An Update on Parkinson’s Disease and Its Psychiatric Complications

At an industry-supported symposium held at the American Association for Geriatric Psychiatry’s 2003 Annual Meeting, experts in the field of psychiatry reviewed various psychiatric complications of Parkinson’s disease, including depression, cognitive impairment, and psychosis. The symposium included an update on Parkinson’s disease and an overview of the occurrence of sleep disorders and anxiety in patients with the disease.

This program was supported by an unrestricted educational grant from AstraZeneca.

Depression in Parkinson’s Disease

According to population-based studies, the prevalence of depression in patients with Parkinson’s disease (PD) ranges from 2.7% to 7.7% (Tandberg, et al. Arch Neurol. 1996;53:175; Hantz, et al. Am J Psychiatry. 1994;151:1010), but can average up to 40% among PD patients visiting outpatient neurologic clinics (Cummings. Am J Psychiatry. 1992; 149:443). Treating this depression is important, because it is associated with impaired cognitive performance, increased functional disability, and decreased quality of life, remarked Albert F.G. Leentjens, MD, PhD, Assistant Professor of Psychiatry at the Memory and Neuropsychiatric Clinic of Maastricht University Hospital, in Maastricht, the Netherlands. Dr. Leentjens’ presentation therefore focused on two clinically relevant questions: how do we diagnose depression in PD patients and what are the treatment options?

Diagnosis of Depression in PD Patients
Diagnosis of depression in patients with PD depends on the specificity and sensitivity of individual depressive symptoms. Substantial data indicate that no specific symptom profile exists for depression in PD patients, although these patients may experience greater anxiety (Menza, et al. Biol Psychiatry. 1993;34:465), dysphoria, and irritability (Brown, et al. Psychol Med. 1988;18:49). While not adequately studied, the sensitivity of depressive symptoms may be more clinically important, because it considers how a depressed PD patient may differ from a nondepressed patient with the disease.

Discriminant Model for Sensitivity of Depression in PD Patients
To rank symptoms according to their sensitivity for depression, Leentjens et al. (J Neuropsychiatry Clin Neurosci. 2003; 15(1):74) constructed a discriminant model in depressed (25%) and nondepressed PD patients. The model was based on the individual scores of the Montgomery-Åsberg Depression Rating Scale (MADRS) and the Hamilton Depression Scale (HAM-D). Core and nonsomatic symptoms were the most sensitive in establishing a diagnosis of major depressive order. The most sen-sitive physical symptoms of depression were early morning wakening (HAM-D only) and reduced appetite (MADRS and HAM-D). Importantly, the only observational item in the MADRS that was rated exclusively by clinicians also was the most sensitive. Dr. Leentjens believes this to be an evidenced-based indicator of the importance of clinical experience.

Pharmacologic Treatment Options for Depression in PD Patients
The options for treating depression in PD patients are limited, Dr. Leentjens reported. According to several placebo-controlled, randomized, double-blind trials of citalopram (Wermuth. Eur J Neurol. 1998;5:235), nortriptyline (Andersen, et al. Acta Neurol Scand. 1980;62:210), and sertraline (Leentjens, et al. Int J Geriatr Psychiatr. 2003;18:1), approximately 70% of PD patients showed improvement in depressive symptoms at 6 to 8 weeks. However, high placebo response rates reduced the differences between the antidepressants and placebo. Thus, Dr. Leentjens questions whether antidepressants are indeed more effective than placebo in depressed PD patients.

Moreover, he pointed out, antidepressant agents are associated with significant side effects. Selective serotonin reuptake inhibitors (SSRIs) may exacerbate motor symptoms, although the American Psychiatric Association (APA) does not consider this to be a major clinical concern, according to Dr. Leentjens. Tricyclic antidepressants (TCAs) may be associated with cognitive impairment, and co-administration of TCAs with selegilene can, in rare cases, cause serotonin syndrome. According to the 2000 guidelines of the APA, no differences exist among antidepressants in the treatment of depression in PD patients, although clinical experience favors SSRIs over TCAs, Dr. Leentjens stated.

Thus, the efficacy of SSRIs and TCAs in the treatment of depression in PD has not been firmly established, according to Dr. Leentjens, but in the absence of better alternatives, the use of these agents in PD is justified.

Cognitive Impairment, Dementia, and Parkinson’s Disease

Cognitive impairment is characterized primarily by the clinical phenomena that are observed, remarked Constantine G. Lyketsos, MD, MHS, Professor of Psychiatry and Behavioral Sciences, Johns Hopkins University School of Medicine, Co-Director of the Division of Geriatric Psychiatry and Neuropsychiatry, and Director of the Johns Hopkins Comprehensive Alzheimer Program, at The Johns Hopkins Medical Institutions, in Baltimore, Maryland. According to DSM-IV, mild cognitive impairment (MCI) is not associated with functional impairment, whereas dementia involves global cognitive decline with functional impact. Parkinson’s dementia is recognized in DSM-IV as a specific dementia syndrome that is not well characterized but is hypothesized to be related to specific pathology in the brain.

Aspects of Cognition
Patients with impaired cortical cognition experience amnesia, aphasia, apraxia, and agnosia. Patients with Parkinson’s disease (PD) tend to exhibit more impairment of frontal, subcortical cognition, which results in dysmnesia; a delay or slowness of thinking; dysexecution (poorly coordinated thinking); and a broader depletion of mental life. MCI, which may be as prevalent as frank dementia in PD, correlates with motor slowing and typically occurs in the early stages of PD at a rate disproportionate to the effects of aging. MCI in PD patients involves disturbances that occur independently in several key domains, such as problem solving, working memory associated with attention, or the ability to participate in a conversation (Dubois, et al. J Neurol. 1996;244:2). Coordination of visuospatial function also is impaired.

Potential Pathways for MCI and Dementia in PD Patients
An estimated 60% of PD patients experience MCI or dementia, with MCI predominant in the early stages of the disease and dementia occurring in the later stages (Mindham, et al. Adv Neurol. 1993;60:470). Noting that the pathophysiology of PD is not fully understood, Dr. Lyketsos outlined three possible pathways involved in the cognitive changes.

Gradual deterioration of the substantia nigra may be central to some aspects of frontal subcortical cognitive changes. Dementia with Lewy bodies (DLB) may be involved, especially if dementia occurs early in suspected PD. Finally, emerging pathologic findings indicate that up to 30% of these patients exhibit Alzheimer’s type manifestations of dementia. This raises intriguing questions: could PD, if present long enough, possibly initiate the Alzheimer’s process? Could the later stages of Parkinson’s dementia be related to Alzheimer’s disease (AD), and would the medications for treatment of AD be helpful in treating the cognitive changes associated with PD? Two studies have examined the efficacy of cholinesterase inhibitors, the only medications approved for AD treatment, in patients with PD.

Use of Cholinesterase Inhibitors to Treat Cognitive Changes in PD
The use of flexible dosing of donepezil 2.5–10 mg/d was studied in PD patients who had dementia or cognitive impairment (Leroi, et al. Johns Hopkins University (JHU) Trial. Under review). The outcomes measured comprised a detailed cognitive battery (including measurement of subcortical changes), as well as motor symptoms (assessed by Unified Parkinson’s Disease Ratings Scale), and clinical tolerability. Initial results indicate that, compared with placebo, donepezil significantly improves memory scores (Mattis Dementia Rating Scale), and exhibits a trend for significantly improving Trails A scores. No worsening of motor disorders was observed. Similar results were obtained by Aarsland et al. (J Neurol Neurosurg Psychiatry. 2002;72:708).

Tolerability was variable in the JHU trial, with 5 of 7 donepezil-treated patients withdrawing early from the study, compared with 1 of 9 patients in the placebo group. However, the side effects of donepezil were readily reversible. Thus, donepezil may improve memory symptoms without worsening motor symptoms, but its use requires careful monitoring.

Dr. Lyketsos stated that the cholinesterase inhibitors rivastigimine and galantamine have not been studied in controlled trials of PD, but rivastigmine is effective in DLB and preliminary data from an open-label trial indicate that it may be effective in Parkinson’s dementia (Korczyn, et al. J Clin Psych. 2002;63:259).

Psychosis in Parkinson’s Disease

Laura Marsh, MD, Associate Professor in the Department of Psy-chiatry and Behavioral Sciences at Johns Hopkins University School of Medicine in Baltimore, Maryland, observed that geriatric psychiatrists often treat the more complex patients with Parkinson’s disease (PD) who have psychosis. These patients typically exhibit associated behavioral disturbances or require nursing home placement. Psychosis in these patients is the primary factor driving their placement in nursing homes and is associated with increased mortality, Dr. Marsh pointed out.

Prevalence of Psychosis in PD
Psychosis has been reported in 8%–40% of PD patients, with 5%–17% in people without significant dementia, and up to 80% in individuals with significant dementia (Greene, et al. Adv Neurology. 1993; 60:703). To the extent that most PD patients experience cognitive deficits that cause functional impairment, many patients with PD can be considered to have a dementia, stated Dr. Marsh. Patients with psychosis may experience hallucinations with or without insight, delusions, or hallucinations, and/or delusions with delirium.

Prevalence of Hallucinations and Delusions
Approximately 40% of PD patients experience minor, complex visual, and/or auditory hallucinations, with an overall lifetime rate of nearly 50% (Fenelon, et al. Brain. 2000;123:733). The prevalence of delusions in PD patients varies from 3%–30% and can be a feature of affective psychoses, often accompanying hallucinations. Patients also may experience persecutory delusions or first-rank symptoms that resemble those of schizophrenia patients.

Risk Factors for Psychosis in PD Patients
Although no single explanation exists for the cause of psychosis in PD patients, it is most commonly related to the use of dopaminergic medications. However, Dr. Marsh believes that the role of nondopaminergic factors is underappreciated. Psychosis in PD reportedly is related to both extrinsic factors (e.g., medications and/or other medical illnesses) and intrinsic factors (Wolters. J Neurol. 2001;248:22). The latter include the primary dopaminergic loss and treatment-induced dopamine hypersensitivity. Degeneration of other neurotransmitter systems, as well as age-related cortical cholinergic loss, also may be important.

Psychiatric Comorbidities in Psychosis Related to PD
One of the extrinsic risk factors for psychosis is psychiatric comorbidity. In a longitudinal study of PD patients, 75% had at least one psychiatric diagnosis and 24% of the total sample had psychotic symptoms (Marsh, et al. Mov Disord. 2002,17(suppl 5):S227). The patients with psychosis often experienced both hallucinations and delusions, and more than 50% had an additional psychiatric condition. Thus, the presence of one nonmotor symptom increases the likelihood of other nonmotor symptoms, which further complicates the management of PD. Moreover, some psychiatric features, e.g., affective lability, may be evident only when the patient’s motor functioning is abnormal. Evaluation of these patients varies, therefore, according to the patient’s antiparkinson dosing regimen and related motor fluctuations.

Treatment Options for Psychosis in PD
The treatment of psychiatric disorders should be balanced against the management of motor function, explained Dr. Marsh. Extrinsic factors (e.g., other medical illnesses) should be treated first and the use of psychoactive drugs (e.g., benzodiazepines) should be eliminated. However, comorbid psychiatric illnesses should still be treated. Dr. Marsh recommends that nonpharmacologic strategies, such as patient education and reassurance, should be used to treat psychosis, with adjustment or elimination of antiparkinson medications, as appropriate.

Although antipsychotic medications may enable dosage increases of antiparkinson medications, conventional antipsychotics can increase parkinsonism. The atypical antipsychotics, which are effective and tolerated at very low doses, generally do not have this problem, but are associated with sedation and orthostasis.

Other strategies for treatment of psychosis include ondansetron (a 5-HT3 inhibitor), and electroconvulsive therapy. Cholinesterase inhibitors can be effective in treating cognitive impairment in PD patients, but the doses require careful monitoring because of variable tolerance. In terms of prevention strategies, Dr. Marsh recommends that the PD regimen be evaluated for appropriate dosing, and mood and sleep disorders should be addressed early.

Sleep, Fatigue, and Anxiety in Patients with Parkinson’s Disease

Patients with Parkinson’s disease (PD) often experience sleep disorders, fatigue, and anxiety, reported Matthew A. Menza, MD, Vice Chairman of the Department of Psychiatry, Chief of the Division of Clinical Psychopharmacology, and Associate Professor of Psychiatry and Neurology at Robert Wood Johnson Medical School in Piscataway, New Jersey. Sleep disorders are characterized by frequent nighttime wakening and daytime naps, and excessive daytime sleepiness (Menza, et al. Psychosomatics. 1995;36:262), whereas fatigue implies a lack of energy. In PD patients, anxiety is closely interlinked with depression: most individuals exhibiting anxiety also have a comorbid depressive disorder (92%) and vice versa (67%) (Menza, et al. Biol Psychiatry. 1993;34:465).

Common Sleep Disorders in PD
Restless legs syndrome in PD patients and, contradictorily, nocturnal muscular rigidity contribute to the difficulty in falling asleep or staying asleep. In addition, the incidence of sleep apnea increases. Excessive daytime sleepiness (EDS), experienced by up to 51% of PD patients, worsens as the disease progresses. Furthermore, dopaminergic agents can induce both sedation and arousal (Hobson, et al. JAMA. 2002; 287:455).

Among PD patients, 3.8% report experiencing sleep attacks (sudden onset of sleep) while driving, with 0.7% stating that they had no warning (Hobson, et al.) Whether sleep attacks may be attributed to dopaminergic drugs or to EDS is not clear, but Dr. Menza advises clinicians to question their patients about these dangerous situations.

Patients also should be questioned about rapid eye movement (REM) behavior disorder, in which skeletal muscles remain active during REM sleep. Thus, patients “act out their dreams.” PD accounts for 27% of REM behavior disorder, with 32% of patients injuring themselves and 64% assaulting their spouses (Stacy. Drugs Aging. 2002; 19:733).

Understanding and Treating Insomnia in PD Patients
Sleep is regulated by a complex interaction of neurotransmitters, many of which are variably affected by PD. Comorbid depression or anxiety, in addition to the normal changes of aging, also contribute to sleep disturbances (Larsen, et al. CNS Drugs. 2001;15:267). Likewise, dopaminergic drugs, which may relieve nighttime rigidity, may disturb the patient’s sleep by acting as stimulators. In early PD, doses of levodopa are stronger predictors of sleep disturbances than is disease severity. In the later stages of the disease, however, movement disorder plays a greater role (Van Hilten, et al. Arch Neurol. 1994; 51:922).

Effective strategies for treating insomnia include the intermittent use of non-benzodiazepine hypnotics; sleep hygiene (e.g., regular sleep hours, physical activity); cognitive behavioral therapy; and stimulus control interventions. Stimulants, such as modafinil or methylphenidate, may be of use but have not been studied in PD.

Fatigue in PD
Up to 40% of PD patients experience significant fatigue, which develops early and is not correlated with disease severity (Shulman, et al. Mov Disord. 2001; 16:507). Fatigue is clearly related to depression and probably to the use of medications, such as dopaminergics and antidepressants. Treatment of fatigue requires a multimodal approach, including cognitive behavioral therapy, the use of stimulants, and mild exercise to combat deconditioning.

Anxiety in PD
Anxiety, which often is present before the onset of PD (Shiba, et al. Mov Disorder. 2000;15:669), has been diagnosed in 65% of PD patients (Menza, et al. Biol Psychiatry. 1993;34:465). Importantly, in the subset of patients who have fluctuations in motor symptoms (“on/off” syndrome), symptoms of anxiety and depression are significantly
(P < 0.001, P < 0.005, respectively) worse when the movement disorder is not well controlled (Figure 1) (Menza, et al. Mov Disorder. 1990;5:148).

Because the management of anxiety in PD has not been studied, anxiety generally is treated as an aspect of depression. Effective treatment of depression has been associated with a significant (P < 0.001) decrease in anxiety (Menza, et al. Neuropsychiatry Clin Neurosci. In press). Other treatment options that may be effective in select patients include deep brain stimulation (Martinez-Martin, et al. Mov Disorder. 2002;17:372) and surgical intervention (Higginson, et al. Neuropsychiatry Neuropsychol Behav Neurol. 2001;14:117).