Grip Strength Criterion Matters: Impact of Average Versus Maximum Handgrip Strength on Sarcopenia Prevalence and Predictive Validity for Low Physical Performance.

Importance: Muscle strength has been elevated to the forefront of sarcopenia diagnosis, with handgrip strength the preferred measure. Extant handgrip protocols adopt different handgrip strength (HGS) criteria. Paucity of direct comparison studies assessing the impact of HGS criterion on prevalence of sarcopenia and predictive validity on physical performance contributes to the lack of standardisation of HGS criteria in sarcopenia diagnosis.

Objectives: Our study aims to compare the effect of average (HGSave) versus maximum (HGSmax) HGS criterion on: (1) prevalence of low HGS and sarcopenia; and (2) association with physical performance at baseline and at 2 years.

Methods: We recruited 200 community dwelling, cognitively intact, and functionally independent older adults. Muscle strength, physical performance measures, cognitive tests and nutritional assessments were performed. Short Physical Performance Battery (SPPB) was administered at baseline and at 2 years. We compared HGSave and HGSmax to assess the prevalence of low HGS and sarcopenia. Univariate analysis was performed comparing baseline characteristics between low and normal groups for each HGS criterion. Significantly different variables were included in logistic regression analysis to examine association of low HGS and SPPB at baseline. Predictive validity of low HGS for SPPB<10 at 2 years was examined by performing logistic regression analysis for HGSave and HGSmax.

Results: The prevalence of low HGS and sarcopenia incorporating HGSave criterion is 40% and 33% respectively, whereas that of HGSmax criterion is 21% and 19.5% respectively. There is moderate agreement between the 2 HGS criteria for sarcopenia diagnosis (kappa=0.604) and poorer agreement for low HGS (kappa=0.570). There was no significant association with baseline SPPB for both HGS criteria. At 2 years, only low HGSmax was significantly associated with low SPPB (adjusted OR 3.91, 95% CI 1.24 - 12.33).

Conclusion: Our study demonstrates that HGS criteria matters in diagnosis of sarcopenia and we support extant HGS protocols using HGSmax criterion in view of better predictive validity for poor physical performance.