BACKGROUND This study evaluated the prognostic value of combined assessment of left ventricular (LV) function and regional myocardial thallium activity in patients with nonrecent myocardial infarction and LV dysfunction. RESULTS Eighty-two patients with previous myocardial infarction (>8 weeks) and echocardiographic evidence of LV dysfunction underwent thallium-201 rest-redistribution tomography and cardiac catheterization. During the follow-up period (mean 25 months) there were 18 cardiac events (14 deaths and 4 nonfatal myocardial infarctions). Multivariate Cox regression analysis on clinical, angiographic, and thallium variables showed that the number of echocardiographic dysfunctional segments with preserved thallium uptake (> or =50% of peak activity; chi-square 11.03; p<0.005) and age (chi-square 8.12, p<0.01) were predictive of poor outcome. At incremental analysis, combined echocardiographic and thallium data provided significant additional information to clinical, thallium, and LV functional data, increasing global chi-square value from 22.4 to 31.5 (p< 0.01). Similarly, combined data gave additional information after considering clinical, echocardiographic, and LV functional data, increasing global chi-square from 17.8 to 22.3 (p <0.05). Differently, the number of diseased vessels at coronary angiography did not add further prognostic information. CONCLUSIONS In patients with previous myocardial infarction and chronic LV dysfunction, the combination of echocardiographic and thallium rest-redistribution imaging data gives prognostic information incremental to those of clinical and LV functional data and to those of each technique considered separately.