Accelerometry-Derived Physical Activity Levels and Mortality in Hemodialysis Patients: A Prospective Cohort Study

Sedentary behavior is common among hemodialysis patients although large-scale data on physical activity (PA) objectively measured by accelerometers are scarce. We (1) compared PA levels between hemodialysis patients and community-dwelling adults, 2) estimated annual PA changes, and 3) determined associations between PA levels and mortality in hemodialysis patients.

Prospective cohort study.

1,030 Japanese hemodialysis patients enrolled for PA measurements between June 2019 and July 2020 and followed for 3 years. PA data in community-dwelling adults were obtained from the Dazaifu-City Survey in Japan.

Light-intensity PA, moderate-intensity PA, vigorous-intensity PA, and steps measured by triaxial accelerometers after enrollment and 1 year later, defined based on 10-second epoch-averaged metabolic equivalents (METs) of 1.6-2.9, 3.0-5.9, and ≥6.0, respectively.

All-cause death.

Cox proportional hazards models with least absolute shrinkage and selection operator (LASSO).

The median light-intensity PA, moderate-intensity PA, vigorous-intensity PA, and steps were 192 (IQR, 128-263) minutes/day, 70 (IQR, 26-169) minutes/week, 0 (IQR, 0-0) minutes/week, and 1,832 (IQR, 680-3,822)/day, respectively. Age- and sex-adjusted geometric means of light-intensity PA, moderate-to-vigorous intensity PA, and steps were 42%, 77%, and 73% lower among hemodialysis patients than community-dwelling adults. The least-squares mean annual changes in light-intensity PA, moderate-intensity PA, and steps were -12.4 minutes/day, -10.7 minutes/week, and -215 steps/day, respectively. The lowest hazard was observed at 216-262 minutes/day for light-intensity PA, 239-291 minutes/week for moderate-intensity PA, and 4,294-6,045/day for steps. LASSO identified light-intensity PA on nondialysis days as the parameter most strongly associated with survival. The lowest mortality risk was observed at about 300 minutes/day of light-intensity PA on nondialysis days compared with about 200 minutes/day on dialysis days.

Observational study design precludes causal inference.

Light-intensity PA on nondialysis days, even for short durations, was associated with better survival in hemodialysis patients.

Sedentary behavior is prevalent in hemodialysis patients. Although current guidelines recommend 150-300 minutes/week of moderate-intensity physical activity (PA) (eg, brisk walking, cycling at a moderate pace, jogging) or 75-150 minutes/week of vigorous-intensity PA (eg, running, sprinting), it is uncertain whether these targets are applicable to hemodialysis patients who have a high prevalence of frailty. In this 3-year prospective cohort study involving 1,030 Japanese hemodialysis patients, light-intensity PA (eg, casual walking, light household activities, gardening, tai-chi) on nondialysis days was most closely associated with better survival among various PA parameters. The lowest mortality risk was observed at about 300 minutes/day of light-intensity PA on nondialysis days. Light-intensity PA, even for short durations, may be associated with favorable outcomes among hemodialysis patients.