Wasting in chronic heart failure (CHF) has long been known but is little investigated. We sought to find out whether the cachectic state in CHF provides additional prognostic information about all-cause mortality.

Between June, 1993, and May, 1995, we studied 171 consecutive patients with CHF (mean age 60 years [SD 11; range 27–86]; 17 female). We assessed exercise capacity (peak oxygen consumption; mean 17·5 mL kg⁻¹ min⁻¹ [6·7]), functional status (New York Heart Association [NYHA] class: 21 class I, 63 class II, 68 class III, 19 class IV), and left-ventricular ejection fraction (mean 30% [SD 15]; n=115). The cachectic status was defined prospectively as a non-intentional documented weight loss of at least 7·5% of previous normal weight (28 patients; range 9–36% or 6–30 kg) during at least 6 months. The Cox proportional-hazards model was used to assess the association of variables with survival, and Kaplan-Meier cumulative survival plots were constructed to estimate the influence of risk factors.

At the end of follow-up in November, 1996, 49 patients had died (after a mean 324 days [SD 283]). The mean follow-up of the survivors was 834 days (SD 186; range 549–1269). The cachectic state was predictive of 18-month mortality independent of age, NYHA class, left-ventricular ejection fraction, and peak oxygen consumption. Mortality in the cachectic patients (n=28) was 18% at 3 months, 29% at 6 months, 39% at 12 months, and 50% at 18 months. Patients who had a peak oxygen consumption below 14 mL kg⁻¹ min⁻¹ (n=53) had mortality at 3, 6, 12, and 18 months of 19%, 30%, 40%, and 51%. 18-month survival was 23% (95% CI 0–46) for the 13 patients with both of these risk factors (cachexia and low peak oxygen consumption) compared with 93% (88–98) in those (n=103) with neither risk factor (p<0·0001).

The cachectic state is a strong independent risk factor for mortality in patients with CHF. Combined with a low peak oxygen consumption, it identifies a subset of patients at extremely high risk of death. Assessment of cachexia should be included in transplant programmes and studies that investigate the effect of interventions by survival analyses.