Background:Sarcopenia is a serious systemic disease reducing overall survival. TAVI is selectively performed in patients with severe aortic stenosis who are not indicated for open cardiac surgery due to severe polymorbidity. Artificial intelligence-assisted assessment of body composition from available CT scans appears to be a simple tool to stratify these patients into low and high risk of all-cause mortality.
Methods:The segmentation of preprocedural CT at the L3 level in patients undergoing TAVI was performed using a neural network (AutoMATiCA) and the obtained parameters (area and density of intramuscular, visceral and subcutaneous fat and muscle) were analyzed using Cox univariate and multivariate models for continuous and categorical variables to determine the regression estimate of survival time. The study was approved by the ethics committee and registered on Clinical Trials (NCT05672160).
Results:866 patients were included (median/IQR: age 79.7/74.9-83.3years; BMI 28.9/26.0-32.6). Survival analysis was performed on all automatically obtained parameters of muscle and fat density and area. Skeletal muscle index (SMI), visceral (VAT) and subcutaneous fat density (SAT) predict overall survival in patients after TAVI: SMI HR 0.987, 95% CI(0.976-0-997); VAT 1.016(1.003-1.029) and SAT 1.015(1.005-1.024), all p
Conclusions:Automatic assessment of body composition helps to estimate the increased risk of death from any cause in patients after TAVI and can thus help in the indication process and periprocedural care of these patients.
Supported by:MUNI/A/1343/2022