Evaluating predictors of appropriate implantable cardioverter-defibrillator (ICD) therapy in patients with idiopathic dilated cardiomyopathy (IDC) may be helpful in developing risk stratification strategies for these patients. Fifty-four patients with IDC underwent ICD implantation and were followed up. Twenty-three patients (42%) had a class I indication for ICD implantation; the remaining patients underwent implantation for multiple risk factors for sudden death including left ventricular dysfunction, nonsustained ventricular tachycardia, syncope, or positive electrophysiologic study results. Clinical, electrocardiographic, and electrophysiologic data were collected. Appropriate ICD therapy was defined as an antitachycardia pacing therapy or shock for tachyarrhythmia determined to be either ventricular tachycardia or ventricular fibrillation. Appropriate ICD therapy was observed in 23 patients (42%). There was a significant difference in use of β-blocker therapy between patients who did and did not have appropriate ICD therapy (p <0.0003). Cox regression analysis identified the following univariate predictors (p <0.1): class I indication (p <0.005) and lack of use of β-blocker therapy (p <0.0007). In multivariate analysis, only lack of β-blocker use (relative risk 0.15, 95% confidence intervals 0.05 to 0.45; p <0.0007) was identified as a predictor of appropriate ICD therapy. Of the patients who received ICD therapy, only 4 (17%) were taking β blockers, whereas 21 of the 31 patients (68%) who did not receive ICD therapy were treated with β blockers (p <0.0003). In patients with IDC selected for ICD implantation, the most consistent predictor of appropriate ICD therapy was lack of β-blocker use. Attempts should be made to administer β blockers to these patients, if tolerated.