Congestive heart failure (CHF) is a highly prevalent condition and its impact will increase further in the future. Due to its poor prognosis, heart failure has been named a “malignant disease”. Several prognostic factors have been identified in CHF patients. Comprehensive prognostic scores have been based on clinical and laboratory parameters such as age, gender, comorbidity, ejection fraction, exercise capacity or haemoglobin concentrations.1 These scores have, however, often ignored the prognostic potential of neuroendocrine markers in CHF. B-type natriuretic peptides seem to be the most powerful neuroendocrine markers and their routine application has found increasing support. Longitudinal studies have demonstrated independent statistical effects of BNP and NT-pro-BNP on mortality rates. In a systematic review, Doust et al2 found that in 23 of 35 multivariable analyses CHF outcomes were predicted by B-type natriuretic peptides at higher significance levels than by any other risk marker. In nine of these analyses, BNP or NT-pro-BNP was the only significant predictor. The observed effect sizes of B-type natriuretic peptides were larger than those of ejection fraction, peak oxygen consumption or a composite score. These findings might stimulate efforts to add B-type natriuretic peptides to conventional prognostic models, although only a few authors have used them in this way. There is, rather, a tendency to replace multidimensional risk assessments with one single laboratory marker, which is thought to facilitate risk assessment in clinical routine. This unidimensional approach might, however, be overly simplistic for a complex disease such as CHF.

In this issue of Heart, Parissis and co-workers3 (see article on page 585) show that the effect of another prognostic marker, namely symptoms of depression, …