Treatment with haloperidol has been shown, in studies using death certificates and prescription files, to be associated with an excess of sudden cardiac deaths, and regulatory warnings highlight this risk in patients with dementia. We used autopsy findings to determine whether the rate of sudden cardiac death is greater in cases of unexpected deaths of patients with dementia treated with haloperidol.

From 1989 through 2013, 1219 patients with a primary diagnosis of dementia with behavioral disturbance were admitted to a psychiatric hospital, and 65 (5.3%) died suddenly. Sixty‐five patients (5.3%) died unexpectedly. Complete post‐mortem examinations after the sudden death were performed in 55 (84.6%) patients. Twenty‐seven of the autopsied cases (49.1%) had been treated with haloperidol orally (2.2mg ± 2.1 mg/day), the only antipsychotic used in this cohort. Univariable comparisons and multivariable regression analyses compared the groups of patients with or without sudden cardiac death.

The leading causes of death were sudden cardiac death (32.7%), myocardial infarction (25.5% of patients), pneumonia (23.6%), and stroke (10.9%). Patients with sudden cardiac death and those with anatomically established cause of death were similar regarding the use of haloperidol (p = 0.5). Sudden cardiac death patients were more likely to suffer from Alzheimer's dementia (p = 0.027) and to have a past history of heart disease (p = 0.0094), and less likely to have been treated with a mood stabilizer (p = 0.024), but none of these variables were independent predictors of sudden cardiac death.

Autopsy data suggest that oral haloperidol is not associated with increased risk of sudden cardiac death in psychiatric inpatients with dementia. Copyright © 2015 John Wiley & Sons, Ltd.