To the Editor: A recent multicenter study¹ reported that the prevalence of impulse-control disorders (ICD) was 2.72-fold higher in Parkinson's disease (PD) patients treated by dopamine agonists than in PD patients not treated by these medications. Several case reports of attempted suicide or suicide in patients treated by dopamine agonists with pathological gambling have been published.² A recent cross-sectional study³ on 128 PD patients examined associations between current suicidal and/or death ideation and neurological, neuropsychological, and psychiatric variables. Interestingly, a history of ICD disorders was not a statistically significant factor, but was associated with a trend (p<0.07). On multivariate logistic regression, major depression, psychosis, and elevated score on the Beck Hopelessness Scale were independent predictors of current suicidal and/or death ideation.

We report the case of a 44-year-old man treated for juvenile idiopathic Parkinson's disease (PD) with pramipexole (0.7 mg three times daily) and l-dopa-benzeraside (62.5 mg three times daily) for 2 years. The dose of pramipexole was then increased to 0.7-mg tablets 4.5 per day. Several days later, he developed severe ICD with hypersexuality: increased libido, visiting of pornographic websites, and adultery. Six months later, pramipexole was decreased, then stopped, and l-dopa was continued alone. Hypersexuality rapidly resolved, but the patient experienced mixed anxiety/depressive disorder, and he was then admitted to the psychiatric department of a general hospital for 3 weeks and was treated with fluoxetine (20 mg/day). At the last psychiatric visit, 4 weeks after discharge from hospital, the patient's main complaint was related to his sexual activity; he said that he had sexual intercourse much less frequently than while he was taking pramipexole. Despite his marked anxiety, he did not present any symptoms of depression. The following day, he experienced a benign fall while jogging and then committed suicide by hanging from a tree, 2 months after withdrawal of pramipexole.

This case report illustrates, for the first time, a potential link between suicide and ICD. Suicide due to depression was excluded by the psychiatric assessment on the day before the patient's death. When pramipexole was stopped, compulsive hypersexuality diminished dramatically and then resolved. The patient was then unable to accept his “normal” sexual activity and experienced a kind of pathological bereavement leading to hopelessness. Another hypothesis could be that the suicide was the result of impulsive behavior secondary to dopamine-agonist treatment, but the fall appeared to be the triggering event in a “hopelessness” patient. This potential cause can be excluded because L-dopa treatment was stable at low dose (187.5 mg/day). However, duration of the illness usually facilitates psychological acceptance and elaboration of the illness by the patient, as proposed by several authors using the term “work of disease” or “disease experience.” The patient's paradoxically increased physical or sexual capacities could considerably interfere with this “work of disease” and increase the risk of suicide. This case report suggests the need to identify patients at risk of suicidal behavior before administering dopamine agonists. The presence of impulse-control disorder and not only depression must therefore be investigated in PD patients who attempt suicide.