The roles of branched-chain amino acids (BCAAs) and aromatic amino acids (AAAs) in sarcopenia remain conflicting, largely due to the overlooked impact of obesity. Whether these AAs differentially associate with sarcopenia without obesity (sarcopenia-only) versus sarcopenic obesity (SO) is unknown. This study evaluated their cross-sectional and prospective associations and explored dietary determinants.

This prospective study included 2994 community-dwelling Chinese older adults. Serum BCAAs and AAAs were quantified at baseline. Multivariable multinomial logistic regression examined cross-sectional (baseline and 14-year) and exploratory prospective (4-year and 14-year) associations with incident sarcopenia-only and SO. Dietary intake was assessed using a validated food frequency questionnaire.

Cross-sectionally, each 1‑SD increment in BCAAs and AAAs was associated with lower sarcopenia-only risk (all ORs < 1, P < 0.05), whereas higher valine (OR = 1.195, P = 0.032) and isoleucine (OR = 1.221, P = 0.016) were associated with increased SO risk. The 14‑year cross‑sectional analysis confirmed the positive association between BCAAs and SO (ORs: 1.604–1.668, all P < 0.01). Prospectively over 4 years, higher baseline BCAAs and tyrosine were associated with lower sarcopenia-only risk (all ORs < 1, P < 0.05). Conversely, higher baseline isoleucine and phenylalanine emerged as short-term risk factors for incident SO (ORs: 1.233–1.331, P < 0.05). Exploratory 14‑year analyses showed directionally consistent associations, with valine inversely associated with sarcopenia‑only (OR = 0.628, P = 0.024) and BCAAs positively associated with SO (ORs: 1.555–1.693, P < 0.01). Dietary patterns differed, with sarcopenia-only showing lower fiber/vegetable intake but higher red/processed meat consumption (all P < 0.05), whereas SO showed elevated hs‑CRP (P < 0.05) and tended to have a higher dietary inflammatory index.

In community‑dwelling older adults, serum BCAAs and AAAs were inversely associated with sarcopenia-only but positively associated with SO. The consistency of these associations across time points supports their utility as potential concurrent metabolic markers for phenotype differentiation. These distinct profiles, alongside dietary correlates, caution against uniform AA supplementation and suggest phenotype-specific nutritional strategies in older adults.