Risk of stroke associated with risperidone in dementia with and without comorbid cardiovascular disease: a population-based matched cohort study

Abstract

Background: Agitation and aggression occur in up to half of people living with dementia over the course of the disease. Although non-pharmacological interventions are used as first-line treatment strategies, antipsychotics may be indicated in severe cases. A major adverse effect of antipsychotics in dementia is stroke; the mechanism of action of atypical antipsychotic risperidone has been linked to cardiovascular disease (CVD) biological pathways in preclinical studies.

Aims: To evaluate the risk of stroke associated with risperidone across different patient subgroups defined by stroke and CVD history.

Method: Anonymised primary care data from the UK-based Clinical Practice Research Datalink were used to identify individuals diagnosed with dementia after the age of 65 years between 2004 and 2023. Risk of stroke over 1 year was compared between individuals initiating risperidone and propensity-score-matched controls across subgroups with and without history of stroke and any CVD.

Results: In the overall cohort (28 403 risperidone users and 136 324 mtatched controls), risperidone was associated with increased risk of stroke (adjusted hazard ratio: 1.28; 95% CI: 1.20–1.37). In the risperidone user group, the incidence rate of stroke was substantially higher in those with a prior history of stroke (incidence rate: 222 per 1000 person-years) and CVD (incidence rate: 94.1 per 1000 person-years) than in the overall cohort (incidence rate: 53.3 per 1000 person-years). Relative risks related to risperidone were similar across all CVD and stroke subgroup comparisons (hazard ratios between 1.23 and 1.44).

Conclusions: People with dementia with a prior history of CVD are at a significant increased risk of stroke, and risperidone further exacerbates this risk. Moreover, risperidone increases risk of stroke in patients without a prior history of CVD. This quantification of stroke risk across subgroups with and without history of CVD may help with communication of risk and aid more judicious prescribing.