Suppr超能文献

经心尖 TAVI:生存、血液动力学、器械和机器学习。10 年经验教训。

Transapical TAVI: Survival, Hemodynamics, Devices and Machine Learning. Lessons Learned After 10-Year Experience.

机构信息

Division of Cardiac Surgery, Department of Cardio-Thoracic-Vascular Sciences and Public Health, University of Padova, Padova, Italy.

Division of Cardiac Surgery, Department of Cardio-Thoracic-Vascular Sciences and Public Health, University of Padova, Padova, Italy.

出版信息

Curr Probl Cardiol. 2023 Aug;48(8):101734. doi: 10.1016/j.cpcardiol.2023.101734. Epub 2023 Apr 10.

Abstract

Aim of this single-center, retrospective study was to assess early and long-term clinical and hemodynamic results of transapical aortic valve implantation (TA-TAVI), and to identify predictors of survival at follow-up. All patients undergoing TA-TAVI for severe aortic valve stenosis at our institution were reviewed. A hybrid approach based on machine-learning techniques was employed to identify survival predictors, using a bagging-decision-tree algorithm and a Random-Forest algorithm, respectively. Two-hundred-thirty-four consecutive patients underwent TA-TAVI (March 2009-May 2019). All cause 30-day mortality was 5.1%. Device success was 95.7%. Median follow-up time was 35.2 months. Kaplan-Meier overall survival rates at 2, 5, and 8 years were 75%, 44%, and 15%, respectively. Structural-valve-deterioration occurred in 25 patients (11.3%) overall. The strongest predictors of survival at follow-up were age, body-mass-index, and ejection fraction. TA-TAVI provided valid early and long-term outcomes. These data support its choice as an optimal alternative access whenever the transfemoral route is not feasible.

摘要

本单中心回顾性研究旨在评估经心尖主动脉瓣植入术(TA-TAVI)的早期和长期临床及血流动力学结果,并确定随访时生存的预测因素。对在我院接受重度主动脉瓣狭窄的 TA-TAVI 的所有患者进行了回顾。采用基于机器学习技术的混合方法,分别使用袋装决策树算法和随机森林算法来识别生存预测因素。234 例连续患者接受了 TA-TAVI(2009 年 3 月至 2019 年 5 月)。全因 30 天死亡率为 5.1%。器械成功率为 95.7%。中位随访时间为 35.2 个月。2、5 和 8 年的 Kaplan-Meier 总生存率分别为 75%、44%和 15%。25 例(11.3%)患者发生结构性瓣膜恶化。随访时生存的最强预测因素是年龄、体重指数和射血分数。TA-TAVI 提供了有效的早期和长期结果。这些数据支持将其作为经股动脉途径不可行时的最佳替代入路选择。

文献检索

告别复杂PubMed语法,用中文像聊天一样搜索,搜遍4000万医学文献。AI智能推荐,让科研检索更轻松。

立即免费搜索

文件翻译

保留排版,准确专业,支持PDF/Word/PPT等文件格式,支持 12+语言互译。

免费翻译文档

深度研究

AI帮你快速写综述,25分钟生成高质量综述,智能提取关键信息,辅助科研写作。

立即免费体验