Prognostic factors of chronic heart failure in NYHA class II or III: value of invasive exercise haemodynamic data.

Abstract

Various parameters have been found useful for establishing the prognosis of patients with chronic heart failure, in particular haemodynamic parameters measured at rest. However, few studies deal with the prognostic value of invasive exercise haemodynamic parameters in such patients. Our aim was to test the value of such parameters to assess the prognosis of patients with chronic heart failure in functional class II or III of the New York Heart Association. Between December 1990 and June 1993, 50 patients fulfilled the criteria for inclusion in the study. The average age and ejection fraction of the population were 54 +/- 2 years and 20 +/- 0.5% respectively. All patients were examined clinically; their invasive haemodynamic parameters were measured at rest and during exercise, and their exercise peak VO2, validated by reaching the anaerobic threshold, was determined. They were followed up for 21.2 +/- 1.17 months (ranges 3-26 months). The assessment criteria employed were mortality and the occurrence of major events (heart failure, pulmonary oedema or severe ventricular arrhythmia). The population consisted of 32 men and 18 women. No gender-related differences were found. Two groups could be distinguished in terms of prognosis: G1: n = 31 living patients with no events, and G2: n = 19 deceased or with major events. G1 and G2 were comparable with regard to age, sex ratio and ejection fraction. Several of the 52 parameters tested were different in the two groups, in particular peak VO2, exercise duration, the workload developed during the cardiopulmonary test and all the invasive exercise haemodynamic parameters except heart rate (P < 0.0001). Haemodynamic parameters at rest had no prognostic value in this series. The ROC curves, the survival curves and the Cox analysis showed that exercise cardiac power output, exercise left ventricular work indices and exercise peak VO2 were the most useful factors for assessing the prognosis of patients with NYHA II or III chronic heart failure. An exercise cardiac power output < 2 watts accurately identified those patients with a short-term poor prognosis, and exercise peak VO2 was almost as accurate. To a lesser extent, the NYHA functional class was also an independent prognostic parameter during multivariate analysis. In conclusion, it appears that invasive haemodynamic parameters are best for determining the prognosis of patients with chronic heart failure. Peak VO2 can, however, be as useful. Moreover this is an easily obtainable non-invasive parameter, which makes it more useful in the evaluation and the follow up of such patients.