人工流产术后发生宫腔粘连风险的列线图预测模型构建与评估-现代医学

人工流产术后发生宫腔粘连风险的列线图预测模型构建与评估

单位：1. 深圳市龙岗区人民医院妇产科, 广东深圳 518172;
2. 深圳市人民医院妇产科, 广东深圳 518172

分类号：R719

出版年·卷·期（页码）：2022·50·第六期（677-682）

摘要：

目的：探讨人工流产术后宫腔粘连发生的危险因素，构建并验证术后宫腔粘连发生风险的列线图预测模型，评估其临床应用价值。方法：选取2018年7月至2020年7月在深圳市龙岗区人民医院进行无痛人工流产术的患者890例作为研究对象，根据患者术后3个月内是否出现宫腔粘连分为宫腔粘连组91例和无宫腔粘连组799例，宫腔粘连组根据粘连程度评分分为轻度粘连、中度粘连和重度粘连。收集两组患者临床资料，通过多因素Logistic回归分析筛选出人工流产术后宫腔粘连发生的独立危险因素，将独立危险因素引入R软件以建立列线图预测模型；采用受试者工作特征(ROC)曲线验证模型区分度，采用Hosmer-Lemeshow拟合优度检验及校准曲线验证模型一致性。结果：人工流产术后宫腔粘连发生率为10.22%(91/890)，其中轻度粘连发生率为5.28%(47/890)，中度粘连发生率为3.37%(30/890)，重度粘连发生率为1.57%(14/890)。与无宫腔粘连组相比，宫腔粘连组患者术后15 d子宫内膜厚度较薄，月经复潮时间较晚，盆腔炎史、子宫手术史比例较高，术后药物治疗比例较低，差异均有统计学意义(P＜0.05)。多因素Logistic回归分析发现，术后15 d子宫内膜厚度≤3.09 mm (OR=1.597，95%CI为1.228~2.077)、月经复潮时间＞35.88 d (OR=1.036，95%CI为1.014~1.059)、盆腔炎史(OR=1.497，95%CI为1.042~2.151)、子宫手术史(OR=1.873，95%CI为1.215~2.888)是人工流产术后宫腔粘连发生的独立危险因素(P＜0.05)，而术后药物治疗(OR=0.527，95%CI为0.327~0.849)是人工流产术后宫腔粘连发生的保护因素(P＜0.05)。ROC曲线下面积为0.778(95%CI为0.718~0.839)，预测模型区分度较好；Hosmer-Lemeshow拟合优度检验(χ²=7.483，P=0.486)，且校准曲线斜率接近1，预测模型一致性较好。结论：本研究构建的预测人工流产术后宫腔粘连发生风险的列线图模型，区分度和一致性均较好。

Objective: To explore the risk factors of intrauterine adhesions after artificial abortion, and to construct and validate the nomogram prediction model for the risk of postoperative intrauterine adhesions, and to evaluate its clinical application value. Methods: A total of 890 patients with painless artificial abortion in Shenzhen Longgang District People's Hospital from July 2018 to July 2020 were selected as the research objects. According to whether intrauterine adhesions occurred within 3 months after operation, they were divided into intrauterine adhesions group(91 cases) and non-intrauterine adhesions group(799 cases), and intrauterine adhesions group were divided into mild adhesion, moderate adhesion and severe adhesion according to the degree of adhesion. Clinical data of the two groups were collected, the independent risk factors of intrauterine adhesions after artificial abortion were screened by multivariate Logistic regression analysis, and were introduced into R software to establish nomogram prediction model. ROC curve was used to verify model discrimination, and Hosmer-Lemeshow goodness-of-fit test and calibration curve were used to verify model consistency. Results: The incidence of intrauterine adhesions was 10.22%(91/890), including 5.28%(47/890) of mild adhesion, 3.37%(30/890) of moderate adhesion and 1.57%(14/890) of severe adhesion respectively. Compared with non-intrauterine adhesions group, the intrauterine adhesion group had thinner endometrial thickness 15 days after operation, later menstrual rehydration time, a higher proportion of pelvic inflammatory disease history and uterine operation history, and a lower proportion of postoperative drug treatment(P<0.05). Multivariate Logistic regression analysis showed that endometrial thickness ≤ 3.09 mm(OR=1.597, 95%CI 1.228-2.077), menstrual recovery time>35.88 d(OR=1.036, 95%CI 1.014-1.059), history of pelvic inflammatory disease(OR=1.497, 95%CI 1.042-2.151), and history of uterine surgery(OR=1.873, 95%CI 1.215-2.888) were independent risk factors for intrauterine adhesions after artificial abortion(P<0.05), the postoperative drug therapy(OR=0.527, 95%CI 0.327-0.849) was the protective factor of intrauterine adhesions after artificial abortion(P<0.05). The area under ROC curve was 0.778(95%CI 0.718-0.839), and the discrimination of prediction model was good; the Hosmer-Lemeshow goodness-of-fit test showed(χ²=7.483, P=0.486), and the slope of calibration curve was close to 1, and the prediction model had good consistency. Conclusion: The nomogram model for predicting the risk of intrauterine adhesions after artificial abortion has good discrimination and consistency.

参考文献：

[1] CARLSSON I,BREDING K,LARSSON P G.Complications related to induced abortion:a combined retrospective and longitudinal follow-up study[J].BMC Womens Health,2018,18(1):158-160.
[2] 吴琼,周伟伟,宗旦棣.人工流产患者术后宫腔粘连风险的知情度调查研究[J].中国性科学,2018,27(3):137-139.
[3] 夏波,赵豆豆,王萌,等.重度宫腔粘连相关危险因素分析[J].中国妇幼健康研究,2021,32(3):355-358.
[4] WANG Y,LYU W,XU W,et al.Asherman syndrome in adenomyosis treated with uterine artery embolization:incidence predictive factors[J].Radiol Med,2020,125(5):437-443.
[5] 薛彩星,李玉萍,靳利利.宫腔镜下黏膜下肌瘤术后宫腔粘连相关因素分析[J].河北医药,2020,42(5):730-732,736.
[6] GIOAN M,FENOLLAR F,LOUNDOU A,et al.Development of a nomogram for individual preterm birth risk evaluation[J].J Gynecol Obstet Hum Reprod,2018,47(10):545-548.
[7] AAGL Advancing Minimally Invasive Gynecology Worldwide.AAGL practice report:practice guidelines for management of intrauterine synechiae[J].J Minim Invasive Gynecol,2010,17(1):1-7.
[8] 中华医学会妇产科学分会.宫腔粘连临床诊疗中国专家共识[J].中华妇产科杂志,2015,50(12):881-887.
[9] 高盼,张娟,薛娟,等.黄体酮联合戊酸雌二醇对稽留流产患者清宫术后宫腔粘连的预防作用[J].中国临床研究,2020,33(2):183-185.
[10] DETTI L,FRANCILLON L,CHRISTIANSEN M E,et al.Early pregnancy ultrasound measurements and prediction of first trimester pregnancy loss:a logistic model[J].Sci Rep,2020,10(1):1545-1547.
[11] CHEN L,LUO D,CHEN X,et al.Development of predictive risk models of postpartum stress urinary incontinence for primiparous and multiparous women[J].Urol Int,2020,104(9-10):824-832.
[12] 郭罗培,隋龙.子宫内膜损伤修复与宫腔粘连形成机制研究进展[J].中国实用妇科与产科杂志,2019,35(6):706-709.
[13] BARADWAN S,BARADWAN A,ALJAROUDI D.The association between menstrual cycle pattern and hysteroscopic march classification with endometrial thickness among infertile women with Asherman syndrome[J].Medicine (Baltimore),2018,97(27):1-4.
[14] 陈惠芳,李月,郝月,等.流产术后并发宫腔粘连的危险因素分析及宫腔镜联合雌孕激素的干预效果研究[J].现代生物医学进展,2019,19(16):3132-3136.
[15] 幸玲丽,杨丽娟,朱文君,等.宫腔镜子宫内膜息肉切除术后宫腔粘连的危险因素分析[J].中国妇幼保健,2019,34(15):3446-3448.
[16] 汤福想,刘玉玲,王晓静.优思悦与芬吗通用于人工流产术后临床效果[J].中国计划生育学杂志,2019,27(12):1604-1606.
[17] 查锦芬,宋华梅,佐满珍.新型透明质酸凝胶预防妊娠中期清宫术后宫腔粘连的研究[J].公共卫生与预防医学,2019,30(3):135-137.

服务与反馈：

【文章下载】【发表评论】【查看评论】【加入收藏】

提示：您还未登录，请登录！点此登录