Previous studies of neurohumoral activation following myocardial infarction have concentrated on the within-hospital phase and have only made measurements at rest. The objectives of this study were to measure neuroendocrine activity in the early convalescent phase of myocardial infarction at rest and during symptom-limited maximal exercise and to study the effects of early drug therapy. We studied 75 patients, mean age 57 (range 37-74) without evidence of overt heart failure, following Q-wave myocardial infarction. Patients were studied a mean of 17 days following myocardial infarction and compared with II age-matched control subjects. Plasma noradrenaline, adrenaline, atrial natriuretic peptide and plasma renin activity were measured at rest, at submaximal and symptom-limited maximal treadmill exercise. At the time of study 40 patients were taking beta-blockers, 19 diuretics and 16 no treatment. Atrial natriuretic peptide levels were higher at rest (P=0.0001) and at symptom-limited exercise (P=0.002) in the patient group than in the control subjects. Although there were no significant resting differences between the patient subgroups, at symptom-limited exercise plasma atrial natriuretic peptide levels were significantly higher in the patients taking beta-blockers than in the other patient groups (P=0.001). Plasma renin activity was no different between the patients and the control subjects at rest or. during exercise. Those patients taking diuretics had higher values at rest (P=0.001) and during exercise (P=0.005) compared with the remaining patients. There were no significant differences in resting or maximal exercise levels of plasma noradrenaline and adrenaline between the patients and the control subjects (all P>0.1). However, catecholamine levels were higher in the patients at submaximal workloads (P<0.005). The neurohumoral response has largely resolved at rest 17 days following myocardial infarction but continued activation persists during exercise.