AIM

Geriatric experts have long debated the inclusion of appendicular lean soft tissue mass (ALSTM) in the diagnostic criteria for sarcopenia. This study examined whether a diagnostic model for sarcopenia that excludes appendicular lean soft tissue mass (ALSTM) is equivalent to one that includes ALSTM in discriminating health-related outcomes.

METHODS

Community-dwelling older adults from the Itabashi Longitudinal Study on Aging in Tokyo, Japan, were included.

Based on the AWGS2019 criteria, we developed two diagnostic models: Model 1, including low handgrip strength/gait speed and low ALSTM assessed by bioelectrical impedance analysis, and Model 2, excluding low ALSTM. Health-related outcomes were self-reported falls and hospitalization in the past year.

Differences in area under the receiver operating characteristic curve (AUC) and 95% confidence intervals (CI) between the two models were estimated using 1000 bootstrap samples.

RESULTS

In total, 2432 participants (median age: 76 years; 53.2% men) without missing data were analyzed. Falls and hospitalization were reported by 17.8% and 15.3% of participants, respectively.

The AUCs [95% CIs] of Models 1 and 2 were 0.61 [0.59, 0.64] and 0.61 [0.58, 0.64] for falls and 0.65 [0.62, 0.68] and 0.64 [0.61, 0.67] for hospitalization, respectively. The differences (Model 2-Model 1) in AUCs [95% CI] between the two models were -0.004 [-0.015, 0.006] for falls and -0.004 [-0.014, 0.004] for hospitalization, with the CIs within the pre-specified equivalence margin (±0.05).

CONCLUSIONS

Diagnostic models excluding ALSTM demonstrated equivalent discrimination ability to those including ALSTM, raising questions about the necessity of ALSTM assessment for sarcopenia diagnosis.