Restenosis remains the main limitation of percutaneous transluminal coronary angioplasty (PTCA). Since it seems likely that restenosis not severe enough to induce ischemia may be better detected with pharmacological testing than with exercise, we investigated whether dipyridamole thallium scintigraphy is better than exercise-electrocardiogram and exercise-thallium in predicting restenosis after PTCA. Noninvasive tests and re-angiography were performed in 61 consecutive patients, 5-6 months after successful single vessel PTCA. Detection of vessel stenosis greater than or equal to 50% was used as angiographic criteria for restenosis. Exercise-induced angina, ST segment depression greater than or equal to 1 mm at exercise-electrocardiogram and reversible perfusion defects in the area supplied by the dilated vessel, during either dipyridamole and exercise-thallium, were considered noninvasive abnormal responses. The overall restenosis rate was 41% (25/61). Angina was the most specific (97%) of all criteria for restenosis, but also one of the least sensitive (40%), slightly better than exercise-ECG (24%). Exercise-thallium had lower sensitivity (72% vs 88%, p less than 0.05) and negative predictive value (82% vs 91%, p less than 0.05) than dipyridamole-thallium. In patients positive at both exercise-thallium and dipyridamole-thallium testing, mean stenosis at follow-up was more severe (73 +/- 23%) than in patients with positive dipyridamole-thallium and negative exercise-thallium (55 +/- 26%) results, but the difference did not reach statistically significant levels. For these reasons, dipyridamole-thallium seems to be an acceptable alternative to exercise thallium to follow patients after initially successful PTCA.