Putative Cut-Points in Sarcopenia Components and Incident Adverse Health Outcomes: An SDOC Analysis

Abstract

Objectives: Analyses performed by the Sarcopenia Definitions and Outcomes Consortium (SDOC) identified cut-points in several metrics of grip strength for consideration in a definition of sarcopenia. We describe the associations between the SDOC-identified metrics of low grip strength (absolute or standardized to body size/composition); low dual-energy x-ray absorptiometry (DXA) lean mass as previously defined in the literature (appendicular lean mass [ALM]/ht² ); and slowness (walking speed <.8 m/s) with subsequent adverse outcomes (falls, hip fractures, mobility limitation, and mortality).

Design: Individual-level, sex-stratified pooled analysis. We calculated odds ratios (ORs) or hazard ratios (HRs) for incident falls, mobility limitation, hip fractures, and mortality. Follow-up time ranged from 1 year for falls to 8.8 ± 2.3 years for mortality.

Setting: Eight prospective observational cohort studies.

Participants: A total of 13,421 community-dwelling men and 4,828 community-dwelling women. MEASUREMENTS Grip strength by hand dynamometry, gait speed, and lean mass by DXA.

Results: Low grip strength (absolute or standardized to body size/composition) was associated with incident outcomes, usually independently of slowness, in both men and women. ORs and HRs generally ranged from 1.2 to 3.0 for those below vs above the cut-point. DXA lean mass was not consistently associated with these outcomes. When considered together, those who had both muscle weakness by absolute grip strength (<35.5 kg in men and <20 kg in women) and slowness were consistently more likely to have a fall, hip fracture, mobility limitation, or die than those without either slowness or muscle weakness.

Conclusion: Older men and women with both muscle weakness and slowness have a higher likelihood of adverse health outcomes. These results support the inclusion of grip strength and walking speed as components in a summary definition of sarcopenia. J Am Geriatr Soc 68:1429-1437, 2020.

Keywords: gait speed; grip strength; hip fracture; mobility limitation; sarcopenia.

Figure 1

Association (odds ratio [OR] or hazards ratio [HR], 95% confidence interval [CI]) of putative sarcopenia variables and adverse outcomes in the Sarcopenia Definitions and Outcomes Consortium cohorts in men. Models are adjusted for age, self-rated health, pain, use of statins, cognitive function, cancer, congestive heart failure, stroke, chronic obstructive pulmonary disease. and diabetes. Hip fracture model is additionally adjusted for bone mineral density. Falls model includes 9,994 men (1,987 [19.9%] falls); mortality model includes 12,856 men (4,109 [32%] deaths); mobility limitation model includes 6,505 men (888 [13.7%] mobility limitation); and the hip fracture model includes 9,512 men (392 [4.1%] hip fractures). ALM, appendicular lean mass; BMI, body mass index; TBF, total body fat.

Figure 2

Association (odds ratio [OR] or hazards ratio [HR], 95% confidence interval [CI]) of putative sarcopenia variables and adverse outcomes in the Sarcopenia Definitions and Outcomes Consortium cohorts in women. Models are adjusted for age, self-rated health, pain, use of statins, cognitive function, cancer, congestive heart failure, stroke, chronic obstructive pulmonary disease, and diabetes. Hip fracture model is additionally adjusted for bone mineral density. Falls model includes 4,551 women (913 [20.1%] falls]); mortality model includes 4,736 women (1,258 [26.6%] deaths); mobility limitation model includes 1,500 women (344 ]22.9%] mobility limitation); and the hip fracture model includes 1,745 women (166 ]9.5%] hip fractures). ALM, appendicular lean mass; BMI, body mass index; TBF, total body fat.

Figure 3

Association (odds ratio [OR], 95% confidence interval [CI]) between grip strength and walking speed combinations and falls, mortality, mobility limitation, and hip fractures in men. Models are adjusted for age, self-rated health, pain, use of statins, cognitive function, cancer, congestive heart failure, stroke, chronic obstructive pulmonary disease, and diabetes. Hip fracture model additionally adjusted for bone mineral density. Falls model includes 9,994 men (1987 [19.9%] falls); mortality model includes 12,856 men (4,109 [32%] deaths); mobility limitation model includes 6,505 men (888 [13.7%] mobility limitation); and the hip fracture model includes 9,512 men (392 [4.1%] hip fractures). Interaction between slowness and muscle weakness was not significant for any model (P > .010).

Figure 4

Association (odds ratio [OR], 95% confidence interval [CI]) between grip strength and walking speed combinations and falls, mortality, mobility limitation, and hip fractures in women. Models are adjusted for age, self-rated health, pain, use of statins, cognitive function, cancer, congestive heart failure, stroke, chronic obstructive pulmonary disease, and diabetes. Hip fracture model additionally adjusted for bone mineral density. Falls model includes 4,551 women ([913] 20.1% falls); Mortality model includes 4,736 women (1,258 [26.6%] deaths); mobility limitation model includes 1,500 women (344 [22.9%] mobility limitation); and the hip fracture model includes 1,745 women (166 [9.5%] hip fractures). Interaction between slowness and muscle weakness was significant for falls (P = .039), borderline significant for hip fracture (P = .076), and not significant for mobility limitation and mortality (P > .10).