Although T wave inversions due to cardiac memory were described already 50 years ago, little is known about the prevalence and about clinical predictors of this phenomenon.After exclusion of 238 patients due to bundle branch block or pacemaker dependency, a total of 325 consecutive patients were enrolled in this study during routine outpatient control of their pacemaker. A 12-lead standard ECG was obtained in all patients during transient inhibition of pacing therapy.Cardiac memory could be documented in 115 of 325 patients (35%) and showed a strong association with the amount of ventricular stimulation. The prevalence of cardiac memory was 9% in patients with ≤25% ventricular stimulation and 86% in patients with ≥75% ventricular stimulation.Cardiac memory was observed in one third of patients following pacemaker implantation. The prevalence of cardiac memory in the ECG with intrinsic rhythm is above 80% in patients with frequent ventricular stimulation. Cardiac memory due to ventricular stimulation is benign and should not be confused with similar T wave inversions due to acute coronary syndrome, severe left ventricular hypertrophy, or myocarditis.