成人先天性心脏病再手术风险模型:STS成人心脏外科数据库分析
该研究在高容量成人先天性心脏病(ACHD)外科项目中开发了复合 morbidity 和 mortality 的风险模型,并在 STS 成人心脏外科数据库(STS-ACSD)中外部验证。结果显示,STS-ACSD 中主要结局发生率高于机构队列(16.7% vs 8.8%),机构模型在 STS-ACSD 中区分度衰减并系统性低估风险。基于 STS-ACSD 中 15 个最具影响力变量新建的模型(极端梯度提升和惩罚逻辑回归)显示出良好区分度,平均交叉验证 AUROC 分别为 0.74 和 0.73。
要点
- ACHD 再手术全国风险高,STS-ACSD 主要结局发生率为 16.7%。
- 机构风险模型在 STS-ACSD 中区分度衰减且系统性低估风险。
- 基于 STS-ACSD 变量新建的机器学习与回归模型 AUROC 达 0.73–0.74。
核心结论
- ACHD 再手术在全国范围内仍属高风险,需精细化风险分层以支持临床决策。
- 整合机器学习与回归的临床可解释风险模型是可行路径。
研究设计与样本
- 在高容量 ACHD 外科项目中用多变量逻辑回归和机器学习开发风险模型。
- 主要结局为手术死亡和主要术后并发症的复合终点,包括机械循环支持、透析、非计划非心脏再手术、神经功能缺损和心脏骤停。
主要发现
- STS-ACSD 中主要结局发生率高于机构队列(16.7% vs 8.8%)。
- 差异部分归因于 STS-ACSD 中紧急/急诊手术比例更高(26.6% vs 5.4%)。
- 机构风险模型在 STS-ACSD 中区分度衰减并系统性低估风险。
- 基于 STS-ACSD 中 15 个最具影响力变量新建模型,极端梯度提升平均交叉验证 AUROC 为 0.74,惩罚逻辑回归 AUROC 为 0.73。
关键数值与推荐
- 新建模型纳入的 15 个变量:主动脉手术、手术状态、血细胞比容、射血分数、肌酐、年龄、白细胞计数、高血压、结缔组织病、血小板、症状、再手术次数、心力衰竭、主要支付方和体重指数。
临床怎么用
- ACHD 再手术风险高,临床决策应参考细化风险分层工具。
- 机构模型外推至全国数据库时需重新校准,避免低估风险。
- 机器学习与回归整合模型可用于术前风险沟通和围术期资源规划。
以上解读依据论文摘要(全文未开放获取或许可不允许翻译转载)。
原文
BACKGROUND: Accurate risk prediction is essential for informed clinician-patient decision-making in adult congenital heart disease (ACHD). We performed extended validation of an institutional risk model for composite morbidity and mortality in reoperative ACHD surgery. METHODS: Risk models developed in a high-volume ACHD surgical program using multivariable logistic regression and machine learning were externally validated in The Society of Thoracic Surgeons Adult Cardiac Surgery Database (STS-ACSD) among patients undergoing reoperative surgery for ACHD, between July 1, 2017 and December 31, 2023. The outcome was a composite of operative mortality and major postoperative morbidity including mechanical circulatory support, dialysis, unplanned noncardiac reoperation, neurologic deficit, and cardiac arrest. Predictor importance was examined using Shapley additive explanations, and model discrimination was assessed with area under the receiver operating characteristic curve (AUROC). RESULTS: The primary outcome was more common in STS-ACSD than in the institutional cohort (16.7% vs 8.8%), attributable in part to a higher proportion of urgent/emergent procedures (26.6% vs 5.4%). Institutional risk model discrimination was attenuated in STS-ACSD with systematic underestimation of risk. A de novo model derived from the 15 most influential STS-ACSD variables (aortic procedure, status, hematocrit, ejection fraction, creatinine, age, white blood cell count, hypertension, connective tissue disorders, platelets, symptoms, reoperation number, heart failure, primary payor, and body mass index) demonstrated good discrimination with mean, cross-validated AUROC 0.74 for extreme gradient boosting and AUROC 0.73 for penalized logistic regression. CONCLUSIONS: Reoperative ACHD surgery remains high-risk nationally, underscoring the need for refined risk stratification to support care. Clinically interpretable risk models integrating machine learning and regression offer a practical path forward.
引用
- Griffeth EM, O'Sullivan D, Dearani JA, et al. Risk Model for Adult Congenital Cardiac Reoperations: Analysis of The Society of Thoracic Surgeons Adult Cardiac Surgery Database. Ann Thorac Surg. 2026. PMID: 42722265. doi:10.1016/j.athoracsur.2026.08.022
供执业医师工作参考,不替代临床判断;数值与分级以指南原文为准。