Authors: Jong S. Kim, Seung-Bong Hong, Keun-Woo Park, Allen T.C. Lee
Categories: Review, Parkinson’s disease, dementia, epilepsy, neurological diseases, psychosis, stroke
Source: Journal of Clinical Neurology (Seoul, Korea)
Neurological diseases often manifest with neuropsychiatric symptoms such as depression, emotional incontinence, anger, apathy and fatigue. In addition, affected patients may also experience psychotic symptoms such as hallucinations and delusions. Various factors contribute to the development of psychotic symptoms, and the mechanisms of psychosis are similar, but still differ among various neurological diseases. Although psychotic symptoms are uncommon, and have been less well investigated, they may annoy patients and their families as well as impair the patients’ quality of life and increase the caregiver burden. Therefore, we need to appropriately identify and treat these psychotic symptoms in patients with neurological diseases.
Keywords: psychosis, neurological diseases, stroke, epilepsy, Parkinson’s disease, dementia
Neuropsychiatric symptoms such as depression, emotional incontinence, anger, and fatigue manifest in more than one-third of neurological diseases.^1^ In addition to these major neuropsychiatric disturbances, patients occasionally show psychotic symptoms such as hallucinations and/or delusions after the occurrence of their neurological diseases such as stroke, epilepsy, dementia, and Parkinson’s disease (PD). Various factors including the lesion location influence the development of these symptoms. These psychotic symptoms annoy patients and their families and are associated with poor recovery, decreased quality of life, and increased mortality, as well as increased caregiver burden. This situation indicates the need to appropriately identify and treat these symptoms. Neurologists often neglect these symptoms, resulting in misdiagnoses and poor treatments. Moreover, the pathogenesis of psychotic symptoms associated with neurological diseases is poorly understood. This narrative review discusses the psychotic symptoms that can occur in major neurological diseases such as stroke, multiple sclerosis (MS), traumatic brain injury (TBI), epilepsy, PD, and dementia. Due to space limitations, other diseases including encephalitis and brain tumors are not covered.
In patients with stroke, psychotic symptoms can manifest as part of complex behavioral syndromes such as delirium or dementia. They can also be caused by associated medical comorbidities, admission to an intensive-care unit, or a side effect of certain drugs. However, the symptoms may occur in isolation for longer periods.
Patients with stroke usually have focal damage in the brain. When a lesion damages the sensory tract, the dysfunctional sensory input—possibly with abnormal neuronal feedback—may produce hallucinations. Unlike in patients with schizophrenia, the hallucinations are usually simple in these cases, such as light flashes or single sounds. However, as illustrated below, the symptoms may be more complex and manifest in a variety of forms such as hyperacusis, palinacousis, palinopsia, or even more-complex ones such as musical hallucinations. In such cases the symptoms may mimic psychosis.
One of the authors observed a 51-year-old female who experienced the sudden onset of perceiving environmental sounds as being louder in her left ear.^2^ Simultaneously a mechanical magnetic resonance imaging (MRI)-noise-like sound was heard in that ear. At the time of admission she slept in the right lateral decubitus position because tactile auricle stimulation aggravated her hyperacusis. Brain MRI revealed an intracerebral hemorrhage (ICH) in the right dorsal pons. She did not have other psychotic symptoms such as delusions. The hyperacusis evident in this patient may have been a hypersensitization phenomenon after partial injury to the central auditory pathway.
We also saw a female patient with ICH in the left middle temporal gyrus.^3^ The superior temporal gyrus, including Heschl’s gyrus, was spared. She intermittently heard “pee-pee,” “tog-tog,” and honey-bee sounds. She also occasionally heard voices repeatedly when the sound was no longer present. For example, when a doctor said, “Your symptoms are much improved,” she heard “much improved” repeatedly afterward (palinacousis). This symptom may be related to a compensatory hyperexcitation of surrounding auditory pathways during the recovery phase of ICH.
Simple auditory hallucinations such as tinnitus may be associated with partial disturbances of the auditory system. It is particularly interesting that complex hallucinations such as musical hallucinations may occur after hearing loss.^4^ Musical hallucinations also occur after tegmental pontine stroke involving the auditory pathways.^5^ The contents of musical hallucinations in these patients are diverse, but most often consist of simple old pop songs or choir music.
Likewise, such complex hallucinations may occur when visual pathways are damaged. One example are so-called peduncular hallucinations occurring after stroke involving the upper brainstem or thalamus.^6^ These patients typically experience visual hallucinations at the end of the day. They see animals or human beings for seconds or minutes, which may move around. The hallucinations are vivid and often colorful. The patients usually realize that they are not real. Although some may believe that the hallucinations are real when they occur, they usually subsequently realize that they were not real. These peduncular hallucinations have been attributed to damage to the pulvinar that contains visual fibers traveling to the parieto-occipital area.
Thus, hallucinations due to damage to the sensory pathways of the brain are usually simple but may also be complex. However, these patients do not experience significant distortion in thought content, which differentiates the symptoms from true psychosis. To diagnose psychosis in patients with stroke, researchers have investigated hallucinations using the Neuropsychiatric Inventory (which includes two sections on the frequency and severity of delusions and hallucinations), the Positive and Negative Symptom Scale, and the Brief Psychotic Rating Scale. However, there is no evidence that these tools are appropriate for diagnosing psychosis in patients with stroke.
Psychosis mimicking schizophrenia occurring after stroke is rare. In a systematic review,^7^ psychosis defined as having either delusions or hallucinations with poor insight was found in 4.9% of patients with stroke. The most common type of psychotic symptoms was delusions, followed by schizophrenic-spectrum disorders and then mood disorders with psychotic features. However, it should be noted that the data were based on only four studies, only three of which applied structured symptom assessments. According to that review, although psychosis-like symptoms can be seen in the early stage of stroke, this is often associated with delirium or confusion. Isolated psychotic symptoms may appear sometime after the onset of stroke. Almeida and Xiao^8^ found that the mean time to the onset of psychosis was 6.1 months after the incident stroke.
Reportedly 0.4%–7.4% of MS patients present with psychotic symptoms, with the prevalence being about threefold higher than that in the general population.^9^ The symptoms include hallucinations, delusions (often paranoid), irritability/agitation, sleep disturbances, grandiosity, and blunted affect. The pathophysiological mechanism underlying psychosis in MS is poorly understood. MRI studies have shown that psychosis in MS is associated with lesions in the frontotemporal area, medial temporal lobe, and periventricular white matter.^10^,^11^ Visual hallucinations have most often been reported in association with temporal-lobe lesions.^10^ Psychosis may be induced by MS medications such as corticosteroids and IFN-ß.
Previous studies^12^ have found schizophrenia-like psychosis in patients with TBI, with the main symptoms including paranoid delusions and auditory/visual hallucinations. One meta-analysis^13^ found a significantly increased incidence of psychosis after TBI relative to no TBI (odds ratio=1.65, 95% confidence interval=1.17–2.32). Those authors also found no relationship between the severity of the head injury and the subsequent risk of psychosis. It was suggested that a genetic predisposition exists in patients who develop psychosis after TBI.
There have been no well-designed studies on the prevalence of psychosis following TBI. One study^14^ examined 40 patients with severe blunt head trauma at 10–15 years after the onset, and found that 8 patients (20%) had developed psychosis, with 5 having frontal- and/or temporal-lobe damage. This suggests that post-TBI psychosis is more common than previously realized. However, this prevalence rate is not generalizable because that study only included patients with very-severe blunt head injury, and also the diagnostic criteria for psychosis were not strictly defined.
The difficulty in assessing the prevalence of post-TBI psychosis is partly attributable to the wide range of time intervals between the onset of TBI and the occurrence of psychosis. Psychosis following TBI generally occurs after a latency of 1–3 years,^15^ which makes it difficult to confirm the causal relationship when psychosis develops in a specific patient who has experienced TBI. The presence of latency suggests that the pathogenesis of psychosis following TBI does not depend on primary injuries but rather on post-TBI changes after the resolution of primary injuries. This issue will be discussed below.
Some studies have investigated the rates of past TBI in patients who were diagnosed as having schizophrenic-spectrum disorders. One study^16^ found that childhood head injuries occurred in 24% of 67 patients with schizophrenia or schizoaffective disorder versus in 12% of 102 unaffected siblings. Moreover, childhood head injuries were associated with a lower age at the onset of the median time between TBI and the first occurrence of psychosis was 12 years. Another study^17^ found that patients with schizophrenia had a threefold higher risk of a prior TBI compared with relatives without psychiatric illness. A previous congenital neurological disorder, a head injury prior to adolescence,^18^ and a positive family history of psychosis^12^ were reported to be risk factors for post-TBI psychosis.
Psychotic symptoms have traditionally been classified according to their temporal relationship with seizure events, as ictal, postictal, and interictal.^19^ A systematic review^20^ found that the pooled estimate of the prevalence of psychosis in patients with epilepsy (PWE) was 5.6%. The prevalence of psychosis in temporal-lobe epilepsy was 7.0%, including 5.2% for interictal psychosis and 2.0% for postictal psychosis.^20^
By definition, ictal psychosis refers to psychosis that is concurrently associated with epileptic discharges in the brain, and scalp electoencephalography (EEG) abnormalities are usually detectable during the symptom.^21^ Most discharges have a focus in the temporal lobe. The psychotic symptoms are typically brief, usually lasting from hours to days. The most common symptoms are visual or auditory illusions often combined with affective changes such as agitation, fear, or paranoia. Other associated psychic phenomena include depersonalization, derealization, and autoscopy.^19^ Insight is usually maintained, and strictly speaking, the manifestation is not that of a true psychosis. Consciousness is usually altered during an ictal psychosis episode, and patients are amnesic during the episode. Nonconvulsive status epilepticus may result in symptoms resembling psychosis.
Postictal psychosis refers to psychosis that develops immediately after a seizure or within 1 week of the return of normal mental function. By definition, the psychotic symptoms must last from 24 hours to 3 months. Postictal psychosis accounts for one-quarter of the cases of psychosis in epilepsy, and occurs in roughly 2% of PWE.^20^ This may be an underestimation considering that two telemetry series found prevalence rates of around 6%.^22^,^23^ In patients with intractable focal epilepsy, the prevalence was reported to be as high as 18%.^24^ Postictal psychosis may be mistaken for an acute confusional state. In 1988, Logsdail and Toone^25^ described the following operational criteria for the diagnosis of postictal
Interictal psychosis (schizophrenia-like psychosis in epilepsy) is a psychotic disorder that would fulfill the diagnostic criteria for schizophrenia if epilepsy is not present. Interictal psychotic disorders are of reasonably long duration and are temporally unrelated to the occurrence of seizures.^26^ The psychosis episodes last from days to weeks, and are usually self-limiting, and differentiating them from postictal psychosis episodes may be difficult. The symptoms include paranoid delusions and auditory hallucinations, but multiple other features including affective symptoms may also occur. Anxiety, feelings of oppression, and withdrawal may be premonitory symptoms heralding the psychosis, and treatment with anxiolytics at this stage may prevent the development of full psychosis.^27^,^28^
Slater and Beard^26^ reported that PWE with chronic interictal psychosis usually experienced paranoid-hallucinatory symptoms, commonly associated with catatonia, affective blunting, and volitional symptoms. Phenomenologically, these symptoms are indistinguishable from schizophrenia, although patients tend to exhibit better preservation of affect, mystical experiences, and visual hallucinations. There is typically no deterioration to the hebephrenic state. Other investigators^29^,^30^ have also noted the largely paranoid hallucinatory characteristics of PWE.
The concept of forced normalization was introduced by Landolt^31^ to explain the puzzling observation that the EEGs of PWE often appeared less pathological when their behavior deteriorated, which has also been documented by others.^28^,^32^ Their observations were 1) the EEG may become closer to but not completely normal; 2) the manifestation is not always of psychosis, with other disturbances such as affective symptoms, an anxiety or dissociative state, and behavioral disturbance also possibly being present; and 3) not all brief interictal psychosis episodes are associated with this phenomenon.^33^ More recently, forced normalization has been considered an uncommon phenomenon occurring in patients with chronic, medically refractory partial epilepsy whose seizures are controlled or reduced with medication. Since forced normalization often occurs after adding an effective antiepileptic drug, the psychosis could be a side effect of medication with EEG improvement being an epiphenomenon.^19^
The reported prevalence of psychotic symptoms in PD has varied markedly, from 16% to 75%.^34^ This large variation might be attributable to inconsistencies across studies in the methods used to diagnose psychosis, severity of PD, place of assessment (community vs. clinic), and inclusion or exclusion of patients with dementia. Several previous studies have used Section 1 of the UPDRS (Unified Parkinson’s Disease Rating Scale) to document the behavioral symptoms, including psychosis, whereas other studies have used the Neuropsychiatry Inventory (NPI).
A population study^35^ found that the frequency of hallucinations and/or delusions was 16% in PD patients. Compared with delusions or hallucinations, vivid dreams were reportedly more closely related to taking a high-dosage dopaminergic drug, and also with less dementia or depression. On the other hand, dementia or depression was more closely associated with hallucinations or delusions. The L-dopa dosage was not closely related to hallucinations or delusions. In addition, psychosis was more common in patients with possible PD than definite PD, such as dementia with Lewy-body disease (DLB) or Parkinson’s disease with dementia (PDD). Another study found that hallucinations improved in PDD after cholinergic treatment.^36^ All of these observations suggest that psychosis in PD is related to cholinergic system dysfunction concomitant with that of the dopaminergic system. Histological studies have shown that cortical Lewy bodies are more prevalent in PDD patients, and a preferential paralimbic distribution of these changes may result in these patients exhibiting psychosis.^35^
Considering the heterogeneous manifestations, a working group of the NINDS (National Institutes of Neurological Disorders and Stroke) and the NIMH (National Institutes of Mental Health) proposed the following diagnostic criteria for PD
It is noteworthy that the criteria for psychosis in PD do not include the alteration of thought that is typically observed in schizophrenia. Thus, a large proportion of so-called psychosis in PD is not identical to that of schizophrenia. Psychotic symptoms are more common in DLB than PD patients, and more resemble psychosis present in schizophrenic-spectrum disorders. Visual hallucinations are a key feature of DLB, and 70%–80% of DLB patients experience hallucinations at some point during their illness.^34^ These visual hallucinations take specific and familiar forms. Particularly characteristic are those involving the presence of one or multiple individuals such as children or small adults within the patient’s surroundings. It is particularly interestingly that these hallucinations tend to vanish when the patient attempts to interact with them. Auditory hallucinations, such as hearing music, are less common. Delusions are often closely linked with hallucinations. Common delusions include jealousy or paranoia, such as patients believing that unfamiliar individuals are entering their homes and engaging in affairs with their spouses. These symptoms may lead to aggressive behaviors especially when they are accompanied by dementia.^34^,^37^
Dementia develops in more than 40% of patients with PD, up to years after the diagnosis. Such PDD is characterized by certain features distinct from typical Alzheimer’s disease (AD). While patients with PDD tend to retain recent memories, their executive function shows greater impairment. The hallmarks of PDD are apathy, lack of initiative, and compromised executive abilities. Patients with PDD typically exhibit a progressive dysexecutive syndrome with attentional deficits and fluctuating cognition, often accompanied by psychotic symptoms.^38^
PDD is believed to result from a combination of subcortical and cortical changes. The factors contributing to the development of dementia,^39^ and perhaps also psychosis, are likely to include dopamine deficiency in the nucleus caudatus and mesocortical regions (due to the degeneration of projections from the substantia nigra and ventral tegmental area) along with cholinergic deficiency in the cortex (due to the degeneration of ascending projections from the nucleus basalis of Meynert), coupled with additional AD pathology and cortical Lewy bodies. Clinical trials have found that cholinesterase inhibitors enhance cognitive function in PDD and DLB.^40^ Cholinesterase inhibitors also lead to a reduction in visual hallucinations,^41^ suggesting that hallucinations in these patients are at least partially attributable to dysfunction of the brain cholinergic system.
Naimark et al.^42^ found that 36 (35.6%) of 101 patients with PDD experienced hallucinations and/or delusions. Patients with psychosis were significantly more likely than their nonpsychotic counterparts to experience insomnia, confusion, agitation, personality changes, and self-care problems, and were less able to manage living in their own home. They also had significantly greater impairments in their cognitive and physical functions. Thus, the psychotic symptoms may result from a complex interplay among cognitive impairment or dementia, preexisting predisposition to psychosis, and perhaps also medications.
It is difficult to assess psychosis in patients with dementia because the symptoms are usually intermixed with other cognitive symptoms. It has been found^43^,^44^ that the presence of psychosis in less common in AD (30%), vascular dementia (15%), and frontotemporal dementia (10%) than in DLB and PDD (50%–75%). While visual hallucinations are most common in patients with DLB and PDD, delusions are more common than hallucinations in AD.^45^,^46^
Psychosis can occur in the early stages of dementia, but it becomes more common as the disease progresses, with its prevalence peaking during the moderate stage.^47^ However, this prevalence variation with severity might simply be due to reliable assessments of psychosis being difficult in severe cases. The manifestations of psychosis can differ with the status of dementia. For example, hallucinations are more common than delusions when the cognitive and functional deficits are more severe.^48^ It has also been shown that persecutory delusions, such as delusions of theft, occur earlier in the course of AD than do misidentification delusions, such as TV sign and phantom boarders, which are associated with advanced dementia and greater cognitive impairment.
The psychotic symptoms in patients with dementia are similar to (but not the same as) those seen in patients with schizophrenia. For example, while auditory hallucinations are most common in schizophrenia, patients with dementia (particularly those with DLB) often experience vivid visual hallucinations. These can include images of people, animals, or insects, and moving objects or patterns (e.g., a river flowing over the bed or rain inside the room). Common delusions in dementia include misidentification and paranoid delusions. Misidentification delusions involve the belief that relatives and friends are being substituted by imposters, whereas paranoid delusions include themes such as theft, abandonment, infidelity, and phantom boarders.^46^
The psychotic symptoms in patients with dementia should be differentiated from other neurological symptoms associated with dementia. For example, if a person with dementia does not recognize themselves in a mirror, this could be a misidentification syndrome associated with dementia rather than a visual hallucination. In addition, memory loss or confabulation may be mistakenly considered to be delusions. Generally speaking, persistence of a false belief suggests the presence of delusions, whereas inconsistent statements are more suggestive of confabulations.^44^ However, differentiating these two conditions can be difficult.
The International Psychogeriatric Association released updated criteria for diagnosing psychosis in both major and mild neurocognitive disorders.^46^ This new guideline is based on the Jeste and Finkel criteria^49^ that were previously widely used to define psychosis in dementia. In brief, 1) visual or auditory hallucinations and/or delusions need to be present; 2) all of the criteria for any major and mild neurocognitive disorder need to be satisfied (with the etiological diagnoses specified); 3) the psychotic symptoms must not have been present continuously prior to the onset of the symptoms of dementia; 4) the psychotic symptoms must have been present, at least intermittently, for at least 1 month; 5) the symptoms must be sufficiently severe to cause some disruption to the functioning of the patient and/or others or pose a threat to the safety of the patient or others; and 6) the patient must not have schizophrenia, delirium, or physical conditions causing the psychosis.
In addition, the following instruments can provide quantitative assessments of psychotic symptoms in patients with 1) the BEHAVE-AD (Behavioral Pathology in Alzheimer’s Disease) tool was developed to evaluate behavioral manifestations in AD, including delusions, hallucinations, activity disturbances, aggression, and mood problems, based on the informant’s account; and 2) the NPI is a comprehensive questionnaire used to examine not only the presence but also the frequency and severity of a wide range of mood and behavioral changes in dementia, including delusions and hallucinations.^50^
Advancements in brain imaging techniques such as MRI have made it easier to identify imaging–clinical correlations. However, partly because psychotic symptoms after a focal brain lesion are uncommon and hence have been neglected by neurologists, it remains unclear whether a brain lesion needs to be at a specific location to cause psychotic symptoms. Difficulties in elucidating the lesion–symptom correlation may also be attributable to a remote effect of a focal brain lesion by increasing the intracranial pressure secondary to brain edema or hydrocephalus.
A focal lesion involving a sensory pathway may produce symptoms mimicking psychosis. Tinnitus is probably the most common type of hallucination, and is usually associated with hearing difficulties due to the involved lesions affecting auditory pathways. Two mechanisms have been suggested for the development of tinnitus.^51^ Firstly, hearing loss results in reduced inhibition and increased excitability in neurons, resulting in tinnitus. Secondly, the tonotopic map in the primary auditory cortex may undergo reorganization to involve a larger territory and increased responsiveness of the neurons in processing unaffected auditory frequencies.^52^ Isolated complex hallucinations (e.g., musical hallucinations) may be associated with alteration of higher level activation of the auditory cortex in the posterior temporal lobes, basal ganglia, cerebellum, and inferior frontal cortex. Thus, segmented sounds that were previously experienced by the patients may be heard repeatedly due to an excessive influence of encoding/recognition modules to perception/imagery modules (in a top-down manner) secondary to an impaired sensory input.^53^
Regarding symptoms similar to schizophreniform psychosis, one study^54^ investigated five patients who developed psychotic symptoms following a right hemisphere stroke. Relative to control patients, the patients with psychosis had a larger frontal-horn ratio and larger third-ventricle ratio in CT. Those authors suggested that preexisting subcortical atrophy and a lesion in the right hemisphere, especially in posterior temporo-parieto-occipital areas, may result in the development of poststroke psychosis. This seems to be consistent with the right hemisphere being more closely associated with the perception of the environment. The regions that were most affected included the right frontal, temporal, and parietal regions as well as the caudate nucleus.^55^,^56^ Involvement of the right inferior frontal gyrus and underlying white matter, including the superior longitudinal fasciculus and anterior corona radiata, has also been emphasized by other authors.^56^ One of the present authors observed three patients who had delusions of infidelity after right occipital infarction (unpublished data). All of these patients were elderly males with cognitive impairments and significant visual field defects. Sensory (visual) deficits may provoke suspicion and delusions in persons with underlying brain atrophy and cognitive decline. The presence of preexisting psychiatric disease or susceptibility to psychosis may also increase the risk of psychotic symptoms developing in these patients.^56^
One notable observation is the prolonged time interval between the brain damage and the onset of psychosis. Bray et al.^57^ addressed the relationship between TBI and psychosis by examining the changes in hippocampal volume and the occurrence of psychotic symptoms over time (at 5 and 12 months post-TBI) in 24 subjects who experienced TBI. Psychotic symptoms were assessed using the Personality Assessment Inventory Psychotic Experiences subscale of schizophrenia. They found that the volume loss of the right hippocampus or hippocampal head was significantly associated with psychotic symptoms. It was hypothesized that degeneration of the right hippocampus leads to psychosis via dysregulation of limbic dopaminergic networks in patients who have experienced TBI. Thus, although some controversy remains, TBI may result in psychosis via the degeneration of specific brain areas such as the hippocampus, especially in individuals with a genetic susceptibility. This theory may also be applied to patients with other types of brain damage such as stroke or MS.
It seems that the neurotoxic effects of epilepsy may explain the association between epilepsy and psychosis.^58^ Various mechanisms proposed include a “kindling” process whereby acute seizure discharges cause changes in brain function perhaps via receptor-based changes and changes in cerebral blood flow, resulting in psychosis.^59^ The ongoing subictal activity in the limbic system that is undetectable in an EEG may also lead to brain changes that result in psychosis.^58^,^60^
Temporal-lobe epilepsy is the most common type of epilepsy that is related to psychotic symptoms and also a condition that is poorly controlled by antiseizure medication. The question of whether epilepsy is a risk factor for psychosis and, conversely, whether psychosis is a risk factor for epilepsy have been repeatedly discussed in the literature. Clarke et al.^61^ found that PWE have a 5.5-fold higher risk of having a psychotic disorder and an 8.5-fold higher risk of having schizophrenia. The risk of developing psychosis was twofold higher in individuals with a parental history of epilepsy than in individuals without such a history. Meanwhile, the risk of being diagnosed with generalized epilepsy was 2.7-fold higher in individuals with a parental history of psychosis than in those without such a history.
Neuropathological, neuroimaging, and genetics studies have shown that patients with schizophrenia and PWE have similar brain structural and genetic abnormalities.^58^ Ventricular enlargement has been demonstrated in both conditions, suggesting that phenotypic manifestations share a biological liability that might be related to abnormalities in neurodevelopment. It is possible that certain genes such as those of the leucine-rich glioma-inactivated family are involved in neurodevelopmental abnormalities that may explain the occurrence of both schizophrenia and partial complex seizures with auditory features.
In addition, antiepileptic medications may play a role in the development of psychosis, that include ethosuximide, topiramate, vigabatrin, zonisamide, topiramate, and levetiracetam.^62^ High starting doses of medication, rapid titration, a past psychiatric history, and a high seizure frequency are reported to be associated with an increased risk of psychotic symptoms.
Studies have shown that hallucinations in PD may be associated with reduced cortical cholinergic neural circuit integration,^63^ disrupted visual processing,^64^ amygdala involvement with alpha-synuclein burden,^65^ and the intrusion of dreams into waking life.^66^ However, the neurobiological underpinnings of psychosis in PD are not fully understood. These symptoms may be caused by a progressive abnormal degeneration of dopamine-producing cells in the substantia nigra and ventral tegmentum area in combination with deterioration of the noradrenergic (locus coeruleus), cholinergic forebrain (nucleus basalis of Meynert), and serotoninergic (dorsal raphe nuclei) systems.^67^
Neuroimaging studies^34^ have proposed two mechanisms. The first theory is that PD patients with visual hallucinations have an aberrant top-to-bottom visual processing system dominating the conventional bottom-to-top visual processing system. This is supported by molecular imaging studies documenting lower glucose metabolism or blood flow in the visual cortex and increased glucose metabolism or increased blood flow in frontal areas. Moreover, gray-matter volume reduction in both dorsal and ventral visual pathways was a consistent finding of volumetry-based studies.^68^
The other theory emphasizes the interplay of three neural the default-mode network (DMN), the ventral attention network (VAN), and the dorsal attention network (DAN).^63^ The DMN is a network of brain regions that are active when a person is not engaged in any mental task. In contrast with the task-negative nature of the DMN, the VAN represents a group of brain regions that are active during stimulus-driven activities. Shifting the attention toward an externally focused goal-directed task activates the DAN with concomitant deactivation of the DMN.^69^ Perceptual errors may occur in the background of abnormal interactions between the DMN and VAN. The abnormalities in these networks may therefore be associated with the emergence of psychosis.
Large-scale neural network abnormalities are associated with a neurotransmitter imbalance. For example, striatal dopamine levels are correlated with the activity of the DMN in healthy individuals.^70^ Hence, patients with advanced PD might have impaired signaling involving dopamine, acetylcholine, and serotonin across the attentional networks. The association of long-term use of dopaminergic agents with the emergence of psychosis supports the putative role of a neurotransmission imbalance in the pathogenesis of psychosis in PD. Abnormal serotonin transmission may also be associated with hallucinations in PD.^71^ Further research into the exact roles of these neurotransmitters on the attentional networks might further elucidate the pathogenesis of psychosis in PD. Finally, PD medications including L-dopa, dopamine agonists, monoamine oxidase inhibitors, anticholinergics, and amantadine have been identified as significant contributors to psychosis in PD.^67^
Thus, PD psychosis appears to arise from a complex interplay between medication exposure, PD pathology, aberrant REM-related phenomena, and comorbid factors such as cognitive impairment and visual disturbances.
Changes in various neurotransmitters including dopaminergic, cholinergic, serotonergic, and glutamatergic ones have been considered as a cause of psychosis in dementia.^43^ Patients with dementia usually have widespread brain pathologies, which makes it difficult to interpret the pathogenesis of psychosis based on where the brain damage is located. Nevertheless, various neuroimaging studies have suggested that delusions in AD are associated with frontal-lobe atrophy, as well as hypometabolism and hypoperfusion over the right frontal lobe.^72^ Further, there is increasing evidence that other parts of the brain also play a role in the development of delusions in AD, including the prefrontal regions, the cingulate cortex, parietal lobe, temporal lobe, hippocampus, parahippocampal cortex, and amygdala. This indicates that different symptoms may be associated with dysfunction in different brain regions.^44^ Studies of AD have found psychosis to be associated with tau and vascular pathologies,^44^,^73^ and that certain genetic polymorphisms are associated with psychosis in AD;^74^ however, more studies are needed to confirm these findings.
Psychosis in various neurological disease can have a significant impact on the quality of life. It can lead to more-rapid deteriorations in neurological functions and physical health, increased stress in caregivers, and earlier admissions to care facilities. A failure to identify and treat psychosis in a timely manner could trigger other emotions including fear and distrust, leading to disturbing and potentially risky behaviors.^7^ Psychosis may also be associated with increased mortality. One retrospective study^8^ found that 6.7% of 1,129 patients with stroke were diagnosed as having psychosis. Those authors found that poststroke psychosis was associated with increased mortality. However, the mechanisms underlying such an association remain known. It is possible that patients with psychosis have greater difficulties in coping with the consequences of stroke. The association may also be due to poor adherence with medical advice or a healthy lifestyle, suicidal ideation, or comorbid medical diseases.
There is a risk of psychosis in patients with dementia being misdiagnosed as schizophrenia. This could lead to the unwarranted administration of antipsychotics, thereby potential exposing the individual to their adverse effects. A holistic approach is crucial to mitigate these risks and enhance the overall well-being of individuals with neurological diseases and psychosis. This requires the collaboration of a multidisciplinary healthcare team that conducts thorough assessments, implements supportive interventions, and provides caregivers with timely education and guidance.
Various medications are used to treat psychotic symptoms in patients with neurological diseases. One problem is that medication side effects such as extrapyramidal symptoms are more concerning in these patients than in psychiatric patients without neurological diseases. For this reason, we will start to discuss the treatment of psychosis in patients with PD and related diseases.
Management strategies for psychosis in PD or DLB patients should be based on potential underlying causes. This involves general measures, addressing specific triggers, adjusting medication dosages, and/or introducing cholinesterase inhibitors for cognitively impaired patients. Firstly, if dopaminergic drugs used for PD treatment is considered excessive, dosages can be reduced. Secondly, because cholinergic circuit abnormalities could contribute to hallucinations, cholinesterase inhibitors (especially rivastigmine) may be tried, especially in patients with dementia. If psychotic symptoms persist, antipsychotics may then be introduced. Notably, typical antipsychotics may worsen PD motor symptoms, while atypical antipsychotics can improve psychiatric symptoms without significantly exacerbating motor symptoms.^75^
Pharmacological approaches in PD and related diseases have been summarized previously.^34^ Pimavanserin is a combined inverse agonist of the 5-HT2A and 5-HT2C receptors that has been approved by the FDA. It is an atypical antipsychotic that uniquely lacks dopamine-blocking properties, and both its efficacy and safety were demonstrated in a randomized trial.^75^ Pimavanserin also resulted in significant improvements in measures of nighttime sleep, daytime wakefulness, and caregiver burden. Clozapine is one of the second-generation antipsychotics that is not usually associated with significant extrapyramidal symptoms. Although uncommon, agranulocytosis may occur, and so optimal blood monitoring is required. The efficacy of quetiapine has not been confirmed due to inconsistent results in different studies. Nevertheless, it is often prescribed, probably due to its relatively good tolerability and lower likelihood of exacerbating PD features. Its side effects include drowsiness, worsening of constipation, and QT prolongation. Studies of PD patients with cognitive impairment and hallucinations have found rivastigmine to be well-tolerated and effective in improving cognitive function as well as behavioral symptoms. There have been several case reports of improvements in psychotic symptoms after treatment with donepezil, and so cholinesterase inhibitors can be used in PDD with hallucinations.
Nonpharmacological approaches can also be considered. Several studies have found electroconvulsive therapy (ECT) to be useful in alleviating the psychotic symptoms in PD. Although additional studies are warranted to elucidate its long-term effects, ECT is probably useful in drug-resistant psychosis in PD. Repetitive transcranial magnetic stimulation (rTMS) is a noninvasive brain stimulation technique that is useful for treating patients with movement disorders and psychotic symptoms. Since rTMS is noninvasive and is not associated with serious adverse events, future trials are warranted to assess its efficacy in the treatment of psychosis in PD.
Nonpharmacological measures need to be considered for patients with dementia before administering antipsychotic medications. The DICE approach is recommended.^76^ Firstly, caregivers are taught to describe or interpret the behaviors caused by dementia. Secondly, factors that may contribute to these behaviors are investigated, which can be related to specific characteristics of the patient, caregiver, and external environment. Thirdly, a management plan is devised that allows the caregiver to address the identifiable factors. Fourthly, the response outcome is evaluated. It is important to note that the presence of delusions or hallucinations does not automatically necessitate pharmacological treatment.
Antipsychotics should be reserved for those who might cause serious harm to themselves or others or for those who continue to exhibit disturbing behaviors despite nonpharmacological interventions. Antipsychotics such as risperidone are used to treat psychosis in patients with dementia. However, they are known to cause various adverse effects, including extrapyramidal symptoms and an increased risk of falls. It is therefore necessary to balance the potential benefits and harms of antipsychotics.^77^,^78^ Pimavanserin has recently shown promise in treating psychosis in various types of dementia.^79^,^80^
Various medications such as haloperidol, risperidone, quetiapine, and olanzapine are used in patients with stroke and other types of focal neurological damage. However, there have been no large, randomized clinical trials. Because patients with stroke are usually elderly and frail, side effects should be carefully monitored that include dizziness, drowsiness, weight gain, falling down and movement disorders. Although psychological treatments are often tried, their efficacy has not been clearly documented. The treatment strategies for psychosis in TBI or MS patients are similar to those with other neurological diseases. Before applying medical therapy, establishing the definitive diagnosis is important because psychotic symptoms may precede neurological diseases. A paper^10^ reported that immunosuppressive therapy initiated after the diagnosis of MS was significantly more effective than antipsychotics that had been given before in treating psychotic symptoms.
This narrative review has presented the psychotic symptoms such as hallucinations and delusions that occur in various neurological diseases. One important issue to address was whether there are differences in psychotic symptoms between neurological disorders and psychiatric diseases. Psychotic symptoms are typically observed in patients with schizophrenia, that is diagnosed using DSM-5 ^81^ the presence of two or more symptoms including 1) delusions, 2) hallucinations, 3) disorganized speech, 4) grossly disorganized or catatonic behavior, and 5) negative symptoms (i.e., diminished emotion, expression, or avolition). Each symptom should present for a significant proportion of a 1-month period (or less if successfully treated), with at least one of items 1–3 appearing.
As discussed above, the most important features of psychotic symptoms in patients with organic brain disease are hallucinations and delusions. They rarely present with items 3 and 4. However, patients with frontal-lobe damage may present with item 5. Thus, aside from the presence of neurological symptoms, the absence of items 3 and 4 may help physicians to differentiate psychotic symptoms due to neurological disease from schizophrenia. There are also differentiating aspects for items 1 and 2. In patients with neurological diseases, hallucinations are mostly unimodal whereas the prevalence of multimodal hallucinations was about twofold higher than that of unimodal hallucinations in patients with schizophrenia,^82^ which included visual, auditory, olfactory, gustatory, and tactile hallucinations. These observations suggest that hallucinations in schizophrenia are caused by dysfunction of a higher integrative center beyond each sensory pathway.^83^
Among various types of hallucinations, patients with schizophrenia most often have auditory hallucinations that usually include human voices. However, this alone is insufficient for differentiating schizophrenia from neurological diseases because, as discussed above, more-complex hallucinations such as musical hallucinations or vivid visual hallucinations can appear in patients who have lesions affecting the sensory pathways. One of the differences appears to be the intensity and bizarreness of associated delusions. Delusions are the firm belief that a patient considers their belief to be true even when this is not supported by their current situation.^84^ The psychotic manifestations in schizophrenia are usually more complex and bizarre than hallucinations associated with organic brain damage, and are clearly associated with desultory thought processes. Patients may even discuss or argue with the hallucinatory figures or be controlled by them, or say that somebody is depriving or inserting thoughts to their brain. These types of manifestations are quite rare in patients with neurological diseases.
Waters et al.^85^ described four features that can be used to differentiate auditory hallucinations in schizophrenia from those in other 1) a failure of self-recognition, whereby the experience is perceived as alien and separate from one’s own thought processes; 2) reduced sense of control over the onset, content, and frequency of hallucinations; 3) the perceived realism of hallucinations; and 4) the contribution of emotions. Auditory hallucinations in schizophrenia are often experienced as separate from one’s own thought processes. In addition, the origin of these events tends to be incorrectly attributed to an external agent, which could be due to the failure of self-monitoring processes whose role is to predict the sensory consequences of one’s own actions. These typically engage memory and other decision-making processes that retrieve and evaluate memory records in order to form a cohesive representation of an experience. It has been proposed that this type of disturbance to the origin-monitoring process results in an incomplete representation of mental events and consequently a failure to identify their origins.
The pathogenesis of psychotic symptoms in major neurological diseases has been described above, and it needs to be compared with that in schizophrenia. Although schizophrenia has been considered a psychiatric rather than a neurological disease, there have been reports of structural brain damage in patients with schizophrenia, which includes enlargement of ventricles, cortical atrophy, cerebral asymmetry, decreased volume of the temporal lobes, and especially decreased volumes of the superior temporal gyrus and hippocampus. Cell loss has been detected histologically in the hippocampus and entorhinal cortex. These data suggest that schizophrenia is a progressive brain disorder.^86^,^87^ The extent of brain changes might also be influenced by antipsychotic medications, substance abuse, and other secondary factors.^87^
A meta-analysis^88^ revealed significant reductions in the thalamus volume bilaterally in patients with schizophrenia compared with healthy control subjects. It has been proposed that alterations in the function of the thalamus are associated with heterogeneous symptoms in schizophrenia. The thalamus is a starting point for several parallel, overlapping networks that extend to the cortex. The changes in the thalamic nodes of these networks are echoed by changes at other points along the pathway, suggesting that schizophrenia is attributable to the disruption of thalamocortical neural networks. Dysfunction in the coordination of the flow of information within and between various neural networks may result in symptoms such as hallucinations, delusions, disorganized behaviors and speech, and thought disorders.^89^ Other regions with altered gray-matter volumes have also been identified, including the insula, inferior parietal lobule, and anterior cingulate cortex.^90^ All of these regions can be broadly regarded as thalamocortical networks.
Ren et al.^91^ examined patients with schizophrenia using structural and functional MRI to reveal alterations of gray-matter volumes in multiple areas of the thalamocortical networks. However, the areas of decreased functional activities were not consistent with those in which volume changes were found. The functional abnormalities were generally located within frontoparietal DMNs, which are thought to be important for decision-making, working memory, and general monitoring of the internal and external environments. Functional deficits in the medial frontal areas might be associated with symptoms such as apathy, lack of emotion, and blunted affect. Alterations of functional activities might be related to acute psychosis, with changes in brain anatomy reflecting more-stable and long-standing alterations. Functional connectivity aberrations in the DMN hubs of the dorsomedial prefrontal cortex and precuneus in schizophrenia were also observed by other authors.^92^
The above-described observations indicate that the pathogenesis of psychotic symptoms appears to be similar in neurological diseases and schizophrenia. As discussed above, even if patients with neurological diseases show hallucinations or delusions, they rarely deteriorate into a hebephrenic or catatonic state. Since these negative symptoms are reportedly associated with frontal-lobe dysfunction, frontal dysfunction might be more widespread or more severe in patients with schizophrenia than in those with neurological disease with psychosis. This is consistent with the observation that psychotic symptoms in patients with widespread cortical pathologies such as DLB, PDD, or AD more closely resemble those of schizophrenia than the symptoms observed in patients with focal brain damage.
This study had some limitations, and so caution is advised when interpreting the presented findings. Firstly, the definition of psychosis in patients with neurological diseases is omitted or incomplete in many literature reports. Even when it has been defined, whether the definition used in psychiatric patients is valid and reliable in patients with neurological diseases has not been properly investigated. Secondly, an apparent psychotic symptoms might not be a real psychotic symptom but rather a misinterpretation of a neurological manifestation. For example, delusions may be attributable to impaired memory or judgement in patients with dementia. Thirdly, schizophrenia is a disease defined only by a constellation of symptoms rather than by objective imaging or pathological data. It might therefore represent a group of heterogeneous diseases, thereby precluding appropriate comparison studies.
In summary, although not common, patients with neurological diseases may present with psychotic symptoms. The symptoms appear to decrease the quality of life, reduce the ability to perform the normal activities of daily living, increase caregiver burden, or even increase mortality. This situation makes early diagnosis and appropriate management essential. The management strategies are similar for different underlying neurological diseases, but they need to be tailored accordingly; for example, for patients with PD and related diseases, agents that have minimal extrapyramidal side effects should be considered. Further studies are needed to elucidate the pathophysiology of psychotic symptoms in neurological diseases, which will also help improve our understanding of the pathophysiology of schizophrenia.
Data sharing not applicable to this article as no datasets were generated or analyzed during the study.
Data sharing not applicable to this article as no datasets were generated or analyzed during the study.