The severity of myocardial infarction (MI) and its functional consequences are difficult to assess in small animals. We searched for criteria to achieve such an assessment in rabbits 1 week after MI. Thirteen large mongrel rabbits (3-4 kg) were anesthetized with pentobarbitone for ligating a branch of the circumflex coronary artery and 7 rabbits were subject to a sham operation without ligation. All sham-operated rabbits and 12 MI animals survived for 1 week, when blood was obtained for biochemical analyses and the baroreflex was tested. Six animals survived to the third week (survivors) and six died earlier (nonsurvivors). The MI size, measured immediately after death, was 42 +/- 3% of the left ventricular mass in nonsurvivors and 20 +/- 7% in survivors. The plasma atrial natriuretic factor (ANF) concentration was correlated linearly with MI size (r = 0.77) over the whole range of infarct sizes and, like the MI size itself, was associated with the risk of early death (critical limit: 80 pM). Plasma renin activity and catecholamines yielded less prognostic information. The baroreflex control of the heart rate (tested using phenylephrine and nitroprusside) of nonsurvivors was severely impaired and the slopes correlated with MI size (r = 0.90 for phenylephrine and r = 0.67 for nitroprusside). The plasma ANF concentration and the baroreflex both accurately reflected MI size and also correctly classified 11/12 rabbits into survivors and nonsurvivors. An ANF- and baroreflex-based stratification of animals for future studies on therapeutic interventions after MI will reduce the number of animals required by at least 65%, making such studies far more feasible than in the past.