Evaluation of cutoffs for low lean mass and slow gait speed in predicting death in the National Health and Nutrition Examination Survey 1999-2004

Abstract

INTRODUCTION: Sarcopenia is commonly defined as loss of muscle mass with limited muscle function or strength. Different cutoffs of low lean mass and slow gait speed have been proposed by different professional working groups. We compared the performance of different cutoffs of low lean mass and slow gait speed in predicting death. METHODS: We analysed data of participants aged 65 years or older from the continuous National Health and Nutrition Examination Survey 1999-2004 (n=2841), and the subsequent follow-up data on mortality up to 31 December 2006. For low lean mass, cutoffs based on appendicular lean mass (ALM) alone, ALM adjusted for body mass index (ALMBMI), and ALM adjusted for height squared (ALMH2) were evaluated. For slow gait speed, the cutoffs based on 0.8 and 1.0 m/s were evaluated. A cox-proportional hazard regression model with adjustment for multiple confounding factors was used for the association analyses. RESULTS: For low lean mass, the cutoffs based on ALMBMI ( < 0.512 in women and < 0.789 in men) showed the most significant association and highest hazard ratio with death (hazard ratio=1.72; 95% confidence interval [CI], 1.28- 2.29). For slow gait speed, all cutoffs tested showed significant association with death in the full model (P < 0.001), while the cutoff of 0.8 m/s showed the highest hazard ratio (2.32; 95% CI, 1.58-3.39). CONCLUSIONS: Low lean mass defined by ALMBMI showed the strongest association with death; while slow gait speed showed significant association with death, with the strongest association being observed for the cutoff of 0.8 m/s. Further studies validating the cutoffs are warranted before using them in clinical settings.