1 This study re-examines the proposal that 3-adrenoceptor blockers with intrinsic sympathomimetic activity decrease plasma noradrenaline levels. 2 Thirteen patients (aged 29-65 years) with uncomplicated essential hypertension were randomly allocated to a three period, double-blind cross-over trial. The treatment periods, each of 3 weeks duration, were composed of placebo, pindolol (5 mg twice daily) and metoprolol (100 mg twice daily), dispensed in identical capsules. At the end of each treatment period, patients were exercised on a bicycle ergometer to a predetermined workload. Blood pressure and heart rate were measured before, immediately on completion of exercise and after 10 min post-exercise rest. Blood samples for plasma noradrenaline and adrenaline determination were also collected at these times. 3 Blood pressures were similar during treatment with pindolol and metoprolol. As expected, heart rate was consistently lower during metoprolol treatment. 4 Basal, pre-exercise plasma noradrenaline and adrenaline were similar at the end of each treatment period. However, the increase following exercise was significantly greater during metoprolol treatment. The post-exercise increase during pindolol treatment was indistinguishable from that in the placebo period. 5 These findings, in a randomised placebo-controlled study, therefore demonstrate that pindolol does not influence basal or exercise-stimulated plasma noradrenaline and adrenaline concentrations. This is best explained by a lack of effect of pindolol on the plasma clearance of catecholamines, which is impaired by ,-adrenoceptor blockers devoid of intrinsic sympathomimetic activity.