超声联合临床高危因素建立胎盘植入性疾病风险模型的研究

Establishment of a risk model of placental accreta spectrum by ultrasound combined with clinical high risk factors

目的基于胎盘植入患者临床高危因素及超声征象建立一种胎盘植入性疾病(PAS)的风险模型,以识别重型胎盘植入。方法回顾性分析2018年1月至2022年6月首都医科大学附属北京妇产医院收治并于分娩过程中临床分型诊断或病理诊断的PAS患者121例。根据植入类型分为轻型组和重型组,比较两组之间的临床高危因素、超声征象的差异。基于临床高危因素及超声征象建立一种PAS的风险模型,预测围产期并发症。结果经临床诊断和(或)胎盘病理诊断的PAS 130例,剔除临床资料不全或超声图像显示不规范9例,余121例纳入研究。121例患者中粘连型64例,植入型39例,穿透型18例。粘连型为轻型,植入型和穿透型统称为重型,其中轻型组64例,重型组57例。两组间前置胎盘、宫腔操作史次数比较差异无统计学意义(均P>0.05);两组间剖宫产次数、子宫肌层变薄、胎盘陷窝、子宫-膀胱交界面异常血管化、子宫膀胱交界面桥接血管、胎盘隆起以及宫颈受累比较差异有统计学意义(均P<0.05)。二元Logistic回归分析结果显示胎盘陷窝、子宫-膀胱交界面异常血管化及剖宫产次数是重型PAS的独立危险因素,基于此建立风险模型,并分别绘制各独立危险因素及风险模型ROC曲线,风险模型AUC为0.826,较其他独立危险因素有良好的诊断效能。结论在PAS高危患者的产前超声分型诊断中,胎盘陷窝、子宫-膀胱交界面异常血管化及剖宫产次数三者联合建立PAS风险模型,对于重型胎盘植入具有良好的诊断效能。

Objective To establish a risk model of placenta accreta spectrum(PAS)based on the clinical risk factors and ultrasound signs of patients with placenta accreta,and identify severe placenta accreta prenatal.Methods A retrospective analysis was performed on 121 PAS patients admitted to Beijing Obstetrics and Gynecology Hospital Affiliated to Capital Medical University from January 2018 to June 2022 who were clinically classified or pathologically diagnosed during delivery.The two groups were divided into light and severe groups according to the implantation type.The clinical risk factors and ultrasound signs between the two groups were compared.A risk model of PAS was established based on the clinical risk factors and ultrasound signs to predict the perinatal complications.Results A total of 130 cases of PAS were clinically diagnosed or pathologically diagnosed with placenta,9 cases with incomplete clinical data or irregular ultrasound images were excluded,and the remaining 121 cases were included in the study.Among the 121 patients,64 cases were placental accreta,39 cases were placental increta,and 18 cases were placenta percreta.The placental accreta was defined as mild group,and the combination of placental increta and placenta percreta were referred to as severe group.There were no significant differences in placenta previa,and the number of uterine cavity operations(all P>0.05).There were significant differences in the number of cesarean section,myometrium thinning,placental lacunae,abnormal vascularization at the utero-bladder junction,bridging vessels at the utero-bladder junction,placental protuberance and cervical involvement(all P<0.05).Binary logistic regression analysis showed that placental lacunae,abnormal vasculization of the utero-bladder interface and the number of cesarean sections were independent risk factors for severe PAS.Based on this,a risk model was established and the ROC curve of each independent risk factor and risk model was plotted respectively.The AUC of the risk model was 0.826,which had better diagnostic efficacy than other independent risk factors.Conclusions In the prenatal ultrasound classification diagnosis of high-risk patients with PAS,the placental lacunae,abnormal vascularization of utero-bladder interface and the number of cesarean section are combined to establish the risk model of PAS,which has a good diagnostic efficacy for severe placenta accreta.