Introduction: Frailty is an important component of risk prognostication in transcatheter aortic valve replacement (TAVR). Objective markers of frailty, including sarcopenia, the modified Frailty Index (mFI), and albumin levels, have emerged, but little is known how such markers compare to each other in predicting outcomes after TAVR. We sought to define and compare these markers in predicting long-term outcomes after TAVR.
Methods: Patients who underwent TAVR at our institution from 2011 to 2016 were included. Indexed cross-sectional areas of the lumbosacral muscles on preoperative computed tomography scans were used to assess sarcopenia. Optimal cutoffs for sarcopenia were defined using a statistically validated method. mFI was calculated using an 11-point scale of clinical characteristics. The primary outcome was 2-year all-cause mortality. Adjusted survival analysis was used to analyze outcomes.
Results: A total of 381 patients were included in this study. Sarcopenia of the psoas muscles was associated with an increased risk of mortality on univariate (HR: 2.3, P = 0.01) and multivariate (HR: 2.5, P = 0.01) analysis. Sarcopenia of the paravertebral muscles was associated with increased risk of mortality only on univariate analysis (HR: 2.1, P = 0.03). Increased preoperative albumin levels were associated with decreased risk of mortality on univariate (HR: 0.3, P < 0.01) and multivariate analysis (HR: 0.3, P < 0.01). The (mFI) was not associated with mortality on univariate or multivariate analysis.
Discussion: Novel cutoffs for sarcopenia of the psoas muscles were determined and associated with decreased survival after TAVR. Sarcopenia and albumin levels may be better tools for risk prediction than mFI in TAVR.
Keywords: TAVR; frailty; outcomes.