国际医药卫生导报 ›› 2024, Vol. 30 ›› Issue (21): 3618-3623.DOI: 10.3760/cma.j.issn.1007-1245.2024.21.019

• 论著 • 上一篇    下一篇

子痫前期产妇阴道试产失败的影响因素及其风险预测列线图模型构建

赵向阳  田轶  魏亚楠   

  1. 汉中市人民医院妇产科,汉中 723000

  • 收稿日期:2024-06-24 出版日期:2024-11-01 发布日期:2024-11-12
  • 通讯作者: 田轶,Email:Ty13892684938@163.com
  • 基金资助:

    陕西省重点研发计划(2022SF-277)

Influencing factors of vaginal delivery failure in women with pre-eclampsia and establishment of a risk prediction nomogram model

Zhao Xiangyang, Tian Yi, Wei Yanan   

  1. Department of Gynecology and Obstetrics, Hanzhong People's Hospital, Hanzhong 723000, China

  • Received:2024-06-24 Online:2024-11-01 Published:2024-11-12
  • Contact: Tian Yi, Email: Ty13892684938@163.com
  • Supported by:

    Key Pan of Research and Development in Shaanxi (2022SF-277)

摘要:

目的 筛查子痫前期阴道试产失败的影响因素,构建子痫前期阴道试产失败的预测模型。方法 回顾性分析2020年1月至2023年10月在汉中市人民医院分娩的137例子痫前期产妇病历资料,年龄(32.29±3.16)岁,依据80/20定律随机分为训练集(107例)和验证集(30例)。训练集107例患者根据分娩方式分为阴道分娩组(85例)和中转剖宫产组(22例)。筛查子痫前期阴道试产失败的影响因素,构建并验证子痫前期阴道试产失败的预测模型。统计学方法采用t检验、χ2检验。结果 中转剖宫产组年龄、乳酸脱氢酶、24 h尿蛋白、T淋巴辅助细胞-22(Th-22)及初产妇占比均高于阴道分娩组[(33.35±2.71)岁比(31.29±3.05)岁、(302.91±50.24)U/L比(259.78±32.61)U/L、(2.41±0.75)g比(1.79±0.34)g、(1.27±0.34)%比(1.06±0.31)%、95.45%(21/22)比76.47%(65/85)],浆细胞样树突细胞(pDC)低于阴道分娩组[(0.13±0.04)%比(0.18±0.06)%],差异均有统计学意义(均P<0.05)。logistic回归分析结果显示,年龄(OR=3.601,95%CI 1.063~12.194)、产次(OR=3.725,95%CI 1.099~12.615)、乳酸脱氢酶(OR=4.088,95%CI 1.207~13.846)、24 h蛋白尿(OR=3.740,95%CI 1.104~12.667)、Th-22(OR=4.088,95%CI 1.207~13.846)、pDC(OR=0.223,95%CI 0.066~0.754)是子痫前期阴道试产失败的影响因素(均P<0.05)。训练集107例患者列线图模型预测子痫前期阴道试产失败的灵敏度为0.804(95%CI 0.715~0.862),特异度为0.749(95%CI 0.675~0.816),曲线下面积为0.832(95%CI 0.731~0.899)。另将验证集30例子痫前期患者带入验证,列线图模型预测子痫前期阴道试产失败的灵敏度为0.756(95%CI 0.682~0.825),特异度为0.773(95%CI 0.701~0.851),曲线下面积为0.804(95%CI 0.707~0.881)。结论 年龄、产次、乳酸脱氢酶、24 h尿蛋白、Th-22、pDC与子痫前期阴道试产失败有关,构建列线图预测模型有助于早期评估子痫前期阴道试产失败的风险。

关键词:

子痫前期, 阴道试产, 剖宫产, 预测模型

Abstract:

Objective To screen the influencing factors of vaginal delivery failure in women with pre-eclampsia, and to construct a predictive model for vaginal delivery failure. Methods The medical records of 137 women with pre-eclampsia who were (32.29±3.16) years old and gave birth in Hanzhong People's Hospital from January 2020 to October 2023 were retrospectively analyzed. According to the 80/20 rule, they were randomly divided into a training set (107 cases) and a validation set (30 cases). According to the delivery modes, the training set were divided into a vaginal delivery group (85 cases) and an intermediate cesarean section group (22 cases). The influencing factors of vaginal trial failure in the women with pre-eclampsia were screened, and a predictive model for vaginal trial failure in the women with pre-eclampsia was constructed and validated. t and χ2 tests were used for the statistical analysis. Results The age, levels of lactate dehydrogenase, 24-hour proteinuria, and Th-22, and proportion of primipara in the intermediate cesarean section group were higher than those in the vaginal delivery group [(33.35±2.71) years vs. (31.29±3.05) years, (302.91±50.24) U/L vs. (259.78±32.61) U/L, (2.41±0.75) g vs. (1.79±0.34)g, (1.27±0.34)% vs. (1.06±0.31)%, and 95.45% (21/22) vs. 76.47% (65/85)]; the plasmacytoid dendritic cell (pDC) in the intermediate cesarean section group was lower than that in the vaginal delivery group [(0.13±0.04)% vs. (0.18±0.06)%]; there were statistical differences (all P<0.05). The logistic regression analysis showed that age (OR=3.601, 95%CI 1.063-12.194), parity (OR=3.725, 95%CI 1.099-12.615), lactate dehydrogenase (OR=4.088, 95%CI 1.207-13.846), 24-hour proteinuria (OR=3.740, 95%CI 1.104-12.667), Th-22 (OR=4.088, 95%CI 1.207-13.846), and pDC (OR=0.223, 95%CI 0.066-0.754) were identified as the influencing factors for vaginal delivery failure in the women with pre-eclampsia (all P<0.05). The sensitivity of the nomogram model in predicting vaginal delivery failure in the 107 women with pre-eclampsia of the training set was 0.804 (95%CI 0.715-0.862), with a specificity of 0.749 (95%CI 0.675-0.816) and an area under the curve of 0.832 (95%CI 0.731-0.899). When validated in the 30 women in the validation set, the nomogram model exhibited a sensitivity of 0.756 (95%CI 0.682-0.825), a specificity of 0.773 (95%CI 0.701-0.851), and an area under the curve of 0.804 (95%CI 0.707-0.881) for predicting vaginal delivery failure in the women with pre-eclampsia. Conclusions Age, parity, lactate dehydrogenase, 24-hour proteinuria, Th-22, and pDC are associated with vaginal delivery failure in women with pre-eclampsia. Constructing a nomogram prediction model contributes to the early assessment of the risk of vaginal delivery failure in women with pre-eclampsia.

Key words:

Pre-eclampsia, Vaginal trial delivery, Cesarean section, Prediction model