新型预后评分ACEF-TBAD的开发与验证:改善B型主动脉夹层患者预后预测
这项多中心回顾性研究纳入547例接受TEVAR的B型主动脉夹层患者,在训练队列(n=382)中开发了结合年龄、肌酐、射血分数、高血压、D-二聚体/纤维蛋白原比值和IL-6的ACEF-TBAD评分,并在外部验证队列(n=165)中验证。该评分预测严重不良事件的AUC在训练集为0.922、验证集为0.829,优于原始ACEF、改良ACEF和EuroSCORE II模型。
要点
- ACEF-TBAD评分在训练集AUC 0.922、验证集AUC 0.829,优于ACEF、改良ACEF和EuroSCORE II。
- 高血压、DFR、ACEF评分和IL-6是TEVAR后严重不良事件的独立危险因素。
- 重分类分析显示风险分层显著改善,训练集NRI 0.366、验证集NRI 0.206。
核心结论
- ACEF-TBAD评分是一种新型、简单的风险分层工具,可早期识别高危患者,有助于个体化治疗并改善患者结局。
研究设计与样本
- 多中心回顾性研究,纳入2015至2020年间接受TEVAR的547例B型主动脉夹层患者,训练队列来自福建医科大学附属协和医院(n=382),验证队列来自两个外部中心(n=165)。
- 采用多变量logistic回归识别独立危险因素,构建结合年龄、肌酐、射血分数、高血压、D-二聚体/纤维蛋白原比值(DFR)和白细胞介素-6(IL-6)的复合风险评分。
- 使用ROC曲线、校准曲线、决策曲线分析和重分类指数评估模型性能。
主要发现
- ACEF-TBAD评分预测严重不良事件的AUC在训练数据集为0.922,在验证数据集为0.829,优于原始ACEF、改良ACEF和EuroSCORE II模型。
- 关键决定因素包括高血压(OR 4.84,95%CI 2.03-11.58,P<.001)、DFR(OR 1.30,95%CI 1.16-1.45,P<.001)、ACEF评分(OR 4.05,95%CI 1.76-9.32,P=.001)和IL-6(OR 1.41,95%CI 1.11-2.32,P<.001)。
- 评分显示良好的校准(P>.05)和净临床获益(决策曲线分析曲线)。
- 重分类分析显示风险分层显著改善,净重分类指数在训练集为0.366、验证集为0.206。
- 生存曲线清晰显示ACEF-TBAD评分能有效将患者分层为不同风险类别。
关键数值与推荐
- ACEF-TBAD评分组成:年龄、肌酐、射血分数、高血压、D-二聚体/纤维蛋白原比值(DFR)和白细胞介素-6(IL-6)。
- 训练数据集AUC 0.922,验证数据集AUC 0.829。
- 高血压OR 4.84(95%CI 2.03-11.58,P<.001);DFR OR 1.30(95%CI 1.16-1.45,P<.001);ACEF评分OR 4.05(95%CI 1.76-9.32,P=.001);IL-6 OR 1.41(95%CI 1.11-2.32,P<.001)。
- 净重分类指数:训练集0.366,验证集0.206。
临床怎么用
- ACEF-TBAD评分可用于接受TEVAR的B型主动脉夹层患者的风险分层,早期识别高危患者,支持个体化治疗决策。
局限性
- 原文未述局限性
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
OBJECTIVE: Type B aortic dissection (TBAD) is a life-threatening aortic disease with an increasing incidence, which requires accurate risk stratification tools for thoracic endovascular aortic repair (TEVAR). This multicenter retrospective study aimed to enhance the risk prediction of severe adverse events post TEVAR in patients with TBAD by improving the Age, Creatinine, and Ejection Fraction (ACEF) score. METHODS: This multicenter retrospective study enrolled 547 patients with TBAD who underwent TEVAR between 2015 and 2020. The training cohort (n = 382) from Fujian Medical University Union Hospital was used for model development, whereas the validation cohort (n = 165) from two external centers evaluated performance. Independent risk factors were identified using multivariate logistic regression. The novel composite risk score (ACEF-TBAD) combined age, creatinine, and ejection fraction with hypertension, D-dimer/fibrinogen ratio (DFR), and interleukin-6 (IL-6). Model performance was assessed using receiver operating characteristic curve, calibration curves, decision curve analysis, and reclassification indices. RESULTS: The ACEF-TBAD score demonstrated superior predictive accuracy compared with the original ACEF, modified ACEF, and EuroSCORE II models, with area under the curve values of 0.922 in the training dataset and 0.829 in the validation dataset. Key determinants included hypertension (odds ratio [OR], 4.84; 95% confidence interval [CI], 2.03-11.58; P < .001), DFR (OR, 1.30; 95% CI, 1.16-1.45; P < .001), ACEF score (OR, 4.05; 95% CI, 1.76-9.32; P = .001), and IL-6 (OR, 1.41; 95% CI, 1.11-2.32; P < .001). The score showed excellent calibration (P > .05) and net clinical benefit (decision curve analysis curve). Reclassification analysis revealed significant improvements in risk stratification (net reclassification index, 0.366 in training; 0.206 in validation). The survival curves clearly demonstrated that the ACEF-TBAD score effectively stratified patients into distinct risk categories, which underscores the clinical utility of the ACEF-TBAD score in predicting severe adverse events and supports its use in risk assessment for patients with TBAD. CONCLUSIONS: The ACEF-TBAD score is a novel and simple risk-stratification tool. This enables early identification of high-risk patients, facilitating personalized treatment and improving patient outcomes.
引用
- Chen X, Xie L, Wu Q, et al. Development and validation of a novel prognostic score to improve prognostic prediction in patients with type B aortic dissection. J Vasc Surg. 2025. PMID: 40541619. doi:10.1016/j.jvs.2025.06.018
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。