(185) Risk of Ischemic Stroke Among a Nationwide Cohort of Patients Receiving Antipsychotics: Role of Α1-Adrenergic and 5-HT1A Antagonism

Friday, August 26, 2022

8:00 AM – 6:00 PM CEST

Location: Hall C

Publication Number: 516

Presenting Author(s)

Fran�ois MONTASTRUC
– MD, PhD, CHU Toulouse - Service de pharmacologie, Toulouse, France

Co-Author(s)

Philippe D. Garcia
– CIC 1436, Team PEPSS « Pharmacologie En Population cohorteS et biobanqueS », Toulouse University Hospital, France, Toulouse, Midi-Pyrenees, France
Christel Renoux, MD PhD
– Associate Professor, Department of Epidemiology, Biostatistics and Occupational Health, McGill University, Montreal, Canada., Canada
Etienne Very
– Department of Psychiatry and Medical Psychology, University Hospital Toulouse (CHU Toulouse), France
Julien Bezin
– University of Bordeaux, Bordeaux, Gironde, France
Agnès Sommet
– Professor, CIC 1436, Team PEPSS « Pharmacologie En Population cohorteS et biobanqueS », Toulouse University Hospital, France, France
Antoine Yrondi
– Professor, Department of Psychiatry and Medical Psychology, University Hospital Toulouse (CHU Toulouse), France
Maryse Lapeyre-Mestre
– CIC 1436, Team PEPSS « Pharmacologie En Population cohorteS et biobanqueS », Toulouse University Hospital, France, France

Background: While antipsychotics are associated with increased risk of cerebral ischemic events (CIE), whether differences exist between antipsychotics is unclear.

Objectives: To compare, in a large national cohort, the risk of early CIE associated with antipsychotic with strong versus weak α1-adrenergic and 5-HT1A antagonisms.

Methods: We performed a population-based cohort study in the General Sample of Beneficiaries (Echantillon Généraliste des Bénéficiaires [EGB]), a random 1/97 permanent sample representative of the French national healthcare insurance database. We assembled a cohort of all adults, new users of antipsychotics with strong α1-adrenergic and 5-HT1A antagonisms (sulpiride, amisulpride, olanzapine, loxapine, aripiprazole, and quetiapine) or weak antagonisms (haloperidol, risperidone, cyamemazine, and chlorpromazine) between January 1st 2009 and December 31th 2018. We used an as-treated exposure definition where patients were considered exposed from the date of the first prescription and censored upon treatment discontinuation or switch. Multivariate and standardized mortality ratios (SMRW) weighted Cox proportional hazards models were used to estimate the hazard ratio (HR) with 95% confidence interval (95% CI) of hospitalisation for CIE (ischemic stroke or transient ischemic attack) associated with strong vs weak antagonists. We estimated the risk of all-cause mortality as a secondary outcome. The study protocol was registered to the European Union electronic Register of Post-Authorisation Studies (EUPAS40828).

Results: The cohort included 11 865 new users of strong antagonists and 7 478 of weak antagonists, with a total follow-up of 4 474.2 person-years and 2 854.6 person-years, respectively. Our study found a standardized incidence for CIE between 0.61 and 0.45 per 100 person-year respectively for our two groups of exposure. Compared with weak antagonists, use of strong antagonists was not associated with an increased risk of CIE in the multivariate Cox model (HR 1.37 [0.61 - 3.04]), and in the weighted Cox model (HR 1.79 [0.96 - 3.35]). However, use of strong antagonists was associated with an increased risk of death in the three models (HR 3.87 [3.08 - 4.86]; HR 1.78 [1.41 - 2.26] and HR 1.30 [1.13 - 1.50], respectively).

Conclusions: In our study, the risk of CIE did not differ between antipsychotics with strong versus weak α1-adrenergic and 5-HT1A antagonisms. However, strong antagonists were associated with a higher risk of death. Further research is needed to identify the pathophysiology of CIE associated with antipsychotics.