovarian cancer surgery, draw ROC curves, and construct a nomogram prediction model.
The clinical data of patients with epithelial ovarian cancer in Shengjing Hospital from August 2019 to August 2022 were included. The levels of WBC, CRP and PCT were statistically analyzed on the first, third and fifth days after surgery, and the ROC was plotted. Multivariate logistic regression analysis determined independent influencing factors, individualized nomogram model for predicting the occurrence of postoperative infectious complications was constructed, and the correction curve was used for verification.
A total of 116 patients were enrolled. The postoperative test levels of WBC, CRP and PCT were compared between two groups, and the differences on POD3 and POD5 were statistically significant. The ROC area on POD5 was 0.739, 0.838 and 0.804, respectively, better than that on POD3. Among them, CRP has the greatest value; The predicted value of the combined test of WBC, CRP and PCT on POD5 was greater than that of a single index on POD5. The nomogram model on POD5 was constructed, and the ROC analysis showed that it had a good degree of differentiation.
WBC, CRP and PCT can effectively predict the occurrence of postoperative infectious complications, among which CRP alone has the greatest diagnostic value on POD5, and the combined test value of the three indicators is higher than that of a single index. The nomogram model constructed by the combined indicators on POD5 can assess the risk individually.
探讨白细胞(WBC)、C反应蛋白(CRP)及降钙素原(PCT)对上皮性卵巢癌术后感染性并发症的早期预测价值,绘制ROC曲线,并构建列线图预测模型。
纳入2019年8月至2022年8月在盛京医院接受治疗的上皮性卵巢癌患者的临床资料。分别于术后第1、3、5天对WBC、CRP及PCT水平进行统计分析,并绘制ROC曲线。采用多因素logistic回归分析确定独立影响因素,构建预测术后感染性并发症发生的个体化列线图模型,并采用校正曲线进行验证。
共纳入116例患者。比较两组患者术后WBC、CRP及PCT的检测水平,术后第3天和第5天差异有统计学意义。术后第5天的ROC曲线下面积分别为0.739、0.838和0.804,优于术后第3天。其中,CRP的价值最大;术后第5天WBC、CRP及PCT联合检测的预测价值大于单项指标。构建了术后第5天的列线图模型,ROC分析显示其具有良好的区分度。
WBC、CRP及PCT可有效预测术后感染性并发症的发生,其中单项指标中CRP在术后第5天的诊断价值最大,三项指标联合检测的价值高于单项指标。术后第5天联合指标构建的列线图模型可对风险进行个体化评估。