Surrogate end points in heart failure.

Abstract

OBJECTIVE: To describe the potential utility of a true surrogate marker of heart failure outcomes, historically investigate the validity of surrogates most commonly evaluated in heart failure, and identify specific end points offering the most potential for future use. DATA SOURCES: A MEDLINE search (1966-June 2001) was completed to identify relevant literature. Additional references were also retrieved from selected articles. Search terms included b-type natriuretic peptide, cardiac remodeling, end-diastolic volume, heart failure, and surrogate end points. DATA SYNTHESIS: By definition, true surrogate end points must predict outcomes associated with disease progression and response to therapy. A validated surrogate measure of mortality would render significant utility in both heart failure drug development and clinical practice. Improvements in traditional functional markers of heart failure, including ejection fraction and exercise capacity, have produced inconsistent results in regard to mortality in a number of clinical trials. Numerous measures of cardiac remodeling and neurohormonal activation, however, have proven to be reliable predictors of disease progression and therapeutic response. These findings have contributed significantly to the current understanding of heart failure pathophysiology and approach to treatment. Specifically, measures such as left-ventricular end-diastolic volume (LVEDV) and, potentially, b-type natriuretic peptide (BNP) concentrations may correlate with disease progression and parallel the mortality reductions observed with angiotensin-converting enzyme inhibitor and beta-blocker therapy. CONCLUSIONS: Currently, LVEDV and plasma BNP offer the greatest potential as surrogate end points in heart failure. Further investigation of these measures is required before routine implementation in drug development and clinical practice can be justified.