Background Patients with psychotic illness have increased risk of cardiovascular death. Arrhythmias [particularly Torsades de Points (TdP)] and sudden cardiac deaths (SCD) are the most serious cardiovascular consequences. There is a recognised association between TdP and QT prolongation; the latter is a known adverse effect of some psychotropic drugs. The potential contribution of ECG abnormalities other than QT prolongation has not recently been subjected to careful investigation and little is known about their prevalence and role in the increased risk of arrhythmias and SCD.

Aim Weaimed to determine the prevalence of ECG changes in patients with mental health disorders, currently receiving antipsychotics drugs, compared with healthy control subjects.

Methods We conducted a prospective cross sectional study, looking at ECGs of patients on atypical antipsychotic medication, compared to ECGs of normal healthy subjects on no medication. ECGs were assessed on a blinded basis by two investigators. Detailed clinical characteristics of each subject were recorded, including age, sex, ECG variables and drug history. ECGs of patients known to have underlying cardiac disease or on cardio-active drugs were excluded from the analysis. QT was automatically corrected using the Bazett formula, with an upper limits of normal set at 440 ms for males, and 460 ms for females. Abnormalities recorded for this study included: sinus bradycardia (<55 beats/min, bpm) tachycardia (>100 bpm); conduction and axis abnormalities; Cornell or Sokolov voltage criteria for left or right ventricular hypertrophy; ST/T wave changes; and QTc prolongation. While not formally an abnormality, we also recorded the prevalence of early repolarisation.

Results ECGs of 131 patients (110 male, age 44 ± 10 years, mean±SD) and 70 normal volunteers (52 male, 47 ± 11 years, p=NS vs patients) were included in the analysis. While heart rate was a significant discriminator [81 ± 16 bpm in patients vs 67 ± 11 in matched controls, p < 0.001], QTc was comparable (412 ± 20 vs 408 ± 22, p=NS). When the cumulative prevalence of ECG abnormalities was computed, this was 33.5% for patients vs 20% for controls, (p = 0.04). Early repolarisation was found in 19% of patients vs 8.5% in the controls (p < 0.05). Detailed results are in the Table 1.

Conclusion Current use of antipsychotics drugs was associated with a trend towards an excess of a number of ECG abnormalities, especially sinus tachycardia. In contrast, QT prolongation did not differ between patients and controls and therefore does not appear likely to be of predictive value for poor cardiovascular outcomes. The higher prevalence of early repolarisation in patients may be of significance. Large scale prospective studies and, ideally, a registry of major and/or fatal arrhythmias in patients on antipsychotic medication would go a long way towards resolving this issue.