Nomogram模型预测剖宫产产后出血患者并发感染风险的研究

The study on the Nomogram model for predicting the risk of concurrent infection in patients with postpartum hemorrhage after cesarean section

目的 构建预测剖宫产产后出血(postpartum hemorrhage, PPH)患者并发感染风险Nomogram模型。方法 选取2021年9月至2023年9月上海交通大学医学院附属第六人民医院收治的222例行剖宫产PPH患者,根据是否感染分为剖宫产PPH感染患者(感染组,111例)和剖宫产PPH未感染患者(未感染组,111例)。统计两组临床资料、不同时间点血常规指标[白细胞、C反应蛋白(C-reactive protein, CRP)、中性粒细胞],分析血常规指标与患者并发感染的相关性;Lasso-Logistic回归分析患者并发感染的影响因素,构建预测患者并发感染风险的Nomogram模型。风险能力指数(C-index)、ROC曲线、校正曲线对该模型进行检验与评价。结果 感染组年龄≥35岁、前置胎盘、胎膜早破患者占比高于未感染组(P<0.05);感染组分娩后1 d、分娩后3 d白细胞、CRP、中性粒细胞高于分娩前,且高于未感染组(P<0.05);分娩后1 d、分娩后3 d白细胞、CRP、中性粒细胞与患者并发感染呈正相关(P<0.05),其中分娩后1 d白细胞、中性粒细胞及分娩后3 d的CRP与患者并发感染的相关性更强。Lasso-Logistic回归结果显示,前置胎盘、胎膜早破、分娩后1 d白细胞、分娩后1 d中性粒细胞、分娩后3 d的CRP与剖宫产PPH患者并发感染显著相关(P<0.05);基于以上因素构建预测患者并发感染风险的Nomogram模型,C-index显示该模型具有较好区分度,ROC曲线显示灵敏度为91.89%,特异度为81.98%,校准曲线与理想曲线拟合良好。结论 剖宫产PPH患者并发感染的影响因素为前置胎盘、胎膜早破、分娩后1 d白细胞、分娩后1 d中性粒细胞、分娩后3 d的CRP,基于以上因素构建的并发感染风险预测Nomogram模型具有良好区分度和预测价值。

Objective To establish a Nomogram model for predicting the risk of co-infection in patients with postpartum hemorrhage(PPH) during cesarean section.Methods 222 cases of cesarean PPH patients admitted to the Shanghai Jiao Tong University School of Medicine Affiliated Sixth People's Hospital from September 2021 to September 2023 were selected and divided into cesarean PPH-infected patients(infected group, 111 patients) and cesarean PPH-uninfected patients(uninfected group, 111 patients) according to whether they were infected or not. Clinical data and blood routine indexes [white blood cells, C-reactive protein(CRP), neutrophils]at different time points of the two groups were analyzed, and the correlation between blood routine indexes and patients with concurrent infection was analyzed. Lasso-Logistic regression analysis was conducted to analyze the influencing factors of co-infection in patients, and a Nomogram model for predicting the risk of co-infection was constructed. C-index, ROC curve and calibration curve were used to test and evaluate the model.Results The proportion of patients aged ≥35 years old, placenta previa and premature rupture of membranes in infected group were higher than those in uninfected group(P<0.05);white blood cell, CRP, and neutrophils at 1 day and 3 days after delivery in the infected group were higher than before delivery and higher than those in the uninfected(P<0.05);white blood cell, CRP,neutrophils on day 1 and day 3 after delivery were positively correlated with concurrent infection(P<0.05), and white blood cells at 1 day after delivery, neutrophils and CRP at 3 days after delivery were more strongly correlated with concurrent infection. Lasso-Logistic regression showed that placenta previa, premature rupture of membranes, white blood cells 1 day after delivery, neutrophils 1 day after delivery and CRP 3 days after delivery were significantly correlated with infection in PPH patients after cesarean section(P<0.05);based on the above factors, a Nomogram model was constructed to predict the risk of co-infection, C-index prediction showed that the model had good differentiation, with a sensitivity of 91.89% and a specificity of 81.98%,the calibration curve fitted well with the ideal curve.Conclusion The influencing factors of concurrent infections in PPH patients delivered by cesarean section are placenta previa, premature rupture of membranes, leukocytes at 1 day after delivery, neutrophils at 1 day after delivery, and CRP at 3 days after delivery, and the Nomogram model for predicting the risk of concurrent infections construct on the basis of the above factors has a good discriminatory degree and predictive value.