Association of gait speed and grip strength with risk of cardiovascular events in patients on haemodialysis: a prospective study

Abstract

BackgroundGait speed (GS) and handgrip strength (HGS), both factors associated with frailty and sarcopenia, are reportedly associated with CV events in the general population. However, little is known about the impact of these factors on the outcome of patients on dialysis. This study aimed to evaluate whether evaluation of GS and HGS could be associated the onset of fatal/non-fatal cardiovascular (CV) events in patients on haemodialysis (HD).MethodsOne-hundred-eighty-two patients with end-stage renal disease (ESRD) undergoing HD at four dialysis clinics in April 2015 provided written informed consent to participate in the study. We excluded patients who had physical disability, were unable to walk without help, or had recently experienced CV events. Usual GS over a 4-m walk and HGS were measured at baseline, and 173 patients (men, 124; women, 49) were divided into sex-specific quartiles according to GS and HGS and were followed-up for fatal/non-fatal CV events for a median of 2\u2009years. We examined the association of GS and HGS with CV events and determined cut-off values using Cox regression analysis adjusted for age, sex, HD duration, history of CVD, and diabetes.ResultsDuring the follow-up period, 46 CV events occurred. Both physical performance factors were significantly associated with CV events. Low GS (<\u20090.82\u2009m/s for men and 0.81\u2009m/s for women) and weak HGS (<\u200929.0\u2009kg for men and 19.7\u2009kg for women) were associated with CV events. For low vs. high GS, the hazard ratio (HR) for CV events was 2.29 [95% confidence interval (CI): 1.20–4.33; P\u2009=\u20090.01], and for low vs. high HGS, the HR was 2.15 [95% CI: 1.00–5.04; P\u2009<\u20090.05]. These HRs remained significant after adjusting for confounding factors, such as sex, age, dialysis vintage, history of CV disease, and diabetes.ConclusionsSlow GS and weak HGS in patients on HD were suggested to be independent predictors of fatal/non-fatal CV events.