Aims: To describe the temporal evolution of neurohumoral activation in survivors of myocardial infarction with left ventricular dysfunction who are initially asymptomatic and to relate this to prognosis.
Methods and results: Patients in the neurohumoral substudy (n = 534) of the Survival and Ventricular Enlargement (SAVE) study had their neurohormones measured at baseline, 3, 12 and 24 months post-infarction, were followed 38 +/- 7 months and had these values related to prognosis. All patients had a left ventricular ejection fraction < or = 40% early post-infarction. Atrial natriuretic peptide, aldosterone, norepinephrine and plasma renin activity decreased progressively over time. Patients with events had a persistent increase in these neurohormones with those dying within the first 24 months of follow-up having the greatest increase. Treatment with captopril affected only plasma renin activity (increase) and aldosterone (decrease). For patients who remained asymptomatic for the first 3 months post-infarction (n = 471), by multivariate analyses (all neurohormones together with non-neurohumoral risk factors), 3-month plasma atrial natriuretic peptide and aldosterone were the most closely related to the development of severe heart failure or to the combined end-points (cardiovascular death, myocardial infarction, or severe heart failure). No neurohormone was related to recurrent myocardial infarction or to cardiovascular mortality. When the last neurohormone measured prior to an event was considered along with non-neurohumoral risk factors (adjusted univariate), atrial natriuretic peptide, aldosterone, norepinephrine and epinephrine were associated with prognosis indicating that a time-dependent analysis identified a closer relationship between neurohormones and events than that identified by 3-month values. However, by multivariate analyses atrial natriuretic peptide was the only neurohormone associated with an event, being associated with the development of severe heart failure (P < 0.001) and the combined end-points (P = 0.022). However, when neurohormones were considered as binary variables, activated or non-activated (defined as > 1.96 SD above the mean of age-matched controls), an association between activation of norepinephrine prior to recurrent myocardial infarction (P < 0.001) and combined end-points (P < 0.01) and between activation of aldosterone and severe heart failure (P < 0.05) was identified.
Conclusions: Neurohumoral activation decreases progressively post-infarction, but only in patients with a good prognosis. In patients with a left ventricular ejection fraction < or = 40% and asymptomatic post-infarction plasma atrial natriuretic peptide at 3 months, aldosterone levels appeared to be the neurohormones most closely associated with prognosis. Increased levels of atrial natriuretic peptide, aldosterone and norepinephrine appear to be temporally most closely associated with events.