生殖医学杂志

目的 分析先兆流产合并绒毛膜下血肿(SCH)患者发生流产和早产的危险因素，并基于危险因素建立列线图模型。方法 回顾性分析2021年6月至2023年6月银川国龙骨科医院收治的147例先兆流产合并SCH孕妇临床资料，根据妊娠结局分为足月分娩组(n=101)、流产组(n=32)和早产组(n=14)。采用多因素Logistic回归分析先兆流产合并SCH患者流产和早产的危险因素，建立列线图预测模型，并对模型预测性能进行验证。结果 多因素分析结果显示，SCH诊断孕周[OR=1.216,95%CI(1.044,1.416),P=0.012]、阴道流血持续时间[OR=0.869,95%CI(0.799,0.945),P=0.001]、血肿持续时间[OR=0.882,95%CI(0.790,0.984),P=0.024]、血肿体积[OR=0.964,95%CI(0.934,0.995),P=0.025]是流产的预测因素，下腹痛[OR=0.140,95%CI(0.028,0.693),P=0.016]、血肿持续时间[OR=0.741,95%CI(0.616,0.892),P=0.001]、血肿体积[OR=0.943,95%CI(0.901,0.988),P=0.013]是早产的危险因素。列线图模型预测流产和早产的曲线下面积分别为0.871和0.896,特异度分别为64.36%和80.20%,敏感度分别为100.00%和85.71%,约登指数分别为0.644和0.659。Hosmer-Lemeshow检验分析结果显示模型预测值和实际值之间无显著性差异(P>0.05)。结论 基于危险因素建立的列线图模型对先兆流产合并SCH患者流产和早产具有一定预测价值，方便医护人员采取医疗干预措施。

Objective:To explore the risk factors of abortion and premature delivery in threatened abortion patients with subchorionic hematoma(SCH),and establish a nomogram model based on the risk factors.Methods:Using a retrospective analysis method, clinical data of 147 threatened abortion patients with SCH from Yinchuang Guolong Orthopedic Hospital between June 2021 and June 2023 were collected. The patients were assigned to term delivery group(n=101),abortion group(n=32)and premature delivery group(n=14). The risk factors of abortion and premature delivery were analyzed by multivariate logistic regression, and a nomogram model was established, the predictive performance of the model was verified.Results:The results of multivariate logistic regression showed that the gestational age of SCH diagnosis[OR=1.216,95%CI(1.044,1.416),P=0.012],the duration of vaginal bleeding[OR=0.869,95%CI(0.799,0.945),P=0.001],the duration of hematoma existence[OR=0.882,95%CI(0.790,0.984),P=0.024],hematoma volume[OR=0.964,95%CI(0.934,0.995),P=0.025] were factors for predicting abortion. Hypogastralgia[OR=0.140,95%CI(0.028,0.693),P=0.016],the duration of hematoma existence[OR=0.741,95%CI(0.616,0.892),P=0.001],hematoma volume[OR=0.943,95%CI(0.901,0.988),P=0.013] were factors for predicting premature delivery. The areas under the curve of nomogram for predicting abortion and premature delivery were 0.871 and 0.896,the specificity was 64.36% and 80.20%,the sensitivity was 100.00% and 85.71%,and the Youden index was 0.644 and 0.659,respectively. The Hosmer-Lemeshow goodness-of-fit test showed that there was no statistically significant difference between the predicted and actual values of the model(P>0.05).Conclusions:A risk factors-based nomogram model has certain predictive value for abortion and premature delivery of threatened abortion patients with SCH,which can facilitate medical staff to take active medical interventions.