Abstract
Abstract
Background
Inflammation and skeletal muscle wasting often coexist in elderly populations, but few studies have examined their relationship in elderly heart failure (HF) patients. This study examined the relationship between inflammation and increased skeletal muscle proteolysis, reduced skeletal mass and strength, and their prognostic implications in elderly HF patients (> 65 years) using a random forest approach.
Methods
We prospectively enrolled consecutive elderly HF patients (n = 78) and age- and sex-matched control subjects (n = 83). We measured the interleukin (IL)-6, C-reactive protein (CRP), and B-type natriuretic peptide (BNP) levels, lower limb muscle mass and strength, and 6-min walk distance. The amount of muscle proteolysis was determined by urinary 3-methylhystidine, normalized by creatinine (3-MH/Cr). The composite endpoint was defined as all-cause death or hospitalizations due to worsening HF.
Results
Compared to controls, elderly HF patients had a significantly higher IL-6, CRP, BNP, and 3-MH/Cr, and exhibited a reduced lower limb muscle mass and strength. A correlation analysis demonstrated significant positive correlations between the inflammatory cytokine levels and 3-MH/Cr and BNP, and negative correlations with the lower limb muscle mass and strength, and 6-min walk distance. During a median follow-up of 2.4-years, 24 patients reached the endpoint. A random forest model revealed that inflammatory cytokines, skeletal muscle wasting, and the BNP had greater effects on the risk prediction. The algorithm achieved an area under the receiver operating characteristic curve of 0.887 (95% CI, 0.772–1.000).
Conclusion
This study provided evidence of the association between inflammation and increased skeletal muscle proteolysis, reduced skeletal mass and strength, and their prognostic roles in elderly HF patients.
Funder
Japan Society for the Promotion of Science
Suzuken Memorial Foundation
Publisher
Springer Science and Business Media LLC
Subject
Cardiology and Cardiovascular Medicine
Cited by
32 articles.
订阅此论文施引文献
订阅此论文施引文献,注册后可以免费订阅5篇论文的施引文献,订阅后可以查看论文全部施引文献