Patients with Parkinson's disease (PD), particularly those exposed to dopamimetic agents such as levodopa, have a high incidence of psychiatric symptoms. Common symptoms are hallucinations,
+13,+14 delusions,
+14—+16 depression,
+16,+17 euphoria,
+16 anxiety,
+16,+17 agitated confusion,
+16 sleep disturbances,
+17,+18 and delirium.
+18
Visual hallucinations are the most common psychotic symptom exhibited by PD patients. Typical hallucinations have been described as nocturnal, nonthreatening images of people or animals, which the patient recognizes, or comes to recognize, as imaginary.
+14 In his 1991 review, Cummings
+19 reported prevalence rates for visual hallucinations in PD patients between 6% and 38%, with most studies reporting about 30%.
Delusions are less common in PD patients than hallucinations. Cummings
+19 reported most prevalence rates between 3% and 17% for delusions in his review of the literature. Moskowitz et al.
+14 reported 8 patients with nonconfusional paranoid delusions. The authors noted that all delusional patients experienced vivid dreams or hallucinations either previous to or concurrent with their delusional state. Cummings
+20 noted a correlation between higher doses of antiparkinsonian medications and the prevalence of delusions.
Both hallucinations and delusions are serious complications of PD. Goetz and Stebbins
+21 found psychotic symptoms to be a greater risk factor for nursing home placement in PD patients than memory problems or even severe motor disabilities. Risk factors for psychosis in PD patients include advancing age,
+22 increased dosage of levodopa,
+23 long-term treatment with levodopa,
+14 anticholinergic therapy, multiple drug therapy,
+13 dementia,
+15 cerebral atrophy,
+23 and preexisting psychiatric conditions.
+18
Here we present 3 cases of delusional misidentification in patients with Parkinson's disease and dementia.
Patient 1 is a 65-year-old married woman diagnosed with Parkinson's disease 3 years ago. She received treatment for depression and anxiety for 2 years prior to the development of her neurological symptoms. On presentation to us, the patient had a mild dementia characterized by generally intact orientation, attention, and fund of knowledge, with impaired short-term memory (1/3 objects after 5 minutes) and visuospatial performance. She complained of visual hallucinations in addition to continued depression and anxiety. Her medications included levodopa and lorazepam. The patient responded initially to a combination of bupropion 75 mg per day and clonazepam 0.5 mg tid. The visual hallucinations, consisting of formed, fleeting images of people, which the patient recognized as imaginary, were considered benign and were not treated. The patient's anxiety decreased over time, and clonazepam was reduced to 0.5 mg bid.
Seven months later, when the family reported that the hallucinations had become severe, she was admitted to neurology for a trial of clozapine. Psychiatric assessment revealed that the patient believed her husband had been replaced by a stranger and that she was in a house that did not belong to her. Occasionally she attempted to throw her husband out of the house, or she packed her own bags and threatened to leave. Her cognitive status had deteriorated significantly. She was disoriented to the month and year. She could no longer reverse the word world or name the President. The patient responded, within 1 week, to a trial of clozapine, 12.5 mg per day, with a significant diminution of her delusional misidentifications.
Over the course of the next year the patient sporadically accused her husband of inauthenticity, and her neurologist increased her clozapine to 18.75 mg per day. The patient's cognitive impairment progressed, as evidenced by a Mini-Mental State Examination score of 15/30. Her clozapine was decreased to 12.5 mg per day because of her husband's report that she was oversedated. On subsequent follow-up her symptoms had worsened, with periods of misidentification of her husband occurring on a daily basis. In the office she referred to her husband as "a friend I'd like to get rid of." Clozapine was increased to 18.75 mg per day.
Patient 2 is a 75-year-old married man with a 9-year history of Parkinson's disease characterized by bradykinesia, gait disturbance, and the absence of a resting tremor. Prior to coming to our institution, about 3 years ago, he developed visual hallucinations and had been treated with thioridazine, which had resulted in a worsening of his parkinsonism. His medication at that time included levodopa, amantadine, deprenyl, and bromocriptine.
At our institution, the neurologist changed his medication to sustained-release levodopa and pergolide. He continued to have psychotic symptoms, including visual hallucinations of men in his house and the related delusional belief that his wife was unfaithful. Over the next 2½ years the patient's psychotic symptoms gradually grew more intense, preventing the neurologist from increasing the parkinsonian medication enough to control an increasing gait disturbance. As a result, the patient was admitted to neurology for a trial of clozapine.
On psychiatric evaluation the patient continued to report that he was seeing other men in the house who were attempting to seduce his wife. He accused her of "hiking up her skirt" in order to attract them. Additionally, the patient was convinced that another woman, similar in appearance to his wife, was entering the house to cook and clean. He never actually misidentified the wife, instead claiming that the other woman appeared when his wife was out on errands. On one occasion, when he and his wife were going to Atlantic City, he insisted on buying three bus tickets because "it wouldn't be fair not to take her (the other woman) along." When specifically asked if the other woman resembled his wife, he stated, "she's no Marilyn Monroe." The patient had a mild dementia, as evidenced by a score of 22/30 on the Mini-Mental State Examination, with mild deficits in orientation, short-term memory, and concentration.
The patient tolerated clozapine, 12.5 mg qhs, for several days and was discharged. According to his wife, his hallucinations and delusions remitted. However, despite a reduction of his clozapine dosage to 6.25 mg qhs, the family elected to discontinue clozapine because of excess daytime sedation. One month later, his psychotic symptoms had returned. In addition to visual hallucinations and the delusion about the other woman, he now claimed to have another house where his possessions had been moved.
Patient 3 is a 66-year-old former businessman with a 6-year history of Parkinson's disease characterized by tremor, bradykinesia, rigidity, and associated orthostatic hypotension. In the last year the patient developed visual hallucinations. His levodopa was reduced, and he was placed on thioridazine. Subsequently, he fell, fractured his hip, and was hospitalized for surgery. In the hospital his hallucinations persisted despite treatment with thioridazine. He was transferred to our hospital for management of his Parkinson's disease.
At the time of transfer the patient was receiving levodopa 350 mg qid, thioridazine 25 mg, and fludrocortisone acetate 0.1 mg bid for orthostatic hypotension. He was unable to stand independently, and he was reported to be disorganized. On the day of admission, his thioridazine was discontinued and he was placed on a reduced dose of sustained-release levodopa. On psychiatric assessment, 2 days following admission, the patient claimed that he was currently located, simultaneously, both at his home in Doylestown, Pennsylvania, and at Beth Israel Hospital in New York City. The patient was correctly oriented to year, month, and date and was able to reverse the spelling of world.
Neuropsychological testing demonstrated that although his verbal ability was in the low average range, his visuospatial ability was in the severely impaired range. For example, he was unable to assemble puzzle pieces (scale score of 2; the norm is 10 with a standard deviation of 3). His verbal memory was moderately impaired, with inability to learn new verbal information over repeated trials. Visuospatial memory was severely deficient. On tasks of abstraction and executive functions the patient's performance was also severely impaired. On the Trail Making Test the patient was completely unable to perform Trails B. Similarly, on a test of verbal fluency (FAS) he generated only 16 words in 3 minutes, a very impaired performance. A head CT revealed mild cortical atrophy.
On psychiatric follow-up, the patient provided a number of different responses regarding spatial orientation: Bedtownship Hospital in the incorporated village of Bedminster; Beth Chanel (a fashion house for distributing shoes); and the kitchen at B'nai Brith Medical.
In addition to disorientation, the patient frequently reported visual hallucinations of wires emanating from the walls, on one occasion stating that they were "involved with Swiss bondkeeping." He also voiced paranoid delusions regarding hospital staff.
The patient tolerated an initial dose of clozapine 12.5 mg qhs without an improvement in his psychiatric symptoms. He was discharged on clozapine 12.5 mg qod alternating with 25.0 mg qod. On a follow-up visit with his neurologist, his visual hallucinations, paranoid delusions, and misidentification of place had all improved significantly. A subsequent attempt to lower the dose of clozapine resulted in a return of his psychotic symptoms.
There have been scattered reports of DMS symptomatology in association with Parkinson's disease in the literature. Lipper
+24 described a patient who developed psychotic symptoms, including Capgras syndrome, after treatment with levodopa, trihexyphenidyl, and bromocriptine. The specific nature of the Capgras delusion was not specified. Chacko et al.,
+25 in a study evaluating clozapine as an antipsychotic in 12 Parkinson's patients, indicated that 1 patient believed that his spouse was an impostor and 1 patient rejected his house. Factor et al.
+26 referred to one of their own Parkinson's patients, an artist, who claimed that his paintings had been substituted with reproductions. This last case is an example of the delusion of inanimate doubles previously described by Anderson and Williams.
+27
In our series, Patient 1 had a classic Capgras delusion, in which she believed her husband was an impostor. The second patient reported the existence of two nearly identical caretakers but did not believe the caregiver had been replaced. Thus, his delusion would be classified as a reduplication for person or for role, rather than Capgras syndrome. Patient 3 could be considered to have a variation of reduplication for place, since he claimed to be in two places at the same time.
In all 3 patients, visual hallucinations preceded or accompanied the onset of the DMS. This is consistent with the findings of Moskovitz et al.,
+14 who proposed a mechanism of dopaminergic kindling that produces a progression of psychotic symptoms, leading from vivid dreams to hallucinations to delusions. The fixed nature and the degree of specificity of the misidentification in all cases confirm that this symptom is best viewed as a primary delusion that cannot be explained solely on the basis of cognitive impairment.
All of our cases of DMS involved dementia. In Patient 1 the development of Capgras was associated with a significant cognitive decline, whereas visual hallucinations had occurred when cognitive impairment was relatively mild.
We have recently reviewed the neuroanatomical correlates and proposed mechanisms of the delusional misidentification syndromes.
+28 In cases with focal lesions, DMS occurs more frequently in association with right hemisphere rather than left hemisphere damage.
+3,+29 However, diffuse and, in particular, bifrontal dysfunction (often in conjunction with right parietal damage) appears to be present in most cases of DMS associated with neurological disease.
+6,+29,+30 Alexander et al.
+5 hypothesized that right hemisphere damage causes difficulty in recognition or identification, particularly when memory is impaired. When such deficits occur in the context of frontal lobe deficits, judgment is diminished and reduplication or delusional misidentification may result.
Right hemisphere impairment is not specifically characteristic of PD. However, frontal lobe dysfunction has been well documented in PD patients,
+17,+31—+33 particularly those with dementia.
+34 This frontal lobe impairment presumably results from decreased function of the basal ganglia and particularly of the caudate nucleus, which provides a major input for the prefrontal cortex.
+32,+33 Taylor et al.
+32 have shown that frontal lobe decline correlates with severity of motor symptoms in PD and decreased response to treatment.
+33 Although excess dopaminergic stimulation of the mesolimbic system may account for visual hallucinations and paranoia in PD patients treated with levodopa,
+14,+35 this phenomenon may not be sufficient to cause delusional misidentification. Frontal lobe impairment, which produces a disturbance in judgment, self-awareness,
+36 and relatedness to elements in the environment such as persons, places, and objects,
+37 may also be required. A study using PET demonstrated significantly greater bilateral orbitofrontal hypometabolism in patients with Alzheimer's disease and DMS than in patients with Alzheimer's without DMS.
+38
Although it has been emphasized that depersonalization/derealization contributes significantly to DMS,
+1 Christodoulou
+1 has shown that other independent factors, such as paranoia, are required. Feinberg and Roane
+37 have noted that DMS and related phenomena associated with depersonalization/derealization are typically seen in patients with bilateral caudal orbital frontal damage. The hypothesis that DMS associated with PD results from a combination of frontal lobe dysfunction and levodopa psychosis is consistent with these previous findings. Our hypothesis would further predict that DMS is common in PD patients with advanced disease.
With regard to treatment, we found clozapine, an antipsychotic with specificity for mesolimbic DA receptors,
+39 to be generally effective and well tolerated in our Parkinson's patients, confirming previous findings.
+25 In our cases, attempts to taper or discontinue clozapine resulted in a return or increase of the DMS, indicating that in PD, DMS can become a chronic condition.
A number of previous authors have stressed that neurological and psychological factors combine to create DMS.
+1,+40 We view DMS in PD as resulting from a combination of psychotic and cognitive disturbances, both with a clear organic basis. However, in many cases of DMS in the literature, including our cases, the misidentified or reduplicated entity is highly personally significant to the patient, suggesting that psychological factors can influence the form that DMS takes.