The incorporation of antitachycardia pacing (ATP) into implantable cardioverter defibrillators (ICDs) has provided a better tolerated alternative to shocks. ATP has been shown to be effective in terminating approximately 80% to 90% of spontaneous ventricular tachycardia (VT) episodes. Although ATP is routinely used, little is known about predictors of ATP failure. Based on the evaluation of stored electrograms, we aimed to prospectively follow patients with ICDs, and to analyze parameters affecting ATP effectiveness. One hundred eighteen consecutive patients received ICDs for standard indications. Before discharge, empirical, standardized ATP therapy was programmed in all patients within VT zones. A total of 1,218 spontaneous tachycardia episodes occurred in 51 patients during a mean follow-up of 24.5 ± 12 months. Among these, 888 VTs were diagnosed. One hundred four fast VTs were detected in the ventricular fibrillation zone and treated with primary shock delivery. ATP was attempted 881 times in the remaining 784 VT episodes. ATP terminated 640 VTs successfully, ATP failed in 55 VTs finally reverted by shocks, and 89 VTs converted to a slower VT outside the VT zone. Fifty-one of these slower VTs reverted spontaneously, and 38 were redetected and treated. Finally, in primary intention-to-treat basis, ATP was successful in 691 VTs (88%) and unsuccessful in 93 VTs (12%). There was no influence of VT cycle length on ATP success rate. Furthermore, ATP efficacy was similar between patients with left ventricular ejection fraction ≤35% or >35%, between daytime and nighttime, as well as between patients with ischemic or nonischemic cardiomyopathy. A faster heart rate immediately preceding the onset of VT (103 ± 19 vs 78 ± 14 beats/min, respectively, hazard ratio 4.08, 95% confidence interval 2.11 to 7.89, p <0.001), and absence of β-blocker therapy (82% vs 93%, respectively, hazard ratio 2.71, 95% confidence interval 1.72 to 4.29, p = 0.02) were found, by Cox proportional-hazard analysis, to be the sole independent predictors of ATP ineffectiveness in ICD recipients. Thus, the present study identified both preceding sinus tachycardia (reflecting an increased sympathetic tone) and lack of β-blocker use as independent risk factors for reduced success of ATP therapy in terminating VT. Therefore, modification of sympathetic tone may be beneficial for patients with ICDs.