Headline
The study titled "Nonlinear association between albumin-globulin ratio and bone mineral density: A cross-sectional study revealing threshold effects and sexual dimorphism in US adults" examines the relationship between the albumin-globulin ratio (AGR) and bone mineral density (BMD) in a large cohort of U.S. adults. Published in the journal *Medicine* (2026), the research investigates whether AGR, a composite biomarker reflecting nutritional and inflammatory status, correlates with BMD at critical skeletal sites, including the femoral neck and lumbar spine (L1–L4). The study highlights nonlinear associations, threshold effects, and differences between sexes, offering insights into potential biomarkers for skeletal health.
Study Design and Participants
This cross-sectional analysis utilized data from 10,101 U.S. adults participating in the National Health and Nutrition Examination Survey (NHANES) between 2005 and 2018. The cohort had a mean age of 48.73 years (standard deviation 16.15), with 53.7% male participants. The study population was predominantly non-Hispanic White males, characterized by higher education levels, socioeconomic status (poverty-to-income ratio: 3.51–3.79), and lower prevalence of diabetes (4.5% vs. 13.2%) and hypertension (21.7% vs. 36.8%) compared to lower AGR quartiles. Participants were selected based on complete data for AGR and BMD measurements. The study employed weighted multivariable linear regression to assess associations between AGR and BMD, adjusting for demographic, socioeconomic, and clinical factors. Generalized additive models identified nonlinear relationships, while piecewise regression quantified threshold effects. Subgroup analyses explored interactions by sex, race, and comorbidities.
Key Findings
The study found a significant inverse association between AGR and BMD at both the femoral neck and lumbar spine (L1–L4). After full adjustment for confounders, a one-unit increase in AGR was linked to reduced femoral neck BMD (β = −0.0354 g/cm², 95% CI: −0.0464 to −0.0243) and lumbar spine BMD (β = −0.0245 g/cm², 95% CI: −0.0369 to −0.0120). Nonlinear analyses revealed inflection points at AGR values of 1.7407 (femoral neck) and 1.1714 (lumbar spine), with distinct associations below and above these thresholds. Women exhibited a 3.1-fold stronger inverse relationship between AGR and BMD compared to men (P interaction < .0001). Participants in the highest AGR quartile had significantly lower BMD than those in the lowest quartile (femoral neck: β = −0.0334, 95% CI: −0.0422 to −0.0246; lumbar spine: β = −0.0210, 95% CI: −0.0310 to −0.0110; both P < .0001). The AGR, as an integrated metric, reflects subclinical inflammatory and nutritional states rather than isolated clinical conditions like hyperalbuminemia or hypoglobulinemia, which were rare in the cohort.
Context and Implications
The study contributes to emerging research linking inflammatory and nutritional biomarkers to skeletal health. While prior studies have examined individual components of the AGR system—such as albumin’s osteogenic properties or sex hormone-binding globulin’s association with BMD—this research evaluates the composite AGR as a potential marker for skeletal homeostasis. The inverse relationship between AGR and BMD contrasts with findings that hypoalbuminemia correlates with reduced BMD, suggesting the AGR captures unique pathophysiological processes. The observed sexual dimorphism, with stronger associations in women, may reflect hormonal or metabolic differences influencing bone metabolism. These findings highlight the complexity of protein metabolism in bone health, where both deficiency and excess could negatively impact BMD. The study’s results imply that AGR could serve as a clinical tool for assessing subclinical inflammatory-nutritional states linked to skeletal health, though further research is needed to clarify underlying mechanisms.
Limitations and Future Directions
The cross-sectional design of the study limits causal inferences, as it cannot establish whether elevated AGR precedes reduced BMD or vice versa. The authors note that prospective validation of threshold effects is necessary to confirm these findings. Additionally, the study’s reliance on NHANES data, while representative of the U.S. population, may not generalize to other demographics or regions. The analysis also did not account for all potential confounders, such as dietary intake or specific inflammatory markers, which could influence BMD. Future research should explore the role of AGR in longitudinal studies and investigate its utility as a biomarker in clinical settings. The authors emphasize the need to elucidate the inflammatory and nutritional pathways connecting AGR to skeletal health, which could inform targeted interventions for bone health maintenance.