The prevalence of cardiac cachexia has previously been estimated to 8-42 %. However, novel treatment strategies for chronic heart failure (CHF) have improved and decreased morbidity and mortality. Therefore, we aimed to reassess the prevalence of cachexia in an outpatient CHF clinic and to characterize a CHF population with and without cachexia with respect to body composition and related biomarkers. From 2008 to 2011, we screened 238 optimally treated, non-diabetic CHF patients for cardiac cachexia, defined as unintentional non-oedematous weight loss of >5 % over ≥6 months. CHF patients (LVEF <45 %) with cachexia (n = 19) and without (n = 19) were compared to controls with prior myocardial infarction and left ventricular ejection fraction (LVEF) >45 % (n = 19). The groups were matched for age, sex, and kidney function. Body composition was assessed by dual energy X-ray absorptiometry. The prevalence of cachexia was 10.5 %. Abdominal fat ± SD (%) was reduced in cachectic CHF: 27.4 ± 10.0 versus 37.5 ± 10.6 % (CHF, no cachexia) and 40.6 ± 8.0 % (controls), (P < 0.001). NT-proBNP levels were inversely correlated to abdominal fat in a multivariate linear regression analysis adjusted for known predictors of NT-proBNP (LVEF and NYHA); (β = -0.28; P = 0.018). Myostatin levels were reduced in cachectic CHF compared to controls (P = 0.013). The prevalence of cachexia in stable CHF, treated according to recent guidelines, is lower than previously anticipated. Body alterations in cachexia consist mainly of reduced abdominal fat mass, and its inverse correlation to NT-proBNP suggests involvement of abdominal lipolysis. Our data do not support a role of circulating myostatin as a biomarker for muscle wasting.