Background Long-term survivors of childhood cancer treated with anthracyclines may have subclinical cardiac dysfunction undetectable at a baseline evaluation. Dobutamine stress echocardiography has been proposed as a more sensitive screening test, but results of previous studies were influenced by selection criteria, infusion protocols, and side effects. Methods We applied a modified dobutamine stress test (from 5 to 10 to 15 μg/kg, infused over a 5-minute period) and evaluated the influence on stress test results of reported risk factors for late cardiac toxicity (female sex, younger age at treatment, higher dose of anthracycline, and longer duration of follow-up). Seventy-one patients (46 male, mean age 15 ± 5 years) treated with anthracyclines (median dose 240 mg/m²) 1 to 16.5 years before and 20 controls (patients’ siblings: 12 male, mean age 19 ± 4 years) were studied. Results No major side effects were recorded. One patient was unable to perform the test because of anxiety. Limiting side effects were infrequent (3%) and occurred at a dobutamine dose of ≥10 μg/kg, when significant changes of hemodynamic and echocardiographic parameters were detectable in all cases. Rest systolic and mean blood pressure and left ventricular fractional shortening were significantly lower in patients than in controls (P <.05), but no differences were found in any of the other indexes of cardiac function between the 2 groups at rest and during each dobutamine infusion step. A similar increase of global left ventricular function (percent of fractional shortening +45% vs +32%) and a decrease of end-systolic stress (–33% vs –29%) were documented. Left ventricular relaxation, early filling, and both relaxation and compliance improved. In all but one patient with reduced global left ventricular function at baseline, time-dependent patterns of hemodynamic and echocardiographic responses to dobutamine were similar. Previously described risk factors for cardiac toxicity did not influence the time changes of the echocardiographic parameters in response to dobutamine. Conclusions Compared with controls, most of our asymptomatic childhood cancer survivors, studied an average of 7 years after treatment with anthracyclines, showed normal baseline cardiac function. Our stress test was feasible and safe. Compared with modalities used in other studies, shorter infusion periods with higher dobutamine doses allowed a higher stress intensity to be reached without reducing patient compliance. At dobutamine stress test the response was comparable in patients and controls except for one patient. Previously reported risk factors for cardiac toxicity had no significant influence on stress test results. (Am Heart J 2000;140:315-23.)