The effect of propranolol administration on regional coronary haemodynamics were investigated in 14 patients with stable exertional angina and isolated left anterior descending artery disease. Thermodilution was used to measure great cardiac vein flow (GCVF) and anterior regional coronary resistance (ARCR) under control conditions, at peak atrial pacing, after i.v. propranolol administration (0.1 mg kg-1) and at the peak of repeated atrial pacing. Propranolol did not change peak pacing heart rate, systolic blood pressure or double product. Peak pacing GCVF decreased slightly but non-significantly after drug administration from 84 +/- 20 to 79 +/- 24 ml min-1, while ARCR increased, but again non-significantly, from 1.36 +/- 0.44 to 1.45 +/- 0.45. Analysis of individual patient responses revealed that propranolol prolonged peak pacing time and hence peak pacing heart rate (from 126 +/- 24 to 140 +/- 23 beats min-1, P less than 0.05) in five patients. In such patients, peak pacing systolic blood pressure was lower than the pre-propranolol atrial pacing (145 +/- 35 vs 165 +/- 33, P less than 0.001) so that double product remained unchanged. Moreover, peak pacing ARCR did not change after propranolol (pre-propranolol 1.47 +/- 0.46, after propranolol 1.40 +/- 0.56 mmHg.ml-1.min, P = ns) while it increased significantly in the nine patients who did not improve after the drug (before propranolol 1.30 +/- 0.44, after propranolol 1.48 +/- 0.41 mmHg.ml-1.min, P less than 0.02). These data suggest that the response to atrial pacing after i.v. propranolol administration is variable as some patients tolerate higher heart rates while others do not.(ABSTRACT TRUNCATED AT 250 WORDS)