The Effect of Routine High-Soy-Protein Soymilk Intake on the Skeletal Muscle Index (SMI) in Japanese Pre-Frail Elderly Individuals with an Ordinary Life: A Post Hoc Analysis of a Randomized Controlled Trial.

BACKGROUND/

OBJECTIVES

The primary outcome of a 12-week randomized controlled trial studying the effects of continuous high-soy-protein (HSP) soymilk consumption in Japanese frail and pre-frail elderly individuals has previously been reported. The authors of this post hoc study aimed to conduct a statistical analysis focusing on changes in the skeletal muscle index (SMI) and evaluate the effect of HSP soymilk intake on appendicular skeletal muscle mass (ASM).

METHODS

A full analysis of all subjects (n = 68, HSP group: n = 34, control group: n = 34) and a subgroup analysis of only pre-frail elderly individuals (n = 58, HSP group: n = 29, control group: n = 29) were performed. The following nine subgroup analyses were conducted: frailty phenotype, basal metabolic rate, walking speed, step counts, total energy expenditure (TEE), total energy intake (TEI), estimated protein intake, age, and sex.

RESULTS

In the overall analysis (primary combined cohort), the SMI showed no significant change between the HSP and control soymilk groups (p = 0.18); however, a significant difference in the change in the SMI between the HSP and control groups was revealed for pre-frail elderly subjects (mean difference in SMI change: 0.08 kg/m2 [95% CI 0.004, 0.15], p = 0.044). Furthermore, in four distinct subgroups restricted to the pre-frail elderly individuals-step counts (≥5000 steps/day), TEE (≥30 kcal/kg BW/day), TEI (≥30 kcal/kg BW/day), and male sex-the HSP group also showed significant differences in SMI change.

CONCLUSIONS

In this study, no significant improvement in the SMI was observed across all subjects (frail and pre-frail elderly individuals); however, an exploratory subgroup analysis suggested that continued daily HSP soymilk intake was associated with potential benefits in pre-frail elderly individuals with high physical activity and energy intake levels. These findings are preliminary and require further research.