BACKGROUND

Sarcopenia, a condition characterized by low muscle mass, plays a critical role in the health of older adults. Early identification of individuals at risk is essential to prevent sarcopenia-related complications. This study aimed to develop a predictive model using readily available clinical nutrition indicators to facilitate early detection.

METHODS

A total of 1,002 participants were categorized into two groups: 819 with normal skeletal muscle mass (SMM) and 183 with low muscle mass (sarcopenia). A predictive model was developed for sarcopenia risk via multivariate logistic regression, and its performance was assessed using four analyses: receiver operating characteristic (ROC) curve analysis, decision curve analysis (DCA), a nomogram chart, and external validation. These methods were used to evaluate the model's discriminative ability and clinical applicability.

RESULTS

In the low-SMM group, more females (55.73% vs. 40.42%) and older individuals (median 61 vs. 55 years) were observed. These patients had lower albumin (41.00 vs. 42.50 g/L) and lymphocyte levels (1.60 vs. 2.02 × 10/L) but higher HDL (1.45 vs. 1.16 mmol/L) and calcium levels (2.24 vs. 2.20 mmol/L) (all  < 0.001). Using LASSO regression, we developed a nutritional AHLC (albumin + HDL cholesterol + lymphocytes + calcium) model for sarcopenia risk prediction. AUROC and DCA analyses, as well as nomogram charts and external validation, confirmed the robustness and clinical relevance of the AHLC model for predicting sarcopenia.

CONCLUSION

Our study employs serum nutrition indicators to aid clinicians in promoting healthier aging. The AHLC model stands out for weight-independent evaluations. This novel approach could assess sarcopenia risk in the Chinese population, thereby enhancing aging and quality of life.