Treatment of psychosis in Parkinson´s disease
Abstract
This is a protocol for a Cochrane Review (Intervention). The objectives are as follows:
To determine clinical efficacy and safety (including deterioration of parkinsonism) from available treatments in Parkinson´s disease psychosis.
Background
Psychosis is a common and disabling nonmotor complication of Parkinson's disease (PD). It is defined as a disturbance of perception and thought and commonly includes hallucinations, delusions, paranoid beliefs, agitation, and delirium. Drug‐induced psychosis (DIP) can be a dose limiting side‐effect even in early monotherapy with levodopa and dopamine agonists in de novo patients and recent double‐blind placebo controlled studies have reported incidence figures of up to 17% over the first five years of treatment even in this uncomplicated group of patients (PDRG‐UK 1993; PSG 2000; Rascol 2000; Rinne 1997). A recent cross‐sectional study reported a prevalence rate of 40% when "minor forms" of psychosis like illusions or transient sensations of presence of a person were included (Fenelon 2000). The frequency of DIP increases with advanced disease and in patients with dementia where it may occur in up to 70% (Fenelon 2000). A variety of other risk factors have been associated with PD psychosis including increased age, cerebral atrophy, presence and severity of depression, abnormal REM sleep regulation, dose of dopaminergic drugs in early stages, combination therapy, and high doses of anticholinergics (Arnulf 2000; Comella 1993; Cummings 1992; Factor 1995; Fischer 1990; Friedman 1991; Klawans 1988; Lieberman 1976; Lieberman 1987; Olanow 1994; Poewe 2003; Timberlake 1978; Vezina 1992). Furthermore, psychosis is one of the main risk factors for nursing home placement and increased mortality of patients with PD (Goetz 1993; Goetz 1995).
The frequency of DIP is higher with dopamine agonist treatment compared to levodopa monotherapy (PSG 2000; Rascol 2000). A recent systematic review suggests that the odds of hallucinations with agonists is more than double that with levodopa (Ives 2004). Enhancing dopaminergic transmission by adding monoamine oxidase‐B (MAO‐B) inhibitors or catechol‐O‐methyl‐transferase (COMT) inhibitors may also occasionally trigger psychotic symptoms, although this has rarely been observed in clinical trials (Poewe 2003; PSG 1997). Anticholinergics are generally believed to be more prone to induce DIP, in particular delirium, compared to amantadine (Poewe 2003). In treating patients with DIP in PD, management of contributing and triggering factors is essential. This includes controlling medical conditions like infections, dehydration, and electrolyte disturbances or reductions of polypharmacy with antiparkinsonian and other centrally active drugs like antimuscarinic antidepressants, anxiolytics, and sedatives. When reducing and simplifying antiparkinsonian combination therapy, drugs with high‐risk benefit ratios regarding cognitive side effects versus antiparkinsonian efficacy should be tapered first, i.e., anticholinergics, amantadine but also deprenyl before reducing dopamine agonists and levodopa. Similarly, dopamine agonists should be tapered before levodopa. Because it is frequently impossible to reduce the dose of the antiparkinsonian drugs to a level that will lead to a resolution of DIP while maintaining sufficient symptomatic motor control, antipsychotic therapy becomes necessary (Poewe 2001).
Objectives
To determine clinical efficacy and safety (including deterioration of parkinsonism) from available treatments in Parkinson´s disease psychosis.
Methods
Criteria for considering studies for this review
Types of studies
Randomised controlled trials (quasi‐randomised studies will be excluded).
Types of participants
Patients of any age and both sexes suffering from Parkinson´s disease and psychotic illnesses. Psychosis is defined as a major psychiatric illness characterised by the occurrence of hallucinations or delusion (DSM III / III‐R/ IV; ICD 9 / 10). Patients with a diagnosis of drug or alcohol misuse will be excluded.
Types of interventions
Trials of typical antipsychotics (e.g. haloperidol, chlorpromazine and newer agents), atypical antipsychotic drugs (clozapine, olanzapine, quetiapine, risperidone, and newer agents) and cholinesterase inhibtors (rivastigmine, donepezil, galantamine, tacrine, and newer agents) at any dose and formulation will be compared to placebo, no treatment or to any other active drugs. In the future, trials of any new classes of drug used in the treatment of psychosis will also be eligible.
Types of outcome measures
The review will assess the following outcome:
1. Behavioural symptoms
1a) binary yes‐no (e.g. hallucinations, delusions, nightmares)
1b) scales (e.g. Clinical Global Impression Scale; Brief Psychiatric Rating Scale; Parkinson's Psychosis Rating Scale; Scale for the Assessment of Positive Symptoms; Positive Subscore of the Positive and Negative Syndrome Scale; Unified Parkinson´s Disease Rating Scale part I item 3)
2. Cognitive functioning (e.g. Mini‐Mental State Examination = MMSE)
3. Adverse effects (e.g. incidence of adverse events; withdrawal due to adverse events; use and change of antiparkinsonian treatment; clinically significant extrapyramidal side effects; change of parkinsonism (e.g. Unified Parkinson´s Disease Rating Scale); sudden and unexpected death)
4. Acceptability of treatment as measured by withdrawals from trials
5. Quality of life (e.g. PDQ 39)
6. Hospital and service outcome (e.g. hospitalisation of people in the community; duration of hospital stay; changes in hospital status)
7. all deaths
8. dead or institutionalisation (i.e. in a hospital or nursing home)
9. Social functioning and economic outcomes (e.g. Clinically significant response in social functioning; cost of care)
Search methods for identification of studies
The review will draw on the search strategy developed for the Movement Disorders Group as a whole. We will conduct searches in electronic databases from its beginning until December 2004. We will identify relevant trials from the following sources:
1. Cochrane Movement Disorders Group Specialised Register (see Review Group's module details for more information), using the terms 'parkinson' and 'psychosis' or 'hallucinosis' or 'hallucinations' or 'delusion' or 'delirium' or
'antipsychotic therapy' or 'antipsychotic treatment' or 'neuroleptic' or 'neuroleptic therapy' or 'neuroleptic treatment';
2. Cochrane Schizophrenia Group Specialised Register (see Review Group's module details for more information), using the terms 'parkinson' and 'psychosis' or 'hallucinosis' or 'hallucinations' or 'delusion' or 'delirium' or 'antipsychotic therapy' or 'antipsychotic treatment' or 'neuroleptic' or 'neuroleptic therapy' or 'neuroleptic treatment';
3. Cochrane Dementia and Cognitive Improvement Group Specialised Register (see Review Group's module details for more information), using the terms 'parkinson' and 'psychosis' or 'hallucinosis' or 'hallucinations' or 'delusion' or 'delirium' or 'antipsychotic therapy' or 'antipsychotic treatment' or 'neuroleptic' or 'neuroleptic therapy' or 'neuroleptic treatment';
4. Cochrane Central Register of Controlled trials (CENTRAL) in The Cochrane Library Issue 4, 2003);
5. MEDLINE;
6. EMBASE;
7. PsycINFO;
8. CINAHL;
9. LILACS;
We will screen titles, keywords, and abstracts of the citations downloaded from the electronic searches, and obtain full copies of reports of potentially suitable trials for further assessment.
The search strategy will also include:
10. Reference lists of located trials, and Psychosis or Cognitive Disorders in Parkinson´s Disease review articles;
11. Handsearch of the proceedings and abstracts from the International Congresses on Parkinson Disease and Movement Disorders, and from the Annual Meetings of the American Academy of Neurology (from 1996 to present)
12. Personal communication with other researchers in the field;
13. Invitation to relevant pharmaceutical companies to release any published or unpublished data they have on file.
14. If necessary, we will contact authors of published trials for further information and unpublished data.
15. Additionally, references of ongoing and recently finished clinical trials will be searched in the National Research Register, the Clinical Center at NIH Clinical Trials Databases, Current Controlled Trials, Centerwatch, Trialscentral.org and Clinicaltrials.gov.
The search strategy for MEDLINE and CENTRAL is given below. This search strategy will be adapted for the other electronic databases (EMBASE, PsycINFO, CINAHL and LILACS):
1. explode Parkinsonian Disorders/
2. parkinson*.tw
3. or/1‐2
4. explode Psychotic Disorders/
5. explode Hallucinations/
6. explode Delusion/
7. explode Delirium/
8. explode Dreams/
9. explode Aggression/
10. explode antipsychotic agents/
11. explode Cognition disorders/
12. nightmar*.tw
13. hallucin*.tw
14. delir*.tw
15. sleep*.tw
16. paranoid* and beliefs*.tw
17. neurolept*.tw
18. antipsychot*.tw
19. explode Central Nervous System Agents AND Parkinsonian Disorders AND (Psychotic Disorders OR Hallucinations OR Delusion OR Delirium OR Dreams OR Aggression)/
20. explode Neurotransmitters and Neurotransmitter Agents AND Parkinsonian Disorders AND (Psychotic Disorders OR Hallucinations OR Delusion OR Delirium OR Dreams OR Aggression)/
21. or/4‐18
22. 3 and 21
23. limit 22 to human
Data collection and analysis
SELECTION OF STUDIES
Two authors will independently assess the studies identified by the search strategy, to identify potentially suitable trials for the review according to the criteria outlined above. Disagreements will be solved by discussion with a third author.
We will independently assess the full papers for the type of participants, the type and dose of medicinal product used, duration of follow up the methodological quality by extracting details of the trial design (parallel, crossover, …), randomisation methods, the number of patients excluded or lost to follow up and the outcome measures stated in the protocol.
We will search sources of bias including: 1) selection bias, including randomisation and by chance differences in groups due to small sample sizes; 2) performance bias; 3) attrition bias; 4) detection bias; 5) selective reporting of results.
The authors will note compliance, dropouts and other exclusions from analysis.
We will independently abstract eligible data onto standardised forms, crosscheck for accuracy and amalgamate them. All results will be expressed as ordinal data or as continuous data depending on the actual measurement in the primary studies.
DATA ANALYSIS
Statistical analyses will be performed using the statistical software provided by the Cochrane Collaboration. We will test heterogeneity between trial results using a standard chi squared test. We will report the results as odds ratios (and 95% confidence intervals) for dichotomous outcomes and weighted mean differences (WMD) for continuous outcomes. The significance of any differences between odds ratios or WMD will be calculated using a standard method (Altman 1996). Where we are not able to combine outcome data from different studies we will give a descriptive summary of the results.
QUALITY ASSESSMENT
Methodological quality for each study will be independently assessed by two authors using a schedule according to the Cochrane Reviewers´ Handbook (Alderson 2004).
Disagreements will be resolved by consensus or by the involvement of a third author. The inter‐rater agreement between authors will be noted. The scores are designed to give readers a general impression of trial quality and will not be used in any quantitative manner. Important aspects of the methodology considered are blinding, placebo control, reporting of dropouts and randomisation. In terms of randomisation, if the methods were clearly described and concealment was felt to be adequate they will be termed 'A'. Trials in which patients were said to be allocated randomly but in which the method of randomisation was unclear will be included and coded B. If studies were randomised but treatment was felt to be inadequately concealed, these will be classed as 'C'. Studies classified as 'C' will be excluded for data analysis.
Rating of the study quality scores will be derived from a list of key methodological topics following a published checklist (Dixon 1997), which was successfully used for a systematic review supported by the Movement Disorders Society (MDS) to assess the efficacy and safety of each of the different antiparkinsonian therapeutic interventions based on the robustness of clinical evidence (MDS 2002). Methodological reliability of the trial will be calculated for each study upon this checklist and a percentage score will be given for the study as an indicator of the overall quality of the study.
Two of three authors will assess quality scores to ensure reliability. If differences occur in the appraisal between the two raters a third rater will be called in for discussion in order to find consensus.
Each of the following questions will be scored as YES = 2, Unclear/possibly = 1, No = 0, or not applicable = N/A, and the total score summed:
Results
1. Is an estimate of the treatment effect given?
2. Is it of clinical importance?
3. Is the estimate of treatment effect sufficiently precise?
Validity: Selection
4. Was the spectrum of patients well defined?
5. Was the diagnosis of the disease well defined?
6. If pragmatic, were suitably broad eligibility criteria used?
7. If explanatory, were eligibility criteria suitably narrow?
Measurement
8. Was assignment to treatments stated to be random?
9. If yes, was the method of randomisation explained and adequate?
10. Were all patients accounted for after randomisation?
11. Were losses to follow up low (less than 10%)?
12. Were the treatment groups similar in important factors at the start of the trial?
13. Were all patients otherwise treated alike?
14. Were patients, health care workers and investigators "blind" to treatment?
15. Was assessment of outcome "blind"?
16. Was the occurrence of side effects explicitly looked for?
17. If yes, were estimates of their frequency/severity given?
Statistical Analysis
18. Was the main analysis on "intention to treat"?
19. If no, was a sensitivity analyses performed?
20. Were additional clinically relevant factors allowed for?
21. Were appropriate statistical methods used?
22. Were any "unusual" methods explained or justified?
23. If subgroup analyses were done, were they explicitly presented as such?
Utility
24. Do the results help me choose treatments?
The overall rating of the quality of a study will be obtained from the sum of the scores divided by the maximum possible score ( = 48).
This will then be expressed as a percentage.
