A Preliminary Nomogram for Sarcopenia Screening in Cardiology Outpatients Incorporating Cardiac Structure, Function, and Clinical Variables.

OBJECTIVE

To investigate associations between routine cardiac measurements and prevalent sarcopenia in cardiology outpatients and develop a preliminary nomogram to support screening.

DESIGN AND

METHODS

This single-center cross-sectional study included 86 adults aged 50 years or older. Participants underwent electrocardiography, transthoracic echocardiography, and sarcopenia assessment using the 2019 Asian Working Group for Sarcopenia criteria. Multivariate logistic regression used univariable screening at p < 0.20 and backward elimination with retention at p < 0.10. Internal validation used 1000 bootstrap resamples with fixed predictor sets. A three-variable model was examined in an exploratory sensitivity analysis.

RESULTS

Sarcopenia was present in 30 participants (34.9%). Age ≥ 65 years (odds ratio [OR] 6.58, 95% confidence interval [CI] 1.83-23.62), diabetes mellitus (OR 4.68, 95% CI 1.38-15.88), and abnormal left ventricular internal diameter at end-systole (LVIDs; OR 5.36, 95% CI 1.01-28.55) were significantly associated with sarcopenia. Abnormal E/A ratio and left ventricular posterior wall thickness at end-diastole (LVPWd) were retained under p < 0.10 but were not statistically significant and remain exploratory. The apparent AUC was 0.848 and the optimism-corrected AUC was 0.816. Corrected calibration intercept and slope were -0.124 and 0.679; the slope below 1 indicates overfitting. Decision curve analysis showed positive net benefit in the development sample without optimism correction.

CONCLUSION

In this small, single-center cross-sectional study, older age, diabetes mellitus, and abnormal LVIDs were associated with prevalent sarcopenia. The internally evaluated five-variable nomogram remains a preliminary screening-support model; E/A ratio and LVPWd findings are exploratory. External validation and clinical impact assessment are required before routine use.