Abstract
BACKGROUND/OBJECTIVES: Aging populations have led to numerous health challenges. The World Health Organization (WHO) proposed "Healthy Aging" to promote elderly health by optimizing Intrinsic Capacity (IC) with vitality as a core component of metabolic homeostasis. The relationships between vitality decline and inflammatory-metabolic indicators (the NLR and TyG index) remain to be investigated. METHODS: This study recruited 986 community-dwelling adults ≥ 60 years old at the Beixingjing Street Community from March 25, 2024, to June 17, 2024, in Shanghai, China. Participants underwent comprehensive face‒to-face assessments with IC evaluations conducted according to the Integrated Care for Older People (ICOPE) guidelines. Vitality was evaluated using the Mini Nutritional Assessment-Short Form (MNA-SF). The study population was divided into two groups based on vitality decline (scores < 12). Multivariable logistic regression was used to analyze associations between vitality decline and other IC domains as well as between vitality decline and inflammatory (NLR) and metabolic indices (TyG, TyG-WC, TyG-BMI, and TyG-WHtR). Three logistic regression models constructed with progressive adjustments were used to assess associations between indices (NLR, TyG, TyG-WC, TyG-BMI, and TyG-WHtR) and vitality decline. Restricted cubic spline (RCS) analyses explored potential nonlinear relationships and threshold effects. Receiver Operating Characteristic (ROC) curve analysis was used to assess the discriminative capacity of different models. RESULTS: The prevalence of vitality decline was 16.43%. Vitality decline was strongly associated with psychological and locomotor capacity decline in fully adjusted models (p < 0.05). An elevated NLR showed a significant association with greater vitality decline (OR = 1.233, p = 0.045). Participants in the highest NLR quartile (Q4) demonstrated significantly greater odds of vitality decline compared to those in the lowest quartile (Q1) (OR = 1.886, p = 0.043). Conversely, unit increases in TyG-derived indices demonstrated protective effects as follows: TyG-WC (OR = 0.988, p < 0.001), TyG-BMI (OR = 0.952, p < 0.001) and TyG-WHtR (OR = 0.120, p < 0.001). In contrast, TyG alone did not reach statistical significance (OR = 0.622, p = 0.078). The highest-quartile (Q4) participants presented a significantly lower risk of vitality decline than the lowest-quartile (Q1) participants as follows: TyG-WC (OR = 0.104, p < 0.001), TyG-BMI (OR = 0.052, p < 0.001), and TyG-WHtR (OR = 0.070, p < 0.001). Interaction terms between NLR, TyG and its indicators were analyzed in separate models and across quartiles. These terms did not show consistent significant associations with vitality decline. RCS analysis with vitality decline as the dependent variable identified threshold effects at TyG-WC = 794.358, TyG-BMI = 209.179, and TyG-WHtR = 4.476. The analysis revealed significant positive linear relationships between NLR and vitality decline, negative linear relationships between TyG and vitality decline, and negative nonlinear relationships between TyG-derived indices and vitality decline. Sensitivity analyses using MNA-SF score as a continuous variable confirmed similar relationship patterns, with NLR showing a significant linear negative association with MNA-SF score and TyG showing a linear positive association. The TyG-derived indices exhibited significant positive nonlinear relationships with MNA-SF scores, with threshold effects emerging at TyG-WC = 666.818, TyG-BMI = 210.997, and TyG-WHtR = 4.405. Effect sizes below the inflection points were substantially larger than those above thresholds, suggesting more pronounced effects at lower values. Among all biomarkers examined, NLR demonstrated the strongest association with vitality decline (AUC = 0.873). Subgroup analyses revealed significant effect modifications exclusively for TyG-derived indices but not for NLR or TyG alone. CONCLUSIONS: This study revealed a high prevalence of vitality decline among older Chinese adults, which was strongly associated with a decrease in psychological and locomotor capacity. NLR showed a negative linear correlation with vitality, whereas TyG-derived parameters showed nonlinear positive associations with a plateau effect. NLR showed optimal discriminative capacity among all indicators for identifying vitality decline.