Introduction

The long-term treatment of Parkinson disease (PD) may be complicated by the development of l-dopa-induced dyskinesia (LID), which is one of the most common and disabling complications that may occur in PD patients. LID affects from 24 to 51% of medication-treated patients with idiopathic PD (Woitalla et al. 2007).

The most of the physicians were dissatisfied with current treatment strategies for dyskinesia (Woitalla et al. 2007).

Atypical neuroleptics, such as clozapine and quetiapina, which have a moderate D2 blocking effect, may improve the dyskinetic state, on the one hand, but exacerbate the extrapyramidal symptoms, on the other (Durif et al. 2004; Baron and Dalton 2003).

Although it is the dopaminergic system that is implicated most, manipulation of other neurotransmitter systems offers the most likely therapeutic approach to reducing dyskinesias and preventing their development (Brotchie 2005). The only currently available drug with a proven efficacy in improving LID is amantadine, whose effect is probably due to its N-methyl-d-aspartate (NMDA) antagonism.

Aripiprazole, a novel antipsychotic medication, appears to mediate its therapeutic effects primarily by partial agonism at the D2 and 5-HT1A receptors and by antagonism at the 5-HT2A receptor (Grunder et al. 2006).

Thanks to its partial agonism, aripripazole modulates dopaminergic and serotoninergic activity, and may thus be able to reduce LID.

Aripripazole has been successfully used to control l-dopa-induced hallucinatory states at a mean dose of 12.8 mg/day (7.5–22.5 mg/day), though with a consequent worsening in parkinsonism (Fernandez et al. 2004).

In an another open-label study, aripiprazole induces a worsening of motor symptoms in 36% of PD patients with psychosis at dosage started 1 mg/day and tritrated up to a maximum dose of 5 mg (Friedman et al. 2006).

The aim of the present study was to evaluate, in an open-label observational study, the long-term effects and safety of very low doses of aripiprazole on LIDs.

Methods

We enrolled, in a 1-year period, 12 patients with Parkinson disease, who had been referred to the Parkinson’s Disease Centre at the Department of Neurosciences of the “La Sapienza” University of Rome (five males and seven females, a mean age of 72.3 ± 7.1 years; disease duration of 15.6 ± 3.5 years; LID duration of 7.1 ± 3.4 years). The inclusion criteria were:

  • diagnosis of Parkinson’s disease according to the United Kingdom Parkinson’s Disease Society Brain Bank criteria,

  • patients had a peak-dose LID,

  • patients that a reduction of dopaminergic therapy induced a significative worsening of the Unified Parkinson’s Disease Rating Scale-III (motor part),

  • patients that had taken or taking amantadine without clinical important benefit.

The exclusion criteria were:

  • patients with supranuclear gaze palsy, signs of upper motor neuron disease, cerebellar signs, prominent autonomic dysfunction, painful or debilitating disorders, previous history of stroke and cognitive impairments, defined as a Mini Mental State Examination of less that 24 were excluded,

  • patients had “plateau” and end-dose LID,

  • patients that had not yet received a reduction of dopaminergic therapy,

  • patients that had not taken amantadine.

The patients were fully informed of the aims of the study and all gave their written informed consent. The Ethic committee of Sapienza University approved the trial. This study was performed using an open-label pilot design. The primary outcome is a significant decrease in the intensity and frequency of LID with lowest signs extrapyramidal.

A worsening of 30% respect to Unified Parkinson’s Disease Rating Scale-III (motor part) scores was considered reason to discontinue the treatment.

Dopa replacement therapy (DRT) was adjusted to a minimum effective dosage of both l-dopa and dopamine agonists; three patients were given amantadine, three patients amantadine and mirtazapine, the remaining six patients were not treated. This last group of six patients had, in the previous months, discontinued amantadine for the following reasons: onset of hallucinations (3 pts.), presence of lower limb “livedo reticularis” (2 pts.) and lack of efficacy (1 pt.). Six patients taking amantadine only or amantadine and mirtazapine continued to do so to avoid any variation in the dopaminergic stimulation that influences the intensity and duration of -induced dyskinesias. Antiparkinsonian therapy and other drug treatments were not modified during the study. All 12 patients were given aripiprazole at doses of 0.625 once a day in the evening.

The patients and their relatives recorded the presence and severity of the dyskinesia and On–Off state on a flowchart hourly every day for a period of 3 days before the baseline evaluation and after 1, 2, 3, 6, 8, 10 and 12 months. Patients were instructed on how to complete the charts, and the ability to clearly distinguish between their own motor states was a requirement for enrollment.

All patients were evaluated by the same physician at the same time of day at both the baseline evaluation and during the full ON phase (defined on the basis of the daily diary given to the patients and the relatives of each patient) using the UPDRS-III (motor part) at baseline and after 1, 2, 3, 6, 8, 10 and 12 months. Videotape recordings were made at the same times during the activating tasks provided by the administration of the UPDRS-III (motor part).

After the medical visit, videotapes were rated independently by two raters, blinded for the baseline conditions of the patients and scheduling visits of the trial, using the abnormal Involuntary Movements Scale (AIMS).

The comparison between means was analyzed using parametric (Student’s t test for paired data) and non-parametric (Wilcoxon test) tests. The statistical analysis was performed by means of repeated-measures analysis of variance for the Unified Parkinson’s Disease Rating Scale-III score and the AIMS score at baseline and after 1, 2, 3, 6, 8, 10 and 12 months in the ten patients who continued the study. Statistical significance was set at 5%; Bonferroni’s correction for multiple comparisons was applied to adjust for point-wise significance thresholds. All statistical analyses were performed using SPSS software (Version 13.0).

Results

The patients were followed for a period of 12 months; they took a aripiprazole dosage of 0.625 mg daily. Two patients abandoned the trial during the first month of therapy and after 8 months following an referred worsening in parkinsonism. The data of these two patients were not analyzed but only reported in the baseline assessment (Table 1). One patient wants to continue the treatment although the parkinsonism was worsening (UPDRS scores at baseline and final were 23 and 40, respectively).

Table 1 Clinical evaluation at baseline and end of study period
Full size table

Blood testing was performed periodically. Haemachrome and leukocyte count were normal in all cases. The low dosage minimized the occurrence of side effects other than extrapyramidal signs. We did not observe any cases of excessive sedation (i.e., somnolence). Blood tests were normal in all cases.

The ten patients who continued taking aripiprazole displayed a significant decrease in the intensity and frequency of dyskinesias (Table 1) in all parts of the body, particularly in trunk movements (Table 2); the balance in five patients who had suffered frequent falls before the administration of aripiprazole also improved following a reduction in trunk dyskinesias.

Table 2 The effect of aripiprazole on different body parts evaluated by aims subscore (number of patients = 10)
Full size table

Complete remission of dyskinesias was observed after the first month of treatment in four of the ten patients (40%) who completed the study; remission in these four patients persisted throughout the study period (Table 1).

Aripriprazole led to a significant reduction in the AIMS scores at 1, 2, 3, 6, 8, 10 and 12 months in ten patients (P < 0.001).

The AIMS data provided from the videotapes were not different from the visit “online” assessment.

Analysis of the daily diary, kept by the patients who continued the therapy, did not reveal any significant change in OFF time (T 0 = 3.67 ± 1.52 h; final = 4.33 ± 1.15 h) (P = ns). Moreover, the ON with troublesome dyskinesia time decreases at limit of significance (T 0 = 4.33 ± 1.15 h; final = 2.33 ± 1.15 h) (P = 0.08), while the ON time without troublesome dyskinesia increases at limit of significance (T 0 = 3 ± 1 h; final = 5 ± 1 h) (P = 0.08).

The assessment of patients on dyskinesia supplied by means of the diary data was in according to the clinician’s evaluation.

Aripiprazole did not significantly change the UPDRS scores at 1, 2, 3, 6, 8, 10 and 12 months in ten patients (P = ns).

A slight worsening was reported which did not, however, reach statistically significance in the motor state at T1 and T6; this worsening in the score may have been due to three patients in whom bradykinesia had worsening, though not sufficiently to suspend aripiprazole.

Discussion

At present, amantadine is the most efficacious treatment in LDS treatment; other drugs, much weaker than amantadine, such as mirtanzapine, can be used in association to potentiate amantadine or alone when amanatadine has been suspended because of the important side effects.

Atypical antipsychotics were used both in open and controlled clinical trials in the treatment of LIDs in severe PD (Durif et al. 2004; Baron and Dalton 2003).

Basal ganglia function is modulated through a wide range of serotoninergic receptors located at different points of the circuit. Clinical studies have demonstrated some improvement in dyskinetic symptoms in PD by means of substances that act on the serotoninergic system, at the 5-HT1A receptors (buspirone, mirtazapine) (Bonifati et al. 1994; Meco et al. 2003) and at the 5-HT2 receptors (ritanserin, mirtazapine) (Meco et al. 1988, 2003).

Aripiprazole, a new atypical antipsychotic drug, has a partial agonist effect at the D2 and 5-HT1A receptors, and an antagonist effect at 5-HT2 (Grunder et al. 2006). This combined action of aripiprazole may compensate for the pulsatile dopaminergic stimulation caused by DRT, and may, clinically speaking, reduce dyskinesias without impairing motor performances. Above all, aripiprazole’s partial agonism on D2 receptors may reduce motor improvement, unlike other antipsychotics.

Our data indicate that aripiprazole at a dose of 0.625 mg daily could be efficacy to control dyskinesias without worsening motor performance or increase the number of Off hours in a subgroup of PD patients who poorly responded to amantadine (alone or in association with mirtazapine) or after amantadine suspension for side effects. Besides being sufficiently tolerated, the drug did not worsen the motor performance or increase the number of On hours. The data obtained are surely limited, being from a open label trial. Other limitations in this study are the difficulties of adequately measuring dyskinesias and small size of patients. In spite of these limitations, our study is important because it is an attempt to give an adjunctive treatment to LID in a very complicated phase of PD, difficult to treat and empowering quality of life.