Sex-age-specific handgrip strength and mortality in an incident hemodialysis cohort: The risk explained by nutrition and comorbidities

Objective: To investigate associations of sex-age-specific handgrip strength by a dynamometer with all-cause mortality and the percent excess risk explained (%ERE) by comorbidities and nutritional indicators in incident maintenance hemodialysis patients. Methods: Prospective cohort of 413 adult patients (165 women, 248 men, 299 <60 years and 114 > 60 years) with <6 months (82% <3 months) on dialysis enrolled in PROHEMO in Salvador, Brazil. Low and high handgrip strength groups were based on sex-age-specific cutoffs (17.8 kg for women <60 years, 13.8 kg for women > 60 years, 29.5 kg for men <60 years, and 21.9 kg for men > 60 years). We used Cox regression to estimate the mortality hazard ratio. The %ERE was determined by the equation (HR1 - HR2)/(HR1 - 1) x 100, in which HR1 represented the hazard ratio in a model with a smaller number of covariates and HR2 represented the hazard ratio in a subsequent model with the inclusion of new covariates plus the variables included in the previous model. Results: The mortality hazard ratio comparing low and high handgrip strength was 2.58 (95% confidence interval: 1.73, 3.85) in the model with sociodemographic factors and vintage and 2.25 (95% confidence interval: 1.49, 3.43) with addition of comorbidities, corresponding to a %ERE of 21%. The hazard ratio was 1.98 (95% confidence interval: 1.29, 3.06) after addition of nutritional indicators corresponding to %ERE of 38%. Results stratified by age and gender followed similar patterns. Conclusion: These results provide support for the assessment of handgrip strength in all maintenance hemodialysis patients for early identification of those who may require special care to improve nutritional status and survival.