Background The risk of arrhythmia is usually assessed by the length of the corrected QT interval (QTc) when patients use antipsychotics. Prolonged QTc intervals are thought to increase the probability of malignant ventricular arrhythmias, and if we focus only on the QTc interval, we may be influenced by a single factor and make decisions that are not conducive to effective treatment. The index of cardiac electrophysiological balance (iCEB) is considered more valuable than the QTc for predicting drug-induced arrhythmias. It has been used in clinical practice, but no studies have observed changes in this index after the use of antipsychotics. Objective To investigate the changes in ventricular repolarization indices and the occurrence of arrhythmias in patients who have been using antipsychotic drugs for a long time, to compare the changes in iCEBc and QTc and to predict abnormal iCEBc values. Methods Patients with schizophrenia who had been hospitalized for more than 4 years and who were receiving atypical antipsychotics underwent a 12-lead synchronized electrocardiogram (ECG) every 2-4 weeks. The baseline data were measured at admission, defined as the baseline (time0), and the most obvious abnormal changes in ventricular depolarization and repolarization measured every 12 months were one-year follow-up (time1), two-year follow-up (time2), three-year follow-up (time3), and four-year follow-up (time4). Repeated measures analysis of variance was used for comparisons. The types and doses of drugs taken at 5 time points were recorded and converted into chlorpromazine equivalents for comparison. The incidence of arrhythmia during the observation cycle was recorded. ResultsThe patients had been treated with antipsychotic medication for 4 years, and the duration of the QRS wave was longer in males than in females. TpTe, TpTe/QRS, TpTe/QT, TpTe/QTc, iCEB, and iCEBc increased significantly with hospital stay, while TpTe, TpTe/QRS, TpTe/QT, and TpTe/QTc exhibited more obvious changes in these indicators in female patients (P < 0.01). The changes in iCEB and iCEBc were more significant in males (P < 0.01). The incidences of arrhythmia (arrhythmic events included premature ventricular beats and premature atrial beats) within 5 time points were 2.5%, 6.25%, 6.25%, 6.25% and 5%, respectively. More than 90% of patients treated with antipsychotics did not have any arrhythmias. No TdP syncope or other cardiovascular symptoms were found in any of the patients. Conclusion After long-term use of antipsychotics, the ventricular repolarization index gradually increased with time. The new ventricular repolarization indices iCEB and iCEBc were more sensitive than the QTc at predicting arrhythmia. According to the abnormal QTc values in men and women, we predict that the abnormal value of the iCEBc in males is 4.528 and that in females is 5.315.