标题：La especificidad de la población impacta en la predicción de los tejidos magros apendiculares para la sarcopenia diagnosticada: un estudio transversal
其他标题：La especificidad de la población impacta en la predicción de los tejidos magros apendiculares para la sarcopenia diagnosticada: un estudio transversal
摘要：Introduction: sarcopenia is a disease characterized by reduced musculoskeletal tissue and muscle strength. The estimation of appendicular lean soft tissue by DXA (ALSTDXA) is one of the criteria for the diagnosis of sarcopenia. However, this method is expensive and not readily avaiable in clinical practice. Anthropometric equations are low-cost and able to accurate predict ALST, but such equations have not been validated for male Brazilian older adults between the ages of 60 to 79 years. To this end, this study sought to validate the existing predictive anthropometric equations for ALST, and to verify its accuracy for the diagnosis of sarcopenia in male Brazilian older adults. Methods: this cross-sectional study recruited and enrolled 25 male older adults (69.3 ± 5.60 years). ALSTDXA and anthropometric measures were determined. ALST estimations with 13 equations were compared to ALSTDXA. The validity of the equations was established when: p > 0.05 (paired t-test); standard error of the estimate (SEE) 0.70. Results: two Indian equations met the criteria (Kulkarini 1: 22.19 ± 3.41 kg; p = 0.134; r² = 0.78; EPE = 1.3 kg. Kulkarini 3: 22.14 ± 3.52 kg; p = 0.135; r² = 0.82; SEE = 1.2 kg). However, these equations presented an average bias (Bland-Altman: 0.54 and 0.48 kg) and ‘false negative’ classification for the ALST index. Thus, three explanatory equations were developed. The most accurate equation demonstrated a high level of agreement (r2adj = 0.87) and validity (r²PRESS = 0.83), a low predictive error (SEEPRESS = 1.53 kg), and an adequate ALST classification. Conclusion: anthropometric models for predicting ALST are valid alternatives for the diagnosis and monitoring of sarcopenia in older adults; however, population specificity affects predictive validity, with risks of false positive/negative misclassification.
