Risk to develop psychotic episodes in Parkinson's disease under dopaminergic treatment: a case control study

Introduction:In the treatment of Parkinson's disease (PD), psychiatric complications are common. Depressive syndromes have a prevalence of up to 50%. Psychoses are rarely observed in untreated PD but are a well known phenomenon in dopaminergically treated PD patients. Known risk factors are higher age, microvascular changes, brain atrophy, cognitive impairment, and polymedication. Case reports have described psychoses under all today commonly used dopamine agonists (DA), but so far no systematic investigation concerning the potential of the different dopaminergic drugs to cause psychosis has been performed.

Methods: We scanned our pool of about 1000 outpatient for psychotic episodes (N=80) and compared them with an age-matched control group of PD patients without psychotic episodes (N=120). Examined parameters were PD medication, number of co-medication, number and kind of co-diseases, white matter lesions, brain atrophy, and dementia. T-test and odds ratio (OR) calculation was used.

Results: The most common psychotic symptom in our patients was optical hallucinations (51%), followed by acustic hallucinations and delusions (8.8%). The number of co-diseases was significantly higher in the psychosis group. There was no significant difference in the number of co-medication between both groups. The risk to develop a psychotic episode is 51.1-fold increased for patients with cortical brain atrophy, 2.6-fold increased for patients with white matter lesions and 2.0-fold increased for patient with dementia. We found the risk to developa psychotic episode to be highest for pergolid (OR 2.2) followed by ropinirole (OR 1.2), pramipexole (OR 0.5), cabergoline (OR 0.3) and apomorphine (OR 0.3). By far the lowest risk was associated with levodopa intake (OR 0.1).

Conclusions: We found that especially cortical brain atrophy is a massive independent risk factor for the development of psychotic episodes in PD. White matter lesions, dementia and the number of co-diseases are further independent risk factors. The risk for developing psychotic episodes under dopaminergic treatment seems to be increased for pergolide. As the DA with the highest risk and the one with the lowest risk are both ergoline derivates, the relative risk for developing psychotic episodes seems not to be a group effect for ergolinic or non-ergolinic DA. The clinical practice to use levodopa in PD patients suffering from psychosis under the treatment with DA seems, according to our data, justified.