探讨胎盘前置状态(PPS)孕妇人工终止妊娠术(ATP)后，发生产后出血(PPH)的初步评分模型建立方法。

采用回顾性研究方法，选择2010年1月至2018年12月，于成都市妇女儿童中心医院接受ATP的142例PPS孕妇为研究对象。导致PPS孕妇发生PPH影响因素的单因素分析，采用χ²检验。将单因素分析结果中差异有统计学意义的因素，并结合已有研究结果与临床经验，对导致PPS孕妇发生PPH的5项可能影响因素，进行多因素非条件logistic回归分析。根据logistic回归分析结果中影响因素的OR值，对PPH影响因素进行赋分。根据此赋分标准，计算本组孕妇PPH风险总评分，并绘制该评分预测孕妇发生PPH的受试者工作特征(ROC)曲线，计算ROC曲线下面积(ROC-AUC)。根据约登指数最大原则，确定PPH风险总评分预测PPS孕妇发生PPH的最佳临界值。本研究遵循的程序符合2013年修订的《世界医学会赫尔辛基宣言》要求。

①对PPS孕妇发生PPH影响因素的单因素分析结果显示，差异有统计学意义的因素包括ATP史(ATP次数≥3次与<3次)、人体质量指数(BMI)≥25 kg/m²与<25 kg/m²、剖宫产术分娩史，以及PPS类型(边缘性、部分性、完全性PPS)与胎盘植入情况(无胎盘黏连、胎盘黏连及胎盘植入) 5项因素。②将这5项因素进行多因素非条件logistic回归分析的结果显示，剖宫产术分娩史(OR＝18.865，95%CI：4.013～88.671，P<0.001)，完全性PPS(OR＝20.148，95%CI：3.387～119.846，P<0.001)，以及胎盘黏连(OR＝44.045，95%CI：9.772～198.523，P<0.001)与胎盘植入(OR＝87.494，95%CI：11.685～655.111，P<0.001)，均为PPS孕妇发生PPH的独立危险因素，并根据上述4项因素的OR值，分别赋予1、1、2及4分。PPH风险总评分预测PPS孕妇发生PPH的ROC-AUC为0.926(95%CI：0.876～0.975，P<0.001)，PPH风险总评分预测PPS孕妇发生PPH的最佳临界值为3分，其预测PPS孕妇发生PPH的敏感度为83.3%，特异度为92.5%，准确率为88.0%。

本研究建立的PPS孕妇发生PPH的评分模型，可以有效预测此类孕妇ATP的PPH风险，利于个体化选择ATP方式。由于本研究为单中心、回顾性研究，尚需要通过多中心、大样本、前瞻性研究进一步证实该评分模型的临床价值。

To establish a preliminary scoring model for postpartum hemorrhage (PPH) in pregnant women with placenta previa status (PPS) after artificial termination of pregnancy (ATP).

A total of 142 pregnant women with PPS who underwent ATP in Chengdu Women′s and Children′s Central Hospital from January 2010 to December 2018 were selected as research subjects by retrospective method. Chi-square test was used for univariate analysis of influencing factors of PPH in pregnant women with PPS after ATP. Then, with the consideration of existing research results, clinical experience and the factors with statistically significant differences in the results of univariate analysis, 5 possible influencing factors of PPH in pregnant women with PPS were analyzed by multivariate unconditional logistic regression analysis. According to the OR value of influencing factors in the results of logistic regression analysis, scores were assigned to the influencing factors of PPH. According to this scoring standard, the total PPH risk scores of all pregnant women were calculated. The receiver operator characteristic (ROC) curve of total PPH risk score for predicting the occurence of PPH in pregnant women with PPS was drawn, and the area under the ROC curve (ROC-AUC) was calculated. According to the maximum principle of Youden index, the optical critical value of total PPH risk score for predicting the occurrence of PPH in pregnant women was determined. The procedures followed in this study were in line with the requirements of the World Medical Association Declaration of Helsinki revised in 2013.

①Univariate analysis of the influencing factors of PPH in pregnant women with PPS showed that factors with statistically significant difference included the history of ATP (≥3 times or <3 times), body mass index (BMI)≥25 kg/m² or <25 kg/m^2, cesarean section history, types of PPS (marginal, partial or complete PPS), and status of placenta accreta (without placenta adhesion, placenta adhesion or placenta accreta). ②Multivariate unconditional logistic regression analysis was performed on the above 5 possible influencing factors of PPH, and the results showed that cesarean section history (OR=18.865, 95%CI: 4.013-88.671, P<0.001), complete PPS (OR=20.148, 95%CI: 3.387-119.846, P<0.001), placenta adhesion (OR=44.045, 95%CI: 9.772-198.523, P<0.001) and placenta accreta (OR=87.494, 95%CI: 11.685-655.111, P<0.001) all were independent risk factors for PPH in pregnant women with PPS, and they were signed to 1 score, 1 score, 2 scores ad 4 scores, respectively according to the OR value of these independent risk factors of PPH. The ROC-AUC of total PPH risk score for predicting the occurrence of PPH in pregnant women with PPS was 0.926 (95%CI: 0.876-0.975, P<0.001). According to the maximum principle of Youden index, the optimal critical value for total PPH risk score for predicting PPH in pregnant women with PPS was 3 scores. At the same time, the sensitivity of total PPH risk score for predicting PPH in pregnant women with PPS was 83.3%, the specificity was 92.5%, and the accuracy was 88.0%.

The predictive scoring model of PPH in pregnant women with PPS established in this study can effectively predict the risk of PPH after ATP, which is beneficial to individual choice of ATP. Because this study just is a single-center and retrospective research, multi-center, large-sample and prospective studies are needed to further confirm the clinical value of this scoring model.