Can drug effects on mortality in heart failure be predicted by any surrogate measure?

Abstract

Clearly at present, the one perfect surrogate marker for mortality remains elusive. Chronic heart failure is a complex syndrome: as such it may perhaps be too simplistic to expect any single parameter to be universally predictive of drug effects on mortality, especially when each drug works by different mechanisms. Nevertheless, neurohormonal antagonists, such as ACE inhibitors and beta-blockers, seem to benefit both mortality and all surrogate markers of mortality. Equally, inotropic drugs and Class I antiarrhythmics appear to worsen both mortality and many surrogates. This is encouraging. However, significant discrepancies exist, particularly for digoxin, ibopamine and hydralazine-nitrates, although it is only with the latter two that diametrically opposite effects occurred, whereby favourable surrogate effects turned into unfavourable mortality effects (or vice versa). It appears appropriate to have guarded optimism about the potential use of these surrogates to predict drug effects in chronic heart failure. Given our current understanding, none of the parameters discussed above is perfect when used alone. Perhaps a battery of surrogates would be more appropriate rather than there being any single surrogate. The most promising surrogates are heart rate variability, QT dispersion and plasma neurohormones, the first two for sudden death and the last one for death from progressive disease.