Gait Speed and Dependence in Activities of Daily Living in Older Adults With Severe Aortic Stenosis

Abstract

In the transcatheter aortic valve replacement era, characterization of functional status in older adults with aortic stenosis (AS) is essential. Gait speed (GS) is emerging as a marker of frailty and predictor of outcomes in older adults undergoing cardiovascular intervention. The objective of this study was to delineate the prevalence of slow GS, evaluate the association of GS with factors used in standard cardiovascular assessments, and evaluate the association of GS with dependence in activities of daily living (ADLs) in older adults with AS. We hypothesized that gait speed would not be associated with clinical factors, but would be associated with ADLs. We evaluated GS, ADLs dependence, and Society of Thoracic Surgery score along with clinical and functional assessments in 102 older adults with AS being evaluated for transcatheter valve. Gait speed <0.5 m/s was considered slow, and GS ≥0.5 m/s was considered preserved. We assessed the association of covariates with GS as well as with ADLs dependence. Median GS was 0.37 m/s (interquartile range, 0.0-0.65 m/s). Sixty-four (63%) subjects had slow GS. Of commonly employed clinical covariates, only prior coronary intervention and serum albumin were weakly associated with GS. However, GS was independently associated with ADLs dependence (Odds ratio: 1.52 [1.21-1.91] for every 0.1 m/s decrease in GS; P = 0.0003). Although the prevalence of slow GS in a population of elderly patients with severe AS being screened for transcatheter valve was high, there were only weak associations between GS and other risk stratifying tools. The strong association between GS and dependent functional status suggests that assessment of gait speed is a useful, objectively measurable, risk stratification tool in this population.