P4495Psoas muscle area and volume and frailty scoring as predictors of outcomes after transcatheter aortic valve implantation

Abstract

Background: The assessment of frailty rely mostly on physical/functional performance tests or subjective questionnaires which are less feasible in very frail patients comparing to sarcopenia, defined as low muscle mass, that can be assessed objectively and relatively quickly by imaging modalities. Purpose: We aimed to determine the long-term predictive value of different frailty scores and objective assessment of sarcopenia in patients undergoing transcatheter aortic valve implantation (TAVI). Methods: Frailty indices according to VARC-2 recommendations [5-meter walk test (5MWT) and hand grip strength] as well as other available scales of frailty [Katz index, elderly mobility index (EMS), Canadian Study of Health and Aging (CSHA) scale, Identification of Seniors at Risk (ISAR) scale] were assessed at baseline. Sarcopenia was evaluated with psoas muscle area (PSA) and volume (PSV) using CT scans. The primary endpoint was 12-month all-cause mortality. Results: We enrolled 153 TAVI patients with analyzable CT scans and complete frailty data. Median of PSA normalized for body surface area (BSA) was 2581.1 (2214.9-2654.9) mm2/m2, and median of normalized PSV was 338.8 (288.1- 365.6) cc/m2. According to 5MWT 13.7% were frail, EMS scale - 5.2%, CSHA scale - 11.1%, Katz index - 12.4% patients, hand grip test - 4.6%, and ISAR scale - 28.7%. At 12 months, all-cause mortality and new-onset atrial fibrillation were highest in the lowest tertile of normalized PSA. In the ROC analysis, all the tested frailty indices, as well as PSA and PSV, were good predictors of 12- month all-cause mortality after TAVI with the highest AUC value for PSA and PSV normalized for BSA. Conclusion: Normalized PSA and PSV are stronger predictors of long-term mortality after TAVI as compared to subjective frailty scores. CT evaluation of psoas muscles could be incorporated to preprocedural comprehensive clinical models used for prediction of outcomes in patients scheduled for TAVI.