Investigating in VigiBase over 6000 cases of pneumonia in clozapine-treated patients in the context of the literature: focus on high lethality and the association with aspiration pneumonia.

Background: The literature associates clozapine with pneumonia/aspiration pneumonia.

Research design and methods: The international pharmacovigilance database (VigiBase™) uses the information component (IC) as statistical signal. VigiBase clozapine reports were analyzed for pneumonia/aspiration pneumonia from introduction to 10 May 2023.

Results: There were 6392 cases of all types of pneumonia (5572 cases of pneumonia, 775 of aspiration pneumonia, and 45 combined). The IC was 3.52 for aspiration pneumonia, introduced as a VigiBase label in 2003, and 1.91 for pneumonia. Patients were reclassified as 3628 with no signs of aspiration and 1533 with signs. Signs of aspiration were strongly associated with some co-medications: olanzapine, odds ratio (OR) = 23.8, 95% confidence interval (CI), 14.9-38.0; risperidone OR = 18.6, CI, 11.4-30.4; valproic acid, OR = 5.5, CI, 4.5-6.6; and benzodiazepines OR = 5.5, CI, 4.5-6.6. In 2415 cases with completed data, fatal outcomes made up 45% (signs of aspiration made no difference), but there was wide variability from 0% (females <45 years of age; duration ≤30 days) to 76% (males >64 years of age; duration >1 year). During the first week, pneumonia was associated with 1) very high titration doses, 2) very small doses in Parkinson's disease, and 3) Japan vs other countries.

Conclusions: In clozapine-treated patients: 1) at least 30% of pneumonia cases may be aspiration pneumonia, 2) stopping some co-medications may decrease the risk of aspiration pneumonia, 3) average lethality in pneumonia was 45% but may be around 75% in geriatric patients with long-term treatment, and 4) safer titrations may sometimes require 5-mg tablets.