Part 3: Neuropsychiatric Symptoms

, ,
Supplements and Featured Publications, Parkinson’s Disease: Quality Assessment and Improvement of Care, Volume 14, Issue 2 Suppl

The nonmotor neuropsychiatric symptoms of Parkinson’s disease, particularly depression, psychosis, and cognitive impairment/dementia, are major contributors to disability and a decline in quality of life. Their effect on patients may be more disabling than motor symptoms. Increasing awareness of the importance of recognizing and treating neuropsychiatric symptoms of this disease in the medical community is a focus of specialists and organizations. This article looks at useful screening measures to help clinicians recognize neuropsychiatric symptoms and offers suggestions for their effective treatment.

(Am J Manag Care. 2008;14:S59-S69)

Parkinson’s disease (PD) is the second most common degenerative neurologic disease. Patients with PD live with significant disability, a poor health-related quality of life (HRQOL), and a higher risk for early death compared with the general population.1-5 Healthcare costs for the management of PD are substantial and expected to rise in the future.6

With the improvements in the treatment of motor symptoms, nonmotor symptoms of PD have been increasingly recognized as a major cause of disability, particularly neuropsychiatric disorders and cognitive impairment.7,8 A recent large patient survey found a greater impact of psychiatric-related symptoms on HRQOL in earlier phases of the disease compared with later stages.9 In studies, nonmotor symptoms, particularly depression and cognitive dysfunction, have been shown to contribute equally or more so to impairment in activities of daily living than limitations imposed by motor impairment.8,10 Collectively, these data suggest the need for earlier evaluation and treatment of nonmotor symptoms in PD, which potentially could improve HRQOL and patient productivity, reduce morbidity, and minimize direct and indirect healthcare costs.

However, most PD-related neuropsychiatric symptoms remain underrecognized and undertreated in clinical practice.11-13 Given the increasing number of PD cases, there is a need to more adequately recognize, diagnose, and treat comorbid psychiatric and cognitive disorders. This article focuses on neuropsychiatric manifestations, which, for the purposes of this article, will include depression, anxiety, sleep disorders, psychosis, cognitive impairment/dementia, and impulse control disorders.


2. Have you lost interest or pleasure in day-today activities?

Treatment. Despite the high prevalence of depression in PD, there are relatively few clinical studies that have evaluated the efficacy of antidepressants in this population. Selective serotonin reuptake inhibitors (SSRIs) and tricyclic antidepressants (TCAs) have been studied most frequently, and in small or open-label studies these agents have shown positive effects, although often modest.10,16,19,20 The paucity of well-controlled studies and/or inadequate trial design appear responsible for the disappointing efficacy experienced in clinical practice, due to the lack of clear guidelines. In a recent literature review and meta-analysis,21 fewer than 30 studies evaluating the effectiveness of antidepressants in PD were identified. Of these studies, 11 were found to be suitable for inclusion in the meta-analysis and only 2 were placebo controlled. All studies involved small populations (N = 8-93), and diagnostic criteria for depression were used in fewer than half. Almost all studies failed to define what constituted a response to treatment. The meta-analysis revealed large effect sizes, reflecting benefit, for both placebo and antidepressant treatment, but these were not significantly different.

The main conclusion of this meta-analysis was that antidepressants have not been adequately evaluated for depression in PD. The large effect size for active treatment suggests that antidepressants have a beneficial effect, but those effects may be nonspecific. Well-conducted, controlled studies with a sufficient number of patients to detect clinically and statistically significant treatment effects are lacking. Similar conclusions were drawn by the AAN in their evidence-based review, citing similar deficiencies in available studies.7 Based on their review of the literature, the AAN did recommend amitriptyline, a TCA, for depression in PD, although TCAs may not be well tolerated by some patients with PD.

It does appear that the most frequently used antidepressants, the SSRIs, are generally well tolerated in patients with PD.20,21 Contrary to prior belief, SSRIs do not appear on average to worsen motor symptoms of PD; in fact, motor symptoms may improve in some patients.10,20,21 In open trials, most PD patients (>85%) were able to complete antidepressant treatment without significant adverse effects.

Table 1

Antidepressants should continue to be prescribed for depression associated with PD, since they can be effective and may help comorbid psychiatric symptoms, including anxiety and sleep disorder symptoms.19 10,16,20,22,23 lists some commonly prescribed antidepressants that have been used specifically in patients with PD and their cost. SSRIs are preferred for initial therapy, because of their good tolerability, and maximal doses should be given, if needed and tolerated, to optimize benefit. There is no good evidence that one SSRI is more effective or causes fewer adverse effects than another.22 However, citalopram, escitalopram, and sertraline are less prone to drug–drug interactions than paroxetine (or fluoxetine, not shown)22 and may be better suited for the patient with PD. Formal cost-effectiveness analyses for PD depression are unavailable, although effective management of depression is a clear priority that will likely lead to long-term cost savings.

TCAs might be considered in patients unresponsive to an SSRI, although their anticholinergic effects are especially problematic in patients with PD since they may also worsen cognition or aggravate orthostatic hypotension.7,10 Other non-SSRI antidepressants, including venlafaxine, duloxetine, mirtazapine, and bupropion, may also be useful to augment or replace therapy in SSRI partial or nonresponders.

In those patients failing to respond adequately, or if depression is severe or complicated by suicidal ideation, referral to a psychiatrist is advised.

AnxietyAnxiety in PD has not been as well studied as depression. However, the majority of patients with depression will also meet criteria for an anxiety disorder and vice versa.3,10 Anxiety can be associated with greater physical and psychological distress than depression in PD.

Anxiety disorders in PD can be categorized as generalized anxiety disorder, panic disorder, social phobia, and obsessive-compulsive disorder (OCD). Generalized anxiety disorder and panic attacks are the most common manifestations. While patients often suffer from avoidance of being seen in a public place due to fear of embarrassment, this does not qualify for the diagnosis of social phobia. True OCD is uncommon. OCD and related symptoms may actually be no more prevalent in PD than in subjects without PD.24

There is some evidence that generalized anxiety or panic attacks in PD are a reaction to the distressing components of the disease (eg, to the discomfort and fear of loss of motor control).16 However, many specialists feel that both psychosocial and neuropathologic factors are contributory.

Patients with PD should be screened for anxiety disorders. Although no well-established instrument is available, the self-administered Beck Anxiety Inventory (BAI) has shown usefulness for distinguishing anxiety from depression. This scale is comprised of 21 items and takes 5 to 10 minutes to administer. The Spielberger State-Trait Anxiety Inventory is also useful to distinguish anxiety, but consists of more questions than the BAI and takes about 10 minutes to complete. Similar to depression, simply asking the patient a few questions about generalized anxiety symptoms, or if they have experienced anxiety attacks, may be equally as useful and informative as the use of rating scales.

Treatment. Behavior modification techniques may benefit some patients with anxiety disorders related to PD. However, pharmacologic therapy is often needed. Certain antidepressants are approved for the treatment of a range of anxiety disorders; these agents include paroxetine, escitalopram, and venlafaxine.22 Benzodiazepines should be considered for more severe anxiety, although patients should be counseled in regard to potential adverse effects, such as sedation, further cognitive impairment, and balance problems, which may increase the risk of falls. Buspirone is well tolerated, but it has not been formally tested for anxiety in PD. Similar to the general population, SSRIs may be helpful in the treatment of panic disorder, social phobia, and OCD in PD.

Disorders of Sleep and Wakefulness

Psychosis in PD is associated with an increase in functional disability, greater caregiver burden often resulting in nursing home placement, and a higher risk of mortality.2,8,10,29 Psychosis in the form of hallucinations or illusions, typically visual, occur in 10% to 40% of patients who are receiving PD pharmacotherapy; delusions are seen less commonly in the subset of patients with hallucinations.2,8,10 These data point to a major role of dopaminergic therapy in the etiology of PD-associated psychosis, and this has been demonstrated.10,16 In this regard, evidence to date is insufficient to suggest that one type of medication is more likely than another to induce psychosis. In contrast, psychotic symptomatology has been observed in 5% to 10% of patients with PD who were not receiving dopaminergic medications, including the time prior to development of levodopa.10,16

Given the percentage of cases occurring in the absence of medications, other factors appear to play a role. A complex interaction of drug therapy, underlying neuropathology, and comorbidities (eg, cognitive or visual impairment) likely contribute to psychosis.10,30 This is supported by the documented risk factors for psychosis in PD; that, in addition to medication exposure, include greater cognitive dysfunction, older age, longer duration of PD, visual disturbances, comorbid depression and anxiety, and sleep disorders, such as RBD.10,16,25

Auditory hallucinations occur less commonly than visual hallucinations.2 In general, patients with psychosis related to PD may be divided into 2 groups. In the first, hallucinations are the only abnormality and are often benign, in that they are untroubling to the patient. These patients may not require treatment.10 The second group experiences more complex and disturbing symptoms, such as hallucinations with concomitant delusions or in the context of delirium.2,10 Delirium can occur in the context of advanced dementia or be induced by concurrent medical conditions or dopaminergic therapy.2 Patients with more complex or severe psychosis usually require treatment. Patients with relatively benign psychosis can advance to the more severe form as the disease progresses.2

Dementia with Lewy bodies (DLB) is also characterized by hallucinations, often with fluctuations in cognition.16 In general, the presence of psychosis early in the course of parkinsonism or prior to the introduction of dopaminergic therapy suggests a diagnosis of DLB.10 Intolerance of antipsychotic medications is also a feature supporting a diagnosis of DLB.31

A valid screening instrument for psychosis in PD is lacking. However, the Parkinson Psychosis Rating Scale has shown some usefulness and is relatively easy to administer.7,10 The Neuropsychiatric Inventory, which includes psychosis subscales, is another instrument that can be clinically useful and can be self-administered by a family member or caregiver. As insight is often retrained, simply asking the patient a few relevant questions during a routine clinic visit may also uncover psychotic symptoms.

Treatment. Treatment of psychosis in PD is complicated by the desire to balance dopaminergic therapy (which can aggravate psychosis) and antipsychotic medications, many of which can aggravate parkinsonism. A review of medications taken by the patients is necessary to eliminate those that might be contributing to psychosis. Nonessential drugs not being taken for PD are the first consideration, especially central nervous system active drugs. This should be followed by a risk–benefit analysis of antiparkinson medications, any of which can contribute to the development of psychotic symptoms.10,32 A risk–benefit model based on expert opinion has been developed for the latter step, which suggests reducing or discontinuing medications in the following order16,32: anticholinergic agents; selegiline (deprenyl); amantadine; dopamine agonists; catetchol-O-methyltransferase inhibitors; controlledrelease levodopa; and immediate-release levodopa. A point will be reached in this process where further reduction in medications will significantly jeopardize motor-function control, and if psychosis is persistent and problematic at this point, then an antipsychotic agent should be added.10,16,32 Typical (or first-generation) antipsychotic agents, such as haloperidol, are not indicated because of their propensity to worsen symptoms of PD.16 The currently recommended agents are the atypical (or second-generation) antipsychotic agents quetiapine and clozapine.7,10,19 While only clozapine has clearly demonstrated efficacy for psychosis in PD, both agents appear to be well tolerated from a motor standpoint.7,19 Except for olanzapine, the other atypical antipsychotic agents have not been well studied in this setting. In available studies, olanzapine did not improve PD psychosis and worsened motor symptoms.7 Quetiapine has been shown effective in treating PD psychosis in open-label studies, but 2 placebo controlled studies have been negative.15,29,33 Although mild worsening of motor symptoms may occur with quetiapine, it is considered the agent of choice for psychosis in PD, based on the results of open-label studies and the clinical experience of movement disorders specialists.10

Clozapine has demonstrated effectiveness in the treatment of psychosis in placebo-controlled studies and does not aggravate motor symptoms.10,16,32,34 However, it can rarely induce agranulocytosis and requires weekly blood counts during the first 6 months of therapy, bimonthly for the next 6 months, and then monthly thereafter.19 This additional clinical surveillance greatly increases cost of care, and this side effect of the medication places the patient at risk for a potentially serious adverse event. This is not required for quetiapine, although atypical antipsychotics have warnings for increased morbidity (cerebrovascular events) and mortality when used in patients with dementia. In general, clozapine is recommended as an alternative agent in patients unresponsive to or unable to olerate quetiapine.10 Cholinesterase inhibitors (eg, rivastigmine) also may have antipsychotic properties in PD dementia and DLB; therefore, they might be considered for use in psychotic patients with comorbid cognitive impairment (see below).10

Cognitive Dysfunction and Dementia

• The Clock Drawing Test (CDT). Commonly used for screening in AD, the CDT is sensitive for assessing visuospatial and executive abilities. However, interpretation may be limited by the presence of micrographia or tremor.

Table 2

It is of importance to distinguish PD from AD and DLB, as the symptom profile, course, and management of each disorder differs (). The presence of dementia and psychosis early in the disease course (ie, within the first year) is highly uncharacteristic in PD and favors a diagnosis of DLB.

Patient referral to a neuropsychologist should be considered if uncertainty arises regarding the presence of cognitive impairment at any point during the course of the disease.

Treatment. Rivastigmine was found to be moderately effective in PD dementia in a large placebocontrolled study16,38 and is US Food and Drug Administration approved for this indication. In small studies, the cholinesterase inhibitor donepezil has improved cognitive function in PD patients with dementia,16,39 and also in patients with dysexecutive syndrome in the absence of dementia.35 The response to rivastigmine and donepezil in these studies supports the contention that dysexecutive syndrome and dementia in PD are, at least in part, related to a cholinergic deficit. Rarely, treatment with cholinesterase inhibitors has caused a worsening of parkinsonism.10

Table 3

Doses, cost of therapy, and principal adverse effects of these 2 agents in patients with PD are shown .16,19,22,23,39 Although somewhat expensive, to the extent that successful treatment is associated with improved functionality and HRQOL, the cost may be offset by reductions in both indirect and direct healthcare costs by minimizing disability, medical resource needs, poor work productivity, and extra caregiver assistance.

A recent cost-effectiveness study40 was based on results of the rivastigmine trial mentioned above.38 Overall, no significant difference between rivastigmine and placebo groups was observed with respect to total costs. The clinical benefits derived by use of rivastigmine, and attendant reduction in caregiver costs, were offset by an increase in direct costs (mainly those related to drug acquisition). This precluded a conclusion regarding cost-effectiveness, despite evidence of improved patient care with rivastigmine. However, the time horizon was limited to 6 months, and the authors concluded that cost reductions secondary to improved clinical outcomes with rivastigmine would probably increase over a longer time frame.

Impulse Control Disorders

Neuropsychiatric symptoms and cognitive impairment in PD are associated with significant disability and poor HRQOL. In particular, depression and cognitive impairment may be more disabling to the patient than motor symptoms, and controlling motor symptoms is no longer the only consideration in patient management. Effective management of neuropsychiatric symptoms can potentially minimize disability, improve HRQOL, reduce caregiver burden, and reduce healthcare resource utilization/healthcare costs, including a reduced or delayed need for long-term care placement. Long-term cost-effectiveness analyses, incorporating an HRQOL component, are needed to confirm these speculations.

Deficiencies cited in the recognition and treatment of neuropsychiatric symptoms suggest the need for improvement in managing these comorbidities. The best available methods for screening and diagnosing neuropsychiatric features, and proper and cost-conscious use of the most effective treatment modalities, must be imparted to clinicians to enable their use in everyday practice. This can be greatly facilitated through educational programs sponsored by managed care, which can update clinicians on current treatment practices and guidelines to improve outcomes in PD.

2. Chaudhuri KR, Healy DG, Schapira AHV. Non-motor symptoms of Parkinson’s disease: diagnosis and management. Lancet Neurol. 2006;5:235-245.

4. Leibson CL, Long KH, Maraganore DM; National Institute for Clinical Excellence. Direct medical costs associated with Parkinson’s disease: a populationbased study. Mov Disord. 2006;21:1864-1871.

6. Huse DM, Schulman K, Orsini L, Castelli-Haley J, Kennedy S, Lenhart G. Burden of illness in Parkinson’s disease. Mov Disord. 2005;20:1449-1454.

8. Weintraub D, Moberg PJ, Duda JE, Katz IR, Stern MB. Effect of psychiatric and other nonmotor symptoms on disability in Parkinson’s disease. J Am Geriatr Soc. 2004;52:784-788.

10. Weintraub D, Stern MB. Psychiatric complications in Parkinson disease. Am J Geriatr Psychiatry. 2005;13:844-851.

12. Shulman LM,Taback RL, Rabinstein AA, Weiner WJ. Non-recognition of depression and other non-motor symptoms in Parkinson’s disease. Parkinsonism Relat Disord. 2002;8:193-197.

14. Reese SL. Psychosocial factors in Parkinson’s disease. Dis Mon. 2007;53:291-295.

16. Barbas NR. Cognitive, affective, and psychiatric features of Parkinson’s disease. Clin Geriatr Med. 2006;22:773-796.

18. Weintraub D, Comella CL, Horn S. Parkinson’s disease— Part 1: pathophysiology, symptoms, burden, diagnosis, and assessment. Am J Manag Care. 2008;14:S40-S48.

20. Weintraub D, Taraborelli D, Moreales KH, Duda JE, Katz IR, Stern MB. Escitalopram for major depression in Parkinson’s disease: an open-label, flexible-dosage study. J Neuropsychiatry Clin Neurosci. 2006;18:377-383.

22. Drugs for psychiatric disorders. Treat Guidel Med Lett. 2006;4:35-46.

24. Maia AF, Pinto AS, Barbosa ER, Menezes PR, Miguel EC. Obsessive-compulsive symptoms, obsessive-compulsive disorder, and related disorders in Parkinson’s disease. J Neuropsychiatry Clin Neurosci. 2003;15:371-374.

26. Korczyn AD. Management of sleep problems in Parkinson’s disease. J Neurol Sci. 2006;248:163-166.

28.Tan EK. Restless legs syndrome and Parkinson’s disease: is there an etiologic link? J Neurol. 2006;253:VII33-VII37.

30. Wolters ECh. PD-related psychosis: pathophysiology with therapeutical strategies. J Neural Transm Suppl. 2006;71:31-37.

32. Friedman JH, Factor SA. Atypical antipsychotics in the treatment of drug-induced psychosis in Parkinson’s disease. Mov Disord. 2000;15:201-211.

34. The Parkinson Study Group. Low-dose clozapine for the treatment of drug-induced psychosis in Parkinson’s disease. N Engl J Med. 1999;340:757-763.

36. Haugarvoll K, Aarsland D,Wentzel-Larsen T, Larsen JP. The influence of cerebrovascular risk factors on incident dementia in patients with Parkinson’s disease. Acta Neurol Scand. 2005;112:386-390.

38. Emre M, Aarsland D, Albanese A, et al. Rivastigmine for dementia associated with Parkinson’s disease. N Engl J Med. 2004;351:2509-2518.

40. Willan AR, Goeree R, Pullenayegum EM, McBurney C, Blackhouse G. Economic evaluation of rivastigmine in patients with Parkinson’s disease dementia. Pharmacoeconomics. 2006;24:93-106.

42. Weintraub D, Siderowf AD, Potenza MN, et al. Association of dopamine agonist use with impulse control disorders in Parkinson disease. Arch Neurol. 2006;63:969-973.