Study objective: We sought to characterize the ECG changes associated with symptomatic β-blocker overdose. Methods: The study population consisted of a prospective cohort of patients reporting to 2 regional poison centers with β-blocker overdose. Each patient received an ECG on presentation and a structured follow-up. The inclusion criteria for symptomatic overdose included heart rate of less than 60 beats/min or systolic blood pressure of less than 90 mm Hg; symptoms consistent with decreased end-organ perfusion; therapeutic intervention with cardioactive medication; and corroboration by 2 of the authors that this was a clear-cut case of symptomatic β-blocker overdose with cardiovascular toxicity. Exclusion criteria included cardioactive coingestants, age younger than 6 years, and no available ECG. Results: Of 167 patients, 13 were determined to have symptomatic exposures. First-degree heart block (>200 ms) was the most common ECG finding (10/12) and also had the greatest likelihood ratio (5.31) when comparing those with symptomatic exposures with those with asymptomatic exposures. Comparing the asymptomatic with the symptomatic groups, the mean PR interval was 167 ms (95% confidence interval [CI] 162 to 171 ms) versus 216 ms (95% CI 193 to 238 ms), the mean QRS interval was 89 ms (95% CI 87 to 91 ms) versus 112 ms (95% CI 92 to 132 ms), the mean QTc interval was 422 ms (95% CI 417 to 428) versus 462 ms (95% CI 434 to 490 ms), and the mean heart rate was 72 beats/min (95% CI 69 to 74 beats/min) versus 66 beats/min (95% CI 59 to 73 beats/min). Two cases of symptomatic acebutolol exposure appeared unique by demonstrating disproportionate prolongation of the QTc interval, an RaVR height of 3 mm or greater, and associated ventricular tachydysrhythmia. Conclusion: The majority of clinically significant β-blocker intoxications demonstrate negative dromotropic effects on ECG. Several ECG differences in acebutolol intoxication might reflect unique pathophysiologic processes relative to other β-blockers. [Ann Emerg Med. 2002;40:603-610.]