The prognostic value of exercise thallium-201 myocardial perfusion imaging has not been studied in an elderly (aged greater than or equal to 70 years) population. Retrospective analysis of 120 consecutive elderly patients undergoing Bruce protocol exercise stress with quantitative planar thallium-201 scintigraphy, followed clinically for a mean of 36 +/- 12 months after testing, revealed a 10% cardiac event rate (6 cardiac deaths from arrhythmia or congestive heart failure, and 5 fatal and 1 nonfatal myocardial infarction). There were no exercise stress-related complications. Survival without cardiac events was associated with greater exercise duration (5.6 +/- 2.4 vs 3.1 +/- 2.4 minutes; p less than 0.0007) and peak exercise heart rate (131 +/- 18 vs 120 +/- 19 beats/min; p less than 0.05). Univariate variables associated with higher cardiac event rates included: (1) peak exercise less than or equal to stage I (18 vs 6%; p = 0.04); (2) maximal ST-segment depression greater than or equal to 2 mm (27 vs 6%; p = 0.003); and (3) presence of a fixed or reversible thallium-201 perfusion defect (18 vs 2%; p = 0.004). Multivariate stepwise logistic regression analysis identified the combination of peak exercise less than or equal to stage I and any thallium-201 perfusion defect as the most powerful predictor of subsequent cardiac events (relative risk = 5.3 at 1 year). Thus, exercise thallium-201 scintigraphy in elderly patients is safe and provides important prognostic information.(ABSTRACT TRUNCATED AT 250 WORDS)