Opportunistic Computed Tomography Imaging for the Assessment of Fatty Muscle Fraction Predicts Outcome in Patients Undergoing Transcatheter Aortic Valve Replacement

Abstract

◼ sarcopenia ◼ transcatheter aortic valve replacement ◼ treatment outcome F railty is considered a major risk factor for adverse outcomes in patients undergoing transcatheter aortic valve replacement (TAVR).Previous studies proposed frailty scales to estimate procedural risk in these patients.However, assessment of these scales is time-consuming, and reported results are inconsistent. 1-3To some extent, this problem might be related to their semiquantitative nature with an inherent risk of interobserver variability.Because this may discourage the use of standardized frailty assessment in wide preinterventional diagnostic workup, international guidelines demand objective markers for estimation of frailty. 2,3Computed tomography (CT) is a substantial part of routine preinterventional workup and beyond that it allows for opportunistic body composition analysis, including assessment of muscle quality as an indicator of muscle function, which is interrelated with frailty. 3,4 This study aimed to investigate the prognostic value of fatty muscle fraction (FMF), measured from routine preinterventional CT, as an objective surrogate for frailty in patients undergoing TAVR for treatment of severe, symptomatic aortic stenosis.The institutional review board approved this retrospective study with waiver of informed consent.Consecutive patients undergoing TAVR at the University Hospital Bonn between 2010 and 2018 were evaluated with CT scans, and skeletal muscle area at the L3/L4 level was determined as previously reported. 4On the basis of densitometric thresholds, skeletal muscle area was separated in areas of fatty and lean muscle and FMF was calculated (Figure ,A).SPSS Statistics 24 (IBM, Armonk, NY, USA) and Prism 8 (GraphPad Software, La Jolla, CA, USA) were used for statistical analysis.The cohort was subdivided into tertiles with cutoff values of >51.8%, 51.8 to 37.3%, and <37.3% to define high, medium, and low FMF, respectively.One-way ANOVA followed by Tukey's multiple comparison tests were used to compare baseline variables and FMF on outcomes.Spearman correlation coefficients were used to test correlation between continuous markers of frailty and FMF.Kaplan-Meier log-rank tests were applied to compare survival curves across the 3 groups, and an adjusted Cox regression including a set of clinically relevant covariates, was fit to test the impact of clinical variables on 1-year mortality.Linearity between covariates and outcome was tested using penalized cubic spline regression analysis.Among 1491 patients undergoing TAVR, 937 had interpretable CT scans and were included in the analytic cohort.Patients with high FMF were less likely to be male (34.0%versus 46.3% versus 66.3%; P<0.001), were older (82.7±5.8 years versus 81.7±5.5 years versus 78.9±6.7 years; P<0.001), and had higher body mass index (26.8±4.7 kg/m 2 ± 26.8±4.9kg/m 2 versus 25.4±4.2kg/m 2 ; P<0.001) compared to patients with medium and low FMF.According to EuroSCORE II, preinterventional risk did not differ significantly between these groups (7.2±6.0%versus 6.6±6.7%versus 6.4±6.3%,P=0.330).

Topics & Concepts