Prognostic impact of peakVO2-changes in stable CHF on chronic beta-blocker treatment

Abstract

Background

Peak oxygen uptake (pVO2) is used for risk stratification in chronic heart failure (CHF), but little is known about the prognostic impact of pVO2-changes in patients on chronic beta-blocker (BBL) therapy. We therefore prospectively evaluated individual pVO2-changes at a 6-month interval in patients all receiving BBL.

Methods

194 patients with stable CHF on stable medication were included (V1) and underwent clinical evaluation and exercise testing. Testing was repeated (V2) at 5.7 ± 1.5 months after V1 and patients were followed > 12 months after V2. Death or hospitalisation due to cardiac reasons was the predefined EP (EPP, end-point positive; n = 62; EPN, end-point negative; n = 113).

Results

Initial characteristics did not differ between EPP and EPN. Multivariate cox regression analysis revealed that change of pVO2 (EPP: − 0.6 ± 2.6 ml/kg min; EPN: + 2.5 ± 3.3 ml/kg min; p < 0.001) was independent to pVO2, LVEF, NTproBNP and NYHA at V2 for prediction of the combined end-point during follow-up. An increase of pVO2 by 10% was identified as an adequate cut-off value for risk stratification and ROC-analysis showed the significant incremental prognostic value of the determination of pVO2 changes in combination with pVO2.

Conclusions

Serial measurements of pVO2 yield additional information for risk stratification in clinically homogenous CHF patients receiving BBL. This is the first study demonstrating this fact within a narrow predefined interval with all patients on BBL.

Introduction

Chronic heart failure (CHF) is the only cardiovascular disease that is characterized by an increasing incidence and prevalence [1], [2]. Fortunately, a number of treatment options have been developed in recent years [3], [4], one of the most important contributions certainly being beta-blockers [5]. These treatment options need to be employed according to the disease severity and individual risk.

Although extensive work has been devoted to identify numerous risk markers, physicians continue to have great difficulties in predicting the individual clinical course of a CHF patient [6]. Besides determination of risk predictors at a given point of time, some attention has been drawn to serial determination of risk markers [7]. This approach seems logic since parameters that reflect the patient's status should change with the status. Therefore, the idea that early changes of these parameters may precede significant clinical deterioration and that detection of these changes may allow early intervention appears tempting.

Most studies originally evaluating prognostic variables in CHF were carried out with only a minority of patients treated with BBL. Recent evidence, however, suggests that the prognostic value of variables used for risk stratification is influenced by BBL treatment itself [8]. Peak oxygen consumption (pVO2) is an established independent risk-predictor for survival and determination of the need of cardiac transplant in patients with severe heart failure [9], [10]. As BBL improve survival without affecting pVO2 [11], the value of pVO2 in predicting outcome in patients treated with BBL had to be confirmed [12]. Little and conflictive information is available to what extent changes from repeated evaluation of pVO2 can be used for prediction of the future clinical course. While some [13], [14] reported a benefit from serial analysis, others [15] could not confirm this.

When interpreting changes of parameters it is crucial to control for the interval of time these changes occur in and for the influence of BBL therapy on prognosis itself. The studies mentioned above had a large variance either of the allowed time interval between determinations or of the percentage of patients on BBL. We therefore aimed to prospectively evaluate the usefulness of relative pVO2-changes at a 6-month interval for further risk stratification in patients all receiving BBL.