Chronic heart failure is a debilitating condition with a poor prognosis despite advances in medical and invasive treatment modalities. Attempts have been made to improve risk stratification in order to single out very-high-risk patients who could benefit from invasive and costly therapies. Unfortunately clinical, echocardiographic, radiological or biochemical measurements performed at rest provide significant diagnostic information but fail to stratify the risk sufficiently. Cardiopulmonary exercise testing alone or in combination with stress echocardiography and biomarkers may offer a more profound insight into the progress of the disease and patient prognosis and aid in the choice of the most appropriate therapy. Peak oxygen consumption (pVO₂) is the most powerful prognostic parameter obtained during cardiopulmonary exercise testing. Ventilation slopes, especially the slope of ventilation to CO₂ production (VE/VCO₂), can have additional prognostic value or be used as a surrogate risk predictor where exercise is sub-maximal. Other cardiopulmonary parameters seem to have, at best, an auxiliary role in risk assessment. In patients with grey zone pVO₂ values the presence of contractile reserve on stress or dobutamine echocardiography can be prognostically favourable. Other echocardiographic findings, such as functional mitral regurgitation, pulmonary hypertension and right ventricle dysfunction, also yield prognostic information. Combining biomarker measurements with pVO₂ obtained during the cardiopulmonary exercise test may further improve risk assessment in heart failure patients. Two markers, B type natriuretic peptides and high sensitivity C-reactive protein (hsCRP), both with established prognostic value in heart failure and cardiovascular diseases in general, are the best candidates for combined assessment with cardiopulmonary test results. Further studies are needed to confirm this notion. (Cardiol J 2007; 14: 340-346)