Beyond BMI: body composition phenotypes and their systemic inflammatory profiles in patients with osteoporotic vertebral compression fracture.

purposeBody mass index (BMI) cannot distinguish fat from muscle mass, masking metabolic heterogeneity. We aimed to characterize body composition phenotypes based on the intersection of BMI and low muscle mass in patients with osteoporotic vertebral compression fracture (OVCF), and to compare systemic inflammatory profiles across phenotypes.

methodsIn this cross-sectional study, 245 hospitalized OVCF patients aged ≥ 50 years were consecutively enrolled. Appendicular muscle mass was measured by dual-energy X-ray absorptiometry, and low muscle mass was defined per the 2025 Asian Working Group for Sarcopenia criteria. Four phenotypes were delineated: normal weight/normal muscle (NW/NM), overweight/obese/normal muscle (OW/OB/NM), normal weight/low muscle (NW/LM), and overweight/obese low muscle (LMO). Systemic inflammatory markers were compared across groups: platelet-to-lymphocyte ratio (PLR) and C-reactive protein (CRP) as positive markers, and prealbumin and albumin as negative markers.

resultsThe overall prevalence of low muscle mass was 53.47% (95% CI: 47.22%-59.61%). Phenotype distribution was: NW/NM 11.02%, OW/OB/NM 35.51%, NW/LM 38.78%, and LMO 14.69%. Significant overall differences were observed for all systemic inflammatory markers: PLR (P < 0.001, ε²=0.059, small), CRP (P = 0.022, ε²=0.028, small), prealbumin (P = 0.007, ε²=0.037, small), and albumin (P = 0.015, ε²=0.031, small). For albumin, prealbumin, and PLR, the OW/OB/NM phenotype consistently exhibited the most favorable profile, while the most adverse values were split between the two low-muscle phenotypes. For CRP, the NW/NM reference group had the lowest median concentration, and all three non-reference phenotypes exceeded 3 mg/L. The key findings persisted in the female-only subgroup and after age adjustment.

conclusionsLow muscle mass is highly prevalent in hospitalized OVCF patients, and body composition phenotypes beyond BMI display modestly differentiated systemic inflammatory profiles. Effect sizes were small across all comparisons. The NW/LM phenotype, hidden to BMI screening, represents the largest subgroup. When benchmarked against clinically validated risk thresholds, albumin and prealbumin levels remained above high-risk cut-offs across all groups, whereas PLR exceeded the sarcopenia risk threshold in both low-muscle phenotypes and CRP exceeded the cardiovascular risk threshold in all non-reference phenotypes. These cross-sectional associations warrant cautious interpretation and require prospective validation before clinical application.