Abstract: To evaluate the efficiency of ultrasound scoring system combined DOI：10.3760/cma.j.cn441417-20240729-01002 收稿日期 2024-07-29 本文编辑 吴相思 引用本文：李西娜, 窦盼, 李令民. 超声评分系统联合MRI对胎盘植入性疾病患者不良妊娠结局的预测模 型构建[J]. 国际医药卫生导报, 2025, 31(1): 5-11. DOI: 10.3760/cma.j.cn441417-20240729-01002. · 5 · 国际医药卫生导报 2025 年1 月第 31 卷第 1 期　IMHGN, January 2025, Vol. 31, No. 1 with magnetic resonance imaging (MRI) in predicting adverse pregnancy outcomes in patients with placenta accreta spectrum disorders. Methods This study was a cross-sectional study. A total of 180 patients with placenta accreta spectrum disorders who underwent cesarean section in Xi'an People's Hospital (Xi'an Fourth Hospital) from January 2021 to January 2024 were selected as the study objects. The patients were assessed using the placenta accreta ultrasound scoring system established by Peking University Third Hospital, and they received MRI before delivery, including 5 MRI signs. The clinical data of all patients were collected, and the adverse pregnancy outcomes were defined as intraoperative uterine rupture, intraoperative bleeding ≥1 500 ml, hysterectomy, or severe postpartum infection. The patients were divided into a normal outcome group (99 cases) and an adverse outcome group (81 cases). Logistic regression analysis was used to analyze the influencing factors of adverse pregnancy outcomes in patients with placenta accreta spectrum disorders, and a nomogram model was constructed. The receiver operating characteristic curve (ROC), calibration curve, and decision curve were used to evaluate the prediction model. Independent sample t test, rank sum test, and χ2 test were used for statistical analysis. Results The normal outcome group was (33.57±3.82) years old, and the adverse outcome group was (34.56± 3.68) years old, without statistically significant difference (P>0.05). The numbers of pregnancies and births in the adverse outcome group were 4 (3, 5) and 1 (1, 2), respectively, which were higher than 3 (2, 4) and 1 (1, 1) in the normal outcome group; the gestational age at delivery in the adverse outcome group was (35.12±2.54) weeks, which was shorter than that in the normal outcome group [(36.68±2.31) weeks]; the ultrasound score in the adverse outcome group was (11.05±1.15) points, which was higher than that in the normal outcome group [(7.00±1.00) points]; the case numbers of placental intravascular growth signs (lakes), T2 dark bands within the placenta, uterine bulging, cervical invasion, and vascular proliferation at the uterine-bladder interface in the adverse outcome group were 64, 67, 62, 18, and 63, respectively, which were more than 37, 36, 32, 4, and 35 in the normal outcome group, with statistically significant differences (all P<0.05). Univariate and multivariate logistic regression analysis showed that gestational week at delivery was an independent protective factor for adverse pregnancy outcomes in patients with placenta accreta spectrum disorders (OR<1, P<0.05); the numbers of pregnancies and births, ultrasound score, placental intravascular growth signs (lakes), T2 dark bands within the placenta, uterine bulging, cervical invasion, and vascular proliferation at the uterine-bladder interface were independent risk factors for adverse pregnancy outcomes (all OR>1, P<0.05). According to the multivariate logistic regression analysis results, a nomogram model was constructed, and ROC analysis was performed on the nomogram model. The sensitivity of the nomogram model was 97.50%, the specificity was 94.90%, the Youden index was 0.924, and the area under the curve (AUC) was 0.980 (0.961-0.999). Conclusions Ultrasound scoring system combined with MRI can effectively predict adverse pregnancy outcomes in patients with placenta accreta spectrum disorders. By using appropriate scoring and imaging standards, physicians can more accurately identify high-risk patients, thereby taking intervention measures to improve pregnancy outcomes. The high sensitivity and specificity of the nomogram model further confirm the clinical application value of this prediction method, providing an important tool for the management of high-risk pregnancies.

Key words: Magnetic resonance imaging, Ultrasound scoring system, Placenta accreta spectrum disorders, Adverse pregnancy outcomes, Prediction model

摘要： 目的　评估超声评分系统联合磁共振成像（MRI）预测胎盘植入性疾病患者不良妊娠结 局的效能。方法　本研究为横断面研究。选取2021年1月至2024年1月于西安市人民医院（西安市 第四医院）接受剖宫产手术的胎盘植入性疾病患者共 180例为研究对象。根据北京大学第三医院建 立的胎盘植入超声评分系统对患者进行产前超声评分，且所有患者产前均接受 MRI并纳入 5项 MRI 征象。收集所有患者临床资料，以术中子宫破裂、术中出血量≥1 500 ml、子宫切除或产后严重感染为 不良妊娠结局，将患者分为正常结局组（99例）及不良结局组（81例）。采用logistic回归分析胎盘植入 性疾病患者不良妊娠结局的影响因素并构建列线图模型。采用受试者操作特征曲线（ROC）、校准曲 线及决策曲线评价预测模型。采用独立样本t检验、秩和检验、χ2 检验进行统计学分析。结果　正常 结局组年龄（33.57±3.82）岁，不良结局组年龄（34.56±3.68）岁，差异无统计学意义（P>0.05）。不良结局 组孕次、产次分别为 4（3，5）、1（1，2），多于正常结局组的 3（2，4）、1（1，1）；不良结局组分娩孕周为 （35.12±2.54）周，短于正常结局组的（36.68±2.31）周；不良结局组超声评分为（11.05±1.15）分，高于正 常结局组的（7.00±1.00）分；不良结局组胎盘内增生血管征象（血窦）、T2相胎盘内黑带、子宫外突、子 宫颈受侵、子宫膀胱交界面增生血管例数分别为64、67、62、18、63，均多于正常结局组的37、36、32、4、 35，差异均有统计学意义（均P<0.05）。单因素、多因素logistic回归分析显示，分娩孕周为胎盘植入性 疾病患者不良妊娠结局的独立保护因素（OR<1，P<0.05），孕次、产次、超声评分、胎盘内增生血管征象 （血窦）、T2相胎盘内黑带、子宫外突、子宫颈受侵、子宫膀胱交界面增生血管为不良妊娠结局的独立 危险因素（均OR>1，P<0.05）。根据多因素logistic回归分析结果构建列线图模型，对列线图模型进行 ROC分析，灵敏度为97.50%，特异度为94.90%，约登指数为0.924，曲线下面积（AUC）为0.980（0.961～ 0.999）。结论　超声评分系统联合MRI可以有效预测胎盘植入性疾病患者的不良妊娠结局。通过适 当的评分和成像标准，能够更准确地识别高风险患者，从而采取干预措施以改善妊娠结局。列线图模 型的高灵敏度和特异度进一步证实了这种预测方法的临床应用价值，为高风险妊娠管理提供了重要 工具。

关键词: 磁共振成像, 超声评分系统, 胎盘植入性疾病, 不良妊娠结局, 预测模型