Development and validation of a prediction model for hypotension after neuraxial anesthesia in preeclamptic parturients: a multicenter retrospective study

建立和验证用于预测先兆子痫产妇椎管内麻醉后低血压的预测模型:一项多中心回顾性研究

阅读:1

Abstract

OBJECTIVE: To develop and validate a multivariate prediction model for hypotension following neuraxial anesthesia in preeclamptic parturients. METHODS: This multicenter retrospective study analyzed 1,402 preeclamptic parturients (gestational age ≥28 weeks) from three tertiary centers (2013-2024). After exclusions (n = 569), 833 patients were allocated to training (n = 495), internal validation (n = 213), and external validation (n = 125) cohorts. Multivariable logistic regression identified independent predictors, with subsequent nomogram construction. Model performance was assessed via discrimination (AUC), calibration (Hosmer-Lemeshow), and clinical utility [decision curve analysis (DCA), clinical impact curves (CIC)]. RESULTS: Seven independent predictors were identified: platelet count (OR 0.920, 95%CI 0.876-0.966), sFlt-1/PlGF ratio (OR 1.039, 95%CI 1.002-1.078), baseline perfusion index (OR 0.221, 95%CI 0.101-0.485), T6 anesthesia level (OR 11.353, 95%CI 1.408-29.320), local anesthetic dose (OR 29.391, 95%CI 4.792-38.270), fetal weight (OR 1.004, p = 0.045), and umbilical artery S/D ratio (OR 9.319, p < 0.001). The nomogram demonstrated robust discrimination (training AUC 0.851; internal validation AUC 0.836; external validation AUC 0.810) and calibration (mean absolute errors: 0.013-0.038). DCA confirmed clinical utility at a 45% risk threshold (net benefit 0.62), capturing 85% of events with 32% false positives. CONCLUSION: This validated prediction model accurately stratifies hypotension risk in preeclamptic parturients receiving neuraxial anesthesia. The nomogram facilitates targeted prophylactic interventions, optimizing resource allocation and maternal hemodynamic stability.

特别声明

1、本页面内容包含部分的内容是基于公开信息的合理引用;引用内容仅为补充信息,不代表本站立场。

2、若认为本页面引用内容涉及侵权,请及时与本站联系,我们将第一时间处理。

3、其他媒体/个人如需使用本页面原创内容,需注明“来源:[生知库]”并获得授权;使用引用内容的,需自行联系原作者获得许可。

4、投稿及合作请联系:info@biocloudy.com。