Assessing the risks of treatment in Parkinson disease psychosis: An in-depth analysis

BACKGROUND: Parkinson disease (PD) psychosis (PDP) is a disabling non-motor symptom. Pharmacologic treatment is limited to pimavanserin, quetiapine, and clozapine, which do not worsen parkinsonism. A Food and Drug Administration black box warning exists for antipsychotics, suggesting increased mortality in elderly patients with dementia. However, the reasons for higher mortality are unknown. AIM: Expanding on prior work exploring mortality in treated PDP patients, we conducted a retrospective comparison to understand the links between treatment regimen, clinical characteristics, and negative outcomes. METHODS: Electronic medical record data extraction included clinically diagnosed PD patients between 4/29/16-4/29/19 and excluded patients with primary psychiatric diagnoses or atypical parkinsonism. Mortality and clinical characteristics during the study period were compared between untreated patients and those receiving pimavanserin, quetiapine, or both agents (combination). Mortality analyses were adjusted for age, sex, levodopa equivalent daily dose (LEDD), and dementia. RESULTS: The pimavanserin group (n = 34) had lower mortality than the untreated group (n = 66) (odds ratio = 0.171, 95% confidence interval: 0.025–0.676, p = 0.026). The untreated group had similar mortality compared to the quetiapine (n = 147) and combination (n = 68) groups. All treated groups had a higher LEDD compared to the untreated group, but no other differences in demographics, hospitalizations, medical comorbidities, medications, or laboratory values were found between the untreated and treated groups. CONCLUSIONS: PDP patients receiving pimavanserin had lower mortality than untreated patients. We found no other clear differences in clinical characteristics to explain the mortality risk. Prospective randomized trials are needed to definitively identify the optimal PDP treatment regimen and associated risks.