Invasive Doppler catheter-derived coronary flow reserve, echocardiographic measurements of left ventricular hypertrophy and intravenous dipyridamole-limited stress thallium-201 scintigraphy were compared in 48 patients (40 were hypertensive or diabetic) with clinical ischemic heart disease and no or minor coronary artery disease. Abnormal vasodilator reserve (ratio less than 3:1) occurred in 50% of the study group and markedly abnormal reserve (less than or equal to 2:1) occurred in 27%. Coronary vasodilator reserve was significantly lower (2.2 +/- 0.8 versus 3.5 +/- 1.3, p = 0.003) and indexed left ventricular mass significantly higher (152.6 +/- 42.2 versus 113.6 +/- 24.0 g, p = 0.0007) in patients with a positive (n = 11) versus a negative (n = 32) thallium perfusion scan. Coronary flow reserve was linearly related in coronary basal flow velocity as follows: y = -0.17x + 4.59; r = -0.57; p = 0.00002. The decrement in flow reserve was not linearly related to the degree of left ventricular hypertrophy. Abnormal vasodilator reserve subsets found in hypertensive patients were defined on the basis of basal flow velocity, indexed left ventricular mass and clinical factors. In this series, diabetes did not cause a detectable additional decrement in flow reserve above that found with hypertension alone. These findings demonstrate that thallium perfusion defects are associated with depressed coronary vasodilator reserve in hypertensive patients without obstructive coronary artery disease. Left ventricular hypertrophy by indexed mass criteria is predictive of which hypertensive patients are likely to have thallium defects.(ABSTRACT TRUNCATED AT 250 WORDS)