Lipid accumulation product and cardiometabolic index as indicators for sarcopenia: A cross-sectional study from NHANES 2011-2018

Abstract

Although evidence suggests that lipid accumulation product (LAP) and cardiometabolic index (CMI) may be associated with the pathogenesis of sarcopenia, their relationship remains unclear. This study aims to investigate their association with sarcopenia. This cross-sectional study analyzed data from 4,172 adults aged 20-59 years from the NHANES 2011-2018 cycles. Log-transformed LAP and CMI were the primary exposure variables. Sarcopenia was defined based on the appendicular skeletal muscle mass divided by body mass index (ASM/BMI) according to FNIH guidelines (< 0.789 in males, < 0.512 in females). Weighted analyses examined the associations between LAP, CMI, and sarcopenia. Multivariable logistic regression, restricted cubic spline (RCS), and threshold analysis were used to assess associations, nonlinear patterns, and potential cutoff points. Subgroup analyses explored associations in specific populations. The dataset was randomly split into training (70%) and validation (30%) sets. LASSO regression was applied to identify key associated factors, followed by a nomogram for estimating the probability of sarcopenia. Model performance was evaluated in both the training and validation sets. Sensitivity analyses were performed using untransformed LAP and CMI to assess the robustness of the findings. Ln-transformed LAP (Ln LAP) and Ln-transformed CMI (Ln CMI) were significantly associated with sarcopenia. Threshold effect analysis identified inflection points (Ln LAP: 4.64, Ln CMI: -0.14) beyond which associations weakened. Individuals in the top quartiles of Ln LAP and Ln CMI exhibited significantly higher odds of sarcopenia (Ln LAP: OR = 8.78, 95% CI: 4.92-15.67; Ln CMI: OR = 4.44, 95% CI: 2.41-8.21). Subgroup analyses revealed stronger associations among adults aged 20-29 and 50-59 years, individuals with higher education levels, and drinkers. Classification models with Ln LAP and Ln CMI performed robustly (AUC: 0.780, 0.768) with high accuracy. Sensitivity analyses confirmed consistent nonlinear associations and dose-response relationships for untransformed LAP and CMI. LAP and CMI showed a positive association with sarcopenia in U.S. adults aged 20-59 years. The developed models highlight this relationship, offering potential guidance for identifying and managing high-risk populations.

Keywords: Cardiometabolic index; Cross-sectional studies; Intra-abdominal fat; Lipid accumulation product; Metabolic syndrome; Muscle strength; NHANES; Sarcopenia.

Fig. 1

Participant selection flowchart.

Fig. 2

RCS analysis demonstrating nonlinear associations of Ln LAP (a) and Ln CMI (b) with sarcopenia.

Fig. 3

Lasso Regression for key variables: (a) regression coefficient paths. (b) cross-validation plot.

Fig. 4

Nomogram for predicting sarcopenia: (a) Model with Ln LAP. (b) Model with Ln CMI.

Fig. 5

Evaluation of classification models on the training set using Ln LAP and Ln CMI: (a) ROC plot. (b) calibration plot. (c) DCA plot.