Samaha Hady, Chalhoub Najib, Tabet Margherita, Smayra Tarek, Sleilaty Ghassan
Department of Cardiovascular Surgery, Hotel-Dieu de France Hospital, Beirut, Lebanon.
Department of Radiology, Hotel-Dieu de France Hospital, Beirut, Lebanon.
Future Cardiol. 2022 Sep;18(9):687-695. doi: 10.2217/fca-2022-0016. Epub 2022 Jul 26.
The topography of vascular and valvular calcification could help accurately predict cardiovascular post-operative complications. The data on these calcifications remains scarce. Identify the topographic distribution of the ascending aortic, left heart valves and coronary calcifications. We extracted 26 variables from 557 patients, hospitalized between 2017 and 2020. The topography of calcification was evaluated by thoracic CT scans. Both multivariate logistic regression and classification and regression tree (CART) were used for statistical analysis. Several comorbidities were associated with vascular or valvular calcification. This study proposes a CART tree for patients according to their age, sex, Euroscore and lipid profile. The proposed classification could represent an important clinical tool. More studies are warranted to better prune the current CART algorithm.
血管和瓣膜钙化的形态学特征有助于准确预测心血管术后并发症。关于这些钙化的数据仍然匮乏。确定升主动脉、左心瓣膜和冠状动脉钙化的形态学分布。我们从2017年至2020年间住院的557例患者中提取了26个变量。钙化的形态学特征通过胸部CT扫描进行评估。多因素逻辑回归分析和分类回归树(CART)均用于统计分析。几种合并症与血管或瓣膜钙化有关。本研究根据患者的年龄、性别、欧洲心脏手术风险评估系统(Euroscore)和血脂谱构建了一个CART树。所提出的分类方法可能是一种重要的临床工具。需要更多的研究来进一步优化当前的CART算法。