Current Psychiatry

Discuss this article at www.facebook.com/CurrentPsychiatry

Alzheimer’s disease (AD) and dementia with Lewy bodies (DLB) are the first and second most common causes of neurodegenerative dementia, respectively.^{“New Alzheimer’s disease guidelines: Implications for clinicians,” Current Psychiatry, March 2012, p. 15-20; http://bit.ly/UNYikk.}

The 2005 report of the DLB Consortium⁵ recognizes central, core, suggestive, and supportive features of DLB (Table 1).^5,10 These features are considered in the context of other confounding clinical conditions and the timing of cognitive and motor symptoms. The revised DLB criteria⁵ require a central feature of progressive cognitive decline. “Probable DLB” is when a patient presents with 2 core features or 1 core feature and ≥1 suggestive features. A diagnosis of “possible DLB” requires 1 core feature or 1 suggestive feature in the presence of progressive cognitive decline.

Table 1

Diagnostic criteria for AD and DLB

Biomarkers for AD, but not DLB

The 2011 diagnostic criteria for AD incorporate biomarkers that can be measured in vivo and reflect speci?c features of disease-related pathophysiologic processes. Biomarkers for AD are divided into 2 categories:¹¹

• amyloid-beta (Aβ) accumulation: abnormal tracer retention on amyloid positron emission topography (PET) imaging and low cerebrospinal fluid (CSF) Aβ42
• neuronal degeneration or injury: elevated CSF tau (total and phosphorylated tau), decreased ?uorodeoxyglucose uptake on PET in temporo-parietal cortices, and atrophy on structural MRI in the hippocampal and temporo-parietal regions.

No clinically applicable genotypic or CSF markers exist to support a DLB diagnosis, but there are many promising candidates, including elevated levels of CSF p-tau 181, CSF levels of alpha- and beta-synuclein,¹² and CSF beta-glucocerebrosidase levels.¹³ PET mapping of brain acetylcholinesterase activity,¹⁴ 123I-2β-carbomethoxy-3β- (4-iodophenyl)-N-(3-fluoropropyl)nortropane single photon emission computed tomography (SPECT) dopamine transporter (DaT) imaging¹⁵ and metaiodobenzylguanidine (MIBG) scintigraphy also are promising methods. DaT scan SPECT is FDA-approved for detecting loss of functional dopaminergic neuron terminals in the striatum and can differentiate between AD and DLB with a sensitivity and specificity of 78% to 88% and 94% to 100%, respectively.¹⁶ This test is covered by Medicare for differentiating AD and DLB.

Differences in presentation

Cognitive impairment. Contrary to the early memory impairment that characterizes AD, memory deficits in DLB usually appear later in the disease course.⁵ Patients with DLB manifest greater attentional, visuospatial, and executive impairments than those with AD, whereas AD causes more profound episodic (declarative) memory impairment than DLB. DLB patients show more preserved consolidation and storage of verbal information than AD patients because of less neuroanatomical and cholinergic compromise in the medial temporal lobe. There is no evidence of significant differences in remote memory, semantic memory, and language (naming and fluency).

Compromised attention in DLB may be the basis for fluctuating cognition, a characteristic of the disease. The greater attentional impairment and reaction time variability in DLB compared with AD is evident during complex tasks for attention and may be a function of the executive and visuospatial demands of the tasks.¹⁷

Executive functions critical to adaptive, goal-directed behavior are more impaired in DLB than AD. DLB patients are more susceptible to distraction and have difficulty engaging in a task and shifting from 1 task to another. This, together with a tendency for confabulation and perseveration, are signs of executive dysfunction.

Neuropsychiatric features. DLB patients are more likely than AD patients to exhibit psychiatric symptoms and have more functional impairment.¹⁸ In an analysis of autopsy-confirmed cases, hallucinations and delusions were more frequent with Lewy body pathology (75%) than AD (21%) at initial clinical evaluation.¹⁸ By the end stages of both illnesses, the degree of psychotic symptoms is comparable.¹⁹ Depression is common in DLB; whether base rates of depressed mood and major depression differ between DLB and AD is uncertain.²⁰

Psychosis in AD can be induced by medication or delirium, or triggered by poor sensory perceptions. Psychotic symptoms occur more frequently during the moderate and advanced stages of AD, when patients present with visual hallucinations, delusions, or delusional misidentifications. As many as 10% to 20% of patients with AD experience hallucinations, typically visual. Delusions occur in 30% to 50% of AD patients, usually in the later stages of the disease. The most common delusional themes are infidelity, theft, and paranoia. Female sex is a risk factor for psychosis in AD. Delusions co-occur with aggression, anxiety, and aberrant motor behavior.

Visual hallucinations—mostly vivid, well-formed, false perceptions of insects, animals, or people—are the defining feature of DLB.²¹ Many patients recognize that they are experiencing visual hallucinations and can ignore them. DLB patients also may experience visual illusions, such as misperceiving household objects as living beings. Delusions—typically paranoid—are common among DLB patients, as are depression and anxiety.¹ Agitation or aggressive behavior tends to occur late in the illness, if at all.

The causes of psychotic symptoms in DLB are not fully understood, but dopamine dysfunction likely is involved in hallucinations, delusions, and agitation, and serotonin dysfunction may be associated with depression and anxiety. Rapid eye movement (REM) sleep/wakefulness dysregulation, in which the dream imagery of REM sleep may occur during wakefulness, also has been proposed as a mechanism for visual hallucinations in DLB.²² In DLB, psychotic symptoms occur early and are a hallmark of this illness, whereas in AD they usually occur in the middle to late stages of the disease.

Motor symptoms. In AD, extrapyramidal symptoms (EPS) are common later in the disease, are strongly correlated with disease severity, and are a strong, independent predictor of depression severity.²³ EPS are more common in DLB than in AD²⁴ and DLB patients are at higher risk of developing EPS even with low doses of typical antipsychotics, compared with AD patients.²⁵

Other symptoms. REM sleep behavior disorder (RBD) is characterized by enacting dreams—often violent—during REM sleep. RBD is common in DLB and many patients also have excessive daytime somnolence. Other sleep disorders in DLB include insomnia, obstructive sleep apnea, central sleep apnea, restless legs syndrome, and periodic limb movements during sleep.

In AD patients, common sleep behaviors include confusion in the early evening (“sundowning”) and frequent nighttime awakenings, often accompanied by wandering.²⁶ Orthostatic hypotension, impotence, urinary incontinence, and constipation are common in DLB. Lack of insight concerning personal cognitive, mood, and behavioral state is highly prevalent in AD patients and more common than in DLB.

Diagnostic evaluation

Because there are no definitive clinical markers for DLB, diagnosis is based on a detailed clinical and family history from the patient and a reliable informant, as well as a physical, neurologic, and mental status examination that looks for associated noncognitive symptoms, and neuropsychological evaluation. Reasons DLB may be misdiagnosed include:

• Some “core” clinical features of DLB may not appear or may overlap with AD.
• Presence and severity of concurrent AD pathology in DLB may modify the clinical presentation, with decreased rates of hallucinations and parkinsonism and increased neurofibrillary tangles.
• Failure to reliably identify fluctuations—variations in cognition and arousal, such as periods of unresponsiveness while awake (“zoning out”), excessive daytime somnolence, and disorganized speech.²⁷

Detecting and characterizing cognitive deficits in dementia patients using neuropsychological testing is important in establishing a clinical diagnosis, determining baseline levels of impairment, forming a prognosis, and initiating disease-specific treatments. Differences in neuropsychological findings in AD and DLB are outlined in Table 2.^16,28-33 Several studies have suggested using these measures to differentiate patients with DLB from those with AD.²⁰

Table 2

Diagnostic testing for Alzheimer’s disease and dementia with Lewy bodies

Evidence is insufficient to support using electroencephalographic and polysomnographic studies when initially evaluating patients with dementia. Brain CT or MRI are recommended as part of the initial evaluation of dementia patients to exclude treatable causes of dementia and help clarify the differential diagnosis. Occipital hypometabolism and hypoperfusion demonstrated on PET and SPECT imaging have high sensitivity and specificity for differentiating AD from DLB.

To diagnose DLB more consistently, look for core features of the disease, RBD, antipsychotic hypersensitivity, and decreased striatal binding at presynaptic DaT sites.¹⁵ Abnormal (low binding) DaT activity is the most reliable diagnostic marker for DLB.³⁴ Myocardial scintigraphy with MIBG is sensitive to pathologic changes of DLB before clinical expression and could overcome the difficulties of using clinical criteria alone to identify patients with DLB.³⁵ MIBG scintigraphy may be preferred to DaT scan because it is less expensive and its sensitivity and specificity to DLB are independent of the presence of parkinsonism.³⁵

For an overview of pharmacotherapy options for patients with AD or DLB, see Box 2.

Box 2

Related Resources

• Hanyu H, Sato T, Hirao K, et al. Differences in clinical course between dementia with Lewy bodies and Alzheimer’s disease. Eur J Neurol. 2009;16(2):212-217.
• Walker Z, McKeith I, Rodda J, et al. Comparison of cognitive decline between dementia with Lewy bodies and Alzheimer’s disease: a cohort study. BMJ Open. 2012;2:e000380.

Drug Brand Names

• Aripiprazole • Abilify
• Clonazepam • Klonopin
• Donepezil • Aricept
• Galantamine • Razadyne, Reminyl
• Levodopa • Dopar, Larodopa
• Memantine • Namenda
• Quetiapine • Seroquel
• Rivastigmine • Exelon

Disclosure

Drs. Bishnoi and Manepalli report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Grossberg serves as a consultant to Forest, Janssen, Novartis, and Pfizer. His department receives research funding from Novartis, Janssen, and Pfizer.

Discuss this article at www.facebook.com/CurrentPsychiatry

Alzheimer’s disease (AD) and dementia with Lewy bodies (DLB) are the first and second most common causes of neurodegenerative dementia, respectively.^{“New Alzheimer’s disease guidelines: Implications for clinicians,” Current Psychiatry, March 2012, p. 15-20; http://bit.ly/UNYikk.}

The 2005 report of the DLB Consortium⁵ recognizes central, core, suggestive, and supportive features of DLB (Table 1).^5,10 These features are considered in the context of other confounding clinical conditions and the timing of cognitive and motor symptoms. The revised DLB criteria⁵ require a central feature of progressive cognitive decline. “Probable DLB” is when a patient presents with 2 core features or 1 core feature and ≥1 suggestive features. A diagnosis of “possible DLB” requires 1 core feature or 1 suggestive feature in the presence of progressive cognitive decline.

Table 1

Diagnostic criteria for AD and DLB

Biomarkers for AD, but not DLB

The 2011 diagnostic criteria for AD incorporate biomarkers that can be measured in vivo and reflect speci?c features of disease-related pathophysiologic processes. Biomarkers for AD are divided into 2 categories:¹¹

• amyloid-beta (Aβ) accumulation: abnormal tracer retention on amyloid positron emission topography (PET) imaging and low cerebrospinal fluid (CSF) Aβ42
• neuronal degeneration or injury: elevated CSF tau (total and phosphorylated tau), decreased ?uorodeoxyglucose uptake on PET in temporo-parietal cortices, and atrophy on structural MRI in the hippocampal and temporo-parietal regions.

No clinically applicable genotypic or CSF markers exist to support a DLB diagnosis, but there are many promising candidates, including elevated levels of CSF p-tau 181, CSF levels of alpha- and beta-synuclein,¹² and CSF beta-glucocerebrosidase levels.¹³ PET mapping of brain acetylcholinesterase activity,¹⁴ 123I-2β-carbomethoxy-3β- (4-iodophenyl)-N-(3-fluoropropyl)nortropane single photon emission computed tomography (SPECT) dopamine transporter (DaT) imaging¹⁵ and metaiodobenzylguanidine (MIBG) scintigraphy also are promising methods. DaT scan SPECT is FDA-approved for detecting loss of functional dopaminergic neuron terminals in the striatum and can differentiate between AD and DLB with a sensitivity and specificity of 78% to 88% and 94% to 100%, respectively.¹⁶ This test is covered by Medicare for differentiating AD and DLB.

Differences in presentation

Cognitive impairment. Contrary to the early memory impairment that characterizes AD, memory deficits in DLB usually appear later in the disease course.⁵ Patients with DLB manifest greater attentional, visuospatial, and executive impairments than those with AD, whereas AD causes more profound episodic (declarative) memory impairment than DLB. DLB patients show more preserved consolidation and storage of verbal information than AD patients because of less neuroanatomical and cholinergic compromise in the medial temporal lobe. There is no evidence of significant differences in remote memory, semantic memory, and language (naming and fluency).

Compromised attention in DLB may be the basis for fluctuating cognition, a characteristic of the disease. The greater attentional impairment and reaction time variability in DLB compared with AD is evident during complex tasks for attention and may be a function of the executive and visuospatial demands of the tasks.¹⁷

Executive functions critical to adaptive, goal-directed behavior are more impaired in DLB than AD. DLB patients are more susceptible to distraction and have difficulty engaging in a task and shifting from 1 task to another. This, together with a tendency for confabulation and perseveration, are signs of executive dysfunction.

Neuropsychiatric features. DLB patients are more likely than AD patients to exhibit psychiatric symptoms and have more functional impairment.¹⁸ In an analysis of autopsy-confirmed cases, hallucinations and delusions were more frequent with Lewy body pathology (75%) than AD (21%) at initial clinical evaluation.¹⁸ By the end stages of both illnesses, the degree of psychotic symptoms is comparable.¹⁹ Depression is common in DLB; whether base rates of depressed mood and major depression differ between DLB and AD is uncertain.²⁰

Psychosis in AD can be induced by medication or delirium, or triggered by poor sensory perceptions. Psychotic symptoms occur more frequently during the moderate and advanced stages of AD, when patients present with visual hallucinations, delusions, or delusional misidentifications. As many as 10% to 20% of patients with AD experience hallucinations, typically visual. Delusions occur in 30% to 50% of AD patients, usually in the later stages of the disease. The most common delusional themes are infidelity, theft, and paranoia. Female sex is a risk factor for psychosis in AD. Delusions co-occur with aggression, anxiety, and aberrant motor behavior.

Visual hallucinations—mostly vivid, well-formed, false perceptions of insects, animals, or people—are the defining feature of DLB.²¹ Many patients recognize that they are experiencing visual hallucinations and can ignore them. DLB patients also may experience visual illusions, such as misperceiving household objects as living beings. Delusions—typically paranoid—are common among DLB patients, as are depression and anxiety.¹ Agitation or aggressive behavior tends to occur late in the illness, if at all.

The causes of psychotic symptoms in DLB are not fully understood, but dopamine dysfunction likely is involved in hallucinations, delusions, and agitation, and serotonin dysfunction may be associated with depression and anxiety. Rapid eye movement (REM) sleep/wakefulness dysregulation, in which the dream imagery of REM sleep may occur during wakefulness, also has been proposed as a mechanism for visual hallucinations in DLB.²² In DLB, psychotic symptoms occur early and are a hallmark of this illness, whereas in AD they usually occur in the middle to late stages of the disease.

Motor symptoms. In AD, extrapyramidal symptoms (EPS) are common later in the disease, are strongly correlated with disease severity, and are a strong, independent predictor of depression severity.²³ EPS are more common in DLB than in AD²⁴ and DLB patients are at higher risk of developing EPS even with low doses of typical antipsychotics, compared with AD patients.²⁵

Other symptoms. REM sleep behavior disorder (RBD) is characterized by enacting dreams—often violent—during REM sleep. RBD is common in DLB and many patients also have excessive daytime somnolence. Other sleep disorders in DLB include insomnia, obstructive sleep apnea, central sleep apnea, restless legs syndrome, and periodic limb movements during sleep.

In AD patients, common sleep behaviors include confusion in the early evening (“sundowning”) and frequent nighttime awakenings, often accompanied by wandering.²⁶ Orthostatic hypotension, impotence, urinary incontinence, and constipation are common in DLB. Lack of insight concerning personal cognitive, mood, and behavioral state is highly prevalent in AD patients and more common than in DLB.

Diagnostic evaluation

Because there are no definitive clinical markers for DLB, diagnosis is based on a detailed clinical and family history from the patient and a reliable informant, as well as a physical, neurologic, and mental status examination that looks for associated noncognitive symptoms, and neuropsychological evaluation. Reasons DLB may be misdiagnosed include:

• Some “core” clinical features of DLB may not appear or may overlap with AD.
• Presence and severity of concurrent AD pathology in DLB may modify the clinical presentation, with decreased rates of hallucinations and parkinsonism and increased neurofibrillary tangles.
• Failure to reliably identify fluctuations—variations in cognition and arousal, such as periods of unresponsiveness while awake (“zoning out”), excessive daytime somnolence, and disorganized speech.²⁷

Detecting and characterizing cognitive deficits in dementia patients using neuropsychological testing is important in establishing a clinical diagnosis, determining baseline levels of impairment, forming a prognosis, and initiating disease-specific treatments. Differences in neuropsychological findings in AD and DLB are outlined in Table 2.^16,28-33 Several studies have suggested using these measures to differentiate patients with DLB from those with AD.²⁰

Table 2

Diagnostic testing for Alzheimer’s disease and dementia with Lewy bodies

Evidence is insufficient to support using electroencephalographic and polysomnographic studies when initially evaluating patients with dementia. Brain CT or MRI are recommended as part of the initial evaluation of dementia patients to exclude treatable causes of dementia and help clarify the differential diagnosis. Occipital hypometabolism and hypoperfusion demonstrated on PET and SPECT imaging have high sensitivity and specificity for differentiating AD from DLB.

To diagnose DLB more consistently, look for core features of the disease, RBD, antipsychotic hypersensitivity, and decreased striatal binding at presynaptic DaT sites.¹⁵ Abnormal (low binding) DaT activity is the most reliable diagnostic marker for DLB.³⁴ Myocardial scintigraphy with MIBG is sensitive to pathologic changes of DLB before clinical expression and could overcome the difficulties of using clinical criteria alone to identify patients with DLB.³⁵ MIBG scintigraphy may be preferred to DaT scan because it is less expensive and its sensitivity and specificity to DLB are independent of the presence of parkinsonism.³⁵

For an overview of pharmacotherapy options for patients with AD or DLB, see Box 2.

Box 2

Related Resources

• Hanyu H, Sato T, Hirao K, et al. Differences in clinical course between dementia with Lewy bodies and Alzheimer’s disease. Eur J Neurol. 2009;16(2):212-217.
• Walker Z, McKeith I, Rodda J, et al. Comparison of cognitive decline between dementia with Lewy bodies and Alzheimer’s disease: a cohort study. BMJ Open. 2012;2:e000380.

Drug Brand Names

• Aripiprazole • Abilify
• Clonazepam • Klonopin
• Donepezil • Aricept
• Galantamine • Razadyne, Reminyl
• Levodopa • Dopar, Larodopa
• Memantine • Namenda
• Quetiapine • Seroquel
• Rivastigmine • Exelon

Disclosure

Drs. Bishnoi and Manepalli report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Grossberg serves as a consultant to Forest, Janssen, Novartis, and Pfizer. His department receives research funding from Novartis, Janssen, and Pfizer.

Discuss this article at www.facebook.com/CurrentPsychiatry

Alzheimer’s disease (AD) and dementia with Lewy bodies (DLB) are the first and second most common causes of neurodegenerative dementia, respectively.^{“New Alzheimer’s disease guidelines: Implications for clinicians,” Current Psychiatry, March 2012, p. 15-20; http://bit.ly/UNYikk.}

The 2005 report of the DLB Consortium⁵ recognizes central, core, suggestive, and supportive features of DLB (Table 1).^5,10 These features are considered in the context of other confounding clinical conditions and the timing of cognitive and motor symptoms. The revised DLB criteria⁵ require a central feature of progressive cognitive decline. “Probable DLB” is when a patient presents with 2 core features or 1 core feature and ≥1 suggestive features. A diagnosis of “possible DLB” requires 1 core feature or 1 suggestive feature in the presence of progressive cognitive decline.

Table 1

Diagnostic criteria for AD and DLB

Biomarkers for AD, but not DLB

The 2011 diagnostic criteria for AD incorporate biomarkers that can be measured in vivo and reflect speci?c features of disease-related pathophysiologic processes. Biomarkers for AD are divided into 2 categories:¹¹

• amyloid-beta (Aβ) accumulation: abnormal tracer retention on amyloid positron emission topography (PET) imaging and low cerebrospinal fluid (CSF) Aβ42
• neuronal degeneration or injury: elevated CSF tau (total and phosphorylated tau), decreased ?uorodeoxyglucose uptake on PET in temporo-parietal cortices, and atrophy on structural MRI in the hippocampal and temporo-parietal regions.

No clinically applicable genotypic or CSF markers exist to support a DLB diagnosis, but there are many promising candidates, including elevated levels of CSF p-tau 181, CSF levels of alpha- and beta-synuclein,¹² and CSF beta-glucocerebrosidase levels.¹³ PET mapping of brain acetylcholinesterase activity,¹⁴ 123I-2β-carbomethoxy-3β- (4-iodophenyl)-N-(3-fluoropropyl)nortropane single photon emission computed tomography (SPECT) dopamine transporter (DaT) imaging¹⁵ and metaiodobenzylguanidine (MIBG) scintigraphy also are promising methods. DaT scan SPECT is FDA-approved for detecting loss of functional dopaminergic neuron terminals in the striatum and can differentiate between AD and DLB with a sensitivity and specificity of 78% to 88% and 94% to 100%, respectively.¹⁶ This test is covered by Medicare for differentiating AD and DLB.

Differences in presentation

Cognitive impairment. Contrary to the early memory impairment that characterizes AD, memory deficits in DLB usually appear later in the disease course.⁵ Patients with DLB manifest greater attentional, visuospatial, and executive impairments than those with AD, whereas AD causes more profound episodic (declarative) memory impairment than DLB. DLB patients show more preserved consolidation and storage of verbal information than AD patients because of less neuroanatomical and cholinergic compromise in the medial temporal lobe. There is no evidence of significant differences in remote memory, semantic memory, and language (naming and fluency).

Compromised attention in DLB may be the basis for fluctuating cognition, a characteristic of the disease. The greater attentional impairment and reaction time variability in DLB compared with AD is evident during complex tasks for attention and may be a function of the executive and visuospatial demands of the tasks.¹⁷

Executive functions critical to adaptive, goal-directed behavior are more impaired in DLB than AD. DLB patients are more susceptible to distraction and have difficulty engaging in a task and shifting from 1 task to another. This, together with a tendency for confabulation and perseveration, are signs of executive dysfunction.

Neuropsychiatric features. DLB patients are more likely than AD patients to exhibit psychiatric symptoms and have more functional impairment.¹⁸ In an analysis of autopsy-confirmed cases, hallucinations and delusions were more frequent with Lewy body pathology (75%) than AD (21%) at initial clinical evaluation.¹⁸ By the end stages of both illnesses, the degree of psychotic symptoms is comparable.¹⁹ Depression is common in DLB; whether base rates of depressed mood and major depression differ between DLB and AD is uncertain.²⁰

Psychosis in AD can be induced by medication or delirium, or triggered by poor sensory perceptions. Psychotic symptoms occur more frequently during the moderate and advanced stages of AD, when patients present with visual hallucinations, delusions, or delusional misidentifications. As many as 10% to 20% of patients with AD experience hallucinations, typically visual. Delusions occur in 30% to 50% of AD patients, usually in the later stages of the disease. The most common delusional themes are infidelity, theft, and paranoia. Female sex is a risk factor for psychosis in AD. Delusions co-occur with aggression, anxiety, and aberrant motor behavior.

Visual hallucinations—mostly vivid, well-formed, false perceptions of insects, animals, or people—are the defining feature of DLB.²¹ Many patients recognize that they are experiencing visual hallucinations and can ignore them. DLB patients also may experience visual illusions, such as misperceiving household objects as living beings. Delusions—typically paranoid—are common among DLB patients, as are depression and anxiety.¹ Agitation or aggressive behavior tends to occur late in the illness, if at all.

The causes of psychotic symptoms in DLB are not fully understood, but dopamine dysfunction likely is involved in hallucinations, delusions, and agitation, and serotonin dysfunction may be associated with depression and anxiety. Rapid eye movement (REM) sleep/wakefulness dysregulation, in which the dream imagery of REM sleep may occur during wakefulness, also has been proposed as a mechanism for visual hallucinations in DLB.²² In DLB, psychotic symptoms occur early and are a hallmark of this illness, whereas in AD they usually occur in the middle to late stages of the disease.

Motor symptoms. In AD, extrapyramidal symptoms (EPS) are common later in the disease, are strongly correlated with disease severity, and are a strong, independent predictor of depression severity.²³ EPS are more common in DLB than in AD²⁴ and DLB patients are at higher risk of developing EPS even with low doses of typical antipsychotics, compared with AD patients.²⁵

Other symptoms. REM sleep behavior disorder (RBD) is characterized by enacting dreams—often violent—during REM sleep. RBD is common in DLB and many patients also have excessive daytime somnolence. Other sleep disorders in DLB include insomnia, obstructive sleep apnea, central sleep apnea, restless legs syndrome, and periodic limb movements during sleep.

In AD patients, common sleep behaviors include confusion in the early evening (“sundowning”) and frequent nighttime awakenings, often accompanied by wandering.²⁶ Orthostatic hypotension, impotence, urinary incontinence, and constipation are common in DLB. Lack of insight concerning personal cognitive, mood, and behavioral state is highly prevalent in AD patients and more common than in DLB.

Diagnostic evaluation

Because there are no definitive clinical markers for DLB, diagnosis is based on a detailed clinical and family history from the patient and a reliable informant, as well as a physical, neurologic, and mental status examination that looks for associated noncognitive symptoms, and neuropsychological evaluation. Reasons DLB may be misdiagnosed include:

• Some “core” clinical features of DLB may not appear or may overlap with AD.
• Presence and severity of concurrent AD pathology in DLB may modify the clinical presentation, with decreased rates of hallucinations and parkinsonism and increased neurofibrillary tangles.
• Failure to reliably identify fluctuations—variations in cognition and arousal, such as periods of unresponsiveness while awake (“zoning out”), excessive daytime somnolence, and disorganized speech.²⁷

Detecting and characterizing cognitive deficits in dementia patients using neuropsychological testing is important in establishing a clinical diagnosis, determining baseline levels of impairment, forming a prognosis, and initiating disease-specific treatments. Differences in neuropsychological findings in AD and DLB are outlined in Table 2.^16,28-33 Several studies have suggested using these measures to differentiate patients with DLB from those with AD.²⁰

Table 2

Diagnostic testing for Alzheimer’s disease and dementia with Lewy bodies

Evidence is insufficient to support using electroencephalographic and polysomnographic studies when initially evaluating patients with dementia. Brain CT or MRI are recommended as part of the initial evaluation of dementia patients to exclude treatable causes of dementia and help clarify the differential diagnosis. Occipital hypometabolism and hypoperfusion demonstrated on PET and SPECT imaging have high sensitivity and specificity for differentiating AD from DLB.

To diagnose DLB more consistently, look for core features of the disease, RBD, antipsychotic hypersensitivity, and decreased striatal binding at presynaptic DaT sites.¹⁵ Abnormal (low binding) DaT activity is the most reliable diagnostic marker for DLB.³⁴ Myocardial scintigraphy with MIBG is sensitive to pathologic changes of DLB before clinical expression and could overcome the difficulties of using clinical criteria alone to identify patients with DLB.³⁵ MIBG scintigraphy may be preferred to DaT scan because it is less expensive and its sensitivity and specificity to DLB are independent of the presence of parkinsonism.³⁵

For an overview of pharmacotherapy options for patients with AD or DLB, see Box 2.

Box 2

Related Resources

• Hanyu H, Sato T, Hirao K, et al. Differences in clinical course between dementia with Lewy bodies and Alzheimer’s disease. Eur J Neurol. 2009;16(2):212-217.
• Walker Z, McKeith I, Rodda J, et al. Comparison of cognitive decline between dementia with Lewy bodies and Alzheimer’s disease: a cohort study. BMJ Open. 2012;2:e000380.

Drug Brand Names

• Aripiprazole • Abilify
• Clonazepam • Klonopin
• Donepezil • Aricept
• Galantamine • Razadyne, Reminyl
• Levodopa • Dopar, Larodopa
• Memantine • Namenda
• Quetiapine • Seroquel
• Rivastigmine • Exelon

Disclosure

Drs. Bishnoi and Manepalli report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Dr. Grossberg serves as a consultant to Forest, Janssen, Novartis, and Pfizer. His department receives research funding from Novartis, Janssen, and Pfizer.

CASE: Delusions and threats

For over 20 months, Ms. I, age 48, sends a psychiatric resident letters and postcards that total approximately 3,000 pages and come from dozens of return addresses. Ms. I expresses romantic feelings toward the resident and believes that he was her physician and prescribed medications, including “mood stabilizers.” The resident never treated Ms. I; to his knowledge, he has never interacted with her.

Ms. I describes the resident’s refusal to continue treating her as “abandonment” and states that she is contemplating self-harm because of this rejection. In her letters, Ms. I admits that she was a long-term patient in a state psychiatric hospital in her home state and suffers from persistent auditory hallucinations. She also wants a romantic relationship with the resident and repeatedly threatens the resident’s female acquaintances and former romantic partners whose relationships she had surmised from news articles available on the Internet. Ms. I also threatens to strangle the resident. The resident sends her multiple written requests that she cease contact, but they are not acknowledged.

The authors’ observations

Stalking—repeated, unwanted attention or communication that would cause a reasonable person fear—is a serious threat for many psychiatric clinicians.¹ Prevalence rates among mental health care providers range from 3% to 21%.^2,3 Most stalkers have engaged in previous stalking behavior.³

Being stalked is highly distressing,⁴ and mental health professionals often do not reveal such experiences to colleagues.⁵ Irrational feelings of guilt or embarrassment, such as being thought to have poorly managed interactions with the stalker, often motivate a self-imposed silence (Table 1).⁶ This isolation may foster anxiety, interfere with receiving problem-solving advice, and increase physical vulnerability. In the case involving Ms. I, the psychiatric resident’s primary responsibility is safeguarding his own physical and psychological welfare.

Clinicians who work in a hospital or other institutional setting who are being stalked should inform their supervisors and the facility’s security personnel. Security personnel may be able to gather data about the stalker, decrease the stalker’s ability to communicate with the victim, and reduce unwanted physical access to the victim by distributing a photo of the stalker or installing a camera or receptionist-controlled door lock in patient entryways. Security personnel also may collaborate with local law enforcement. Having a third party respond to a stalker’s aggressive behavior—rather than the victim responding directly—avoids rewarding the stalker, which may generate further unwanted contact.⁷ Any intervention by the victim may increase the risk of violence, creating an “intervention dilemma.” Resnick⁸ argues that before deciding how best to address the stalker’s behavior, a stalking victim must “first separate the risk of continued stalking from the risk that the stalker will commit a violent act.”

Mental health professionals in private practice who are being stalked should consider retaining an attorney. An attorney often can maintain privacy of communications regarding the stalker via the attorney-client and attorney-work product privileges, which may help during legal proceedings.

Table 1

Factors that can impede psychiatrists from reporting stalking

RESPONSE: Involving police

Over 2 months, Ms. I phones the resident’s home 105 times (the resident screens the calls). During 1 call, she states that she is hidden in a closet in her home and will hurt herself unless the resident “resumes” her psychiatric care. The resident contacts police in his city and Ms. I’s community, but authorities are reluctant to act when he acknowledges that he is not Ms. I’s psychiatrist and does not know her. Police officers in Ms. I’s hometown tell the resident no one answered the door when they visited her home. They state that they would enter the residence forcibly only if Ms. I’s physician or a family member asked them to do so, and because the resident admits that he is not her psychiatrist, they cannot take further action. Ms. I leaves the resident a phone message several hours later to inform him she is safe.

The authors’ observations

Stalking-induced countertransference responses may lead a psychiatrist to unwittingly place himself in harm’s way. For example, intense rage at a stalker’s request for treatment may generate guilt that motivates the psychiatrist to agree to treat the stalker. Feelings of helplessness may produce a frantic desire to do something even when such activity is ill-advised. Psychiatrists may develop a tolerance for antisocial or threatening behavior—which is common in mental health settings—and could accept unnecessary risks.

A psychiatrist who is being stalked may be able to assist a mentally ill stalker in a way that does not create a duty to treat and does not expose the psychiatrist to harm, such as contacting a mobile crisis intervention team, a mental health professional who recently treated the stalker, a family member of the stalker, or law enforcement personnel. A psychiatrist who is thrust from the role of helper to victim and must protect his or her own well-being instead of attending to a patient’s welfare is prone to suffer substantial countertransference distress.

The situation with Ms. I was particularly challenging because the resident did not know her complete history and therefore had little information to gauge how likely she was to act on her aggressive threats. Factors that predict future violence include:

• a history of violence
• significant prior criminality
• young age at first arrest
• concomitant substance abuse
• male sex.⁹

Unfortunately, other than sex, this data regarding Ms. I could not be readily obtained.

A psychiatrist’s duty

Although sympathetic to his stalker’s distress, the resident did not want to treat this woman, nor was he ethically or legally obligated to do so. An individual’s wish to be treated by a particular psychiatrist does not create a duty for the psychiatrist to satisfy this wish.¹⁰ State-based “Good Samaritan” laws encourage physicians to assist those in acute need by shielding them from liability, as long as they reasonably act within the scope of their expertise.¹¹ However, they do not require a physician to care for an individual in acute need. A delusional wish for treatment or a false belief of already being in treatment does not create a duty to care for a person.

OUTCOME: Seeking help

Ms. I’s phone calls and letters continue. The resident discusses the situation with his associate residency director, who refers him to the hospital’s legal and investigative staffs. Based on advice from the hospital’s private investigator, the resident sends Ms. I a formal “cease and desist” letter that threatens her with legal action and possible jail time. The staff at the front desk of the clinic where the resident works and the hospital’s security department are instructed to watch for a visitor with Ms. I’s name and description, although the hospital’s investigator is unable to obtain a photograph of her. Shortly after the resident sends the letter, Ms. I ceases communication.

The authors’ observations

This case is unusual because most stalking victims know their stalkers. Identifying a stalker’s motivation can be helpful in formulating a risk assessment. One classification system recognizes 5 categories of stalkers: rejected, intimacy seeking, incompetent, resentful, and predatory (Table 2).¹ Rejected stalkers appear to pose the greatest risk of violence and homicide.⁸ However, all stalkers may pose a risk of violence and therefore all stalking behavior should be treated seriously.

Table 2

Classification of stalkers

Responding to a stalker

The approach should be tailored to the stalker’s characteristics.¹² Silence—ie, lack of acknowledgement of a stalker’s intrusions—is one tactic.¹³ Consistent and persistent lack of engagement may bore the stalker, but also may provoke frustration or narcissistic or paranoia-fueled rage, and increased efforts to interact with the mental health professional. Other responses include:

• obtaining a protection or restraining order
• promoting the stalker’s participation in adversarial civil litigation, such as a lawsuit
• issuing verbal counterthreats.

Restraining orders are controversial and assessments of their effectiveness vary.¹⁴ How well a restraining order works may depend on the stalker’s:

• ability to appreciate reality, and how likely he or she is to experience anxiety when confronted with adverse consequences of his or her actions
• how consistently, rapidly, and harshly the criminal justice system responds to violations of restraining orders.

Restraining orders also may provide the victim a false sense of security.¹⁵ One of her letters revealed that Ms. I violated a criminal plea arrangement years earlier, which suggests she was capable of violating a restraining order.

Litigation. A stalker may initiate civil litigation against the victim to feel that he or she has an impact on the victim, which may reduce the stalker’s risk of violence if he or she is emotionally engaged in the litigation. Based on the authors’ experience, as long as the stalker is talking, he or she generally is less likely to act out violently and terminate a satisfying process. Adversarial civil litigation could give a stalker the opportunity to be “close” to the victim and a means of expressing aggressive wishes. The benefit of litigation lasts only as long as the case persists and the stalker believes he or she may prevail. In one of her letters, Ms. I bragged that she had represented herself as a pro se litigant in a complex civil matter, suggesting that she might be constructively channeled into litigation.

Promoting litigation carries significant risk.¹⁶ Being a defendant in pro se litigation may be emotionally and financially stressful. This approach may be desirable if the psychiatrist’s institution is willing to offer substantial support. For example, an institution may provide legal assistance—including helping to defray the cost of litigation—and litigation-related scheduling flexibility. An attorney may serve as a boundary between the victim and the pro se litigant’s sometimes ceaseless, time-devouring, anxiety-inducing legal maneuvers.

Counterthreats. Warning a stalker that he or she will face severe civil and criminal consequences if his or her behavior continues can make clear that his or her conduct is unacceptable.¹⁷ Such warnings may be delivered verbally or in writing by a legal representative, law enforcement personnel, a private security agent, or the victim.

Issuing a counterthreat can be risky. Stalkers with antisocial or narcissistic personality features may perceive a counterthreat as narcissistically diminishing, and to save face will escalate their stalking in retaliation. Avoid counterthreats if you believe the stalker might be psychotic because destabilizing such an individual—such as by precipitating a short psychotic episode—may increase unpredictability and diminish their responsive to interventions.

Ms. I’s contact with the resident lasted approximately 20 months, slightly less than the average 26 months reported in a survey of mental health professionals.³ Because stalkers are unpredictable, the psychiatric resident remains cautious.

Related Resources

• National Center for Victims of Crime. Stalking resource center. www.victimsofcrime.org/our-programs/stalking-resource-center.
• Mullen PE, Pathé M, Purcell R. Stalkers and their victims. New York, NY: Cambridge University Press; 2009.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: Delusions and threats

For over 20 months, Ms. I, age 48, sends a psychiatric resident letters and postcards that total approximately 3,000 pages and come from dozens of return addresses. Ms. I expresses romantic feelings toward the resident and believes that he was her physician and prescribed medications, including “mood stabilizers.” The resident never treated Ms. I; to his knowledge, he has never interacted with her.

Ms. I describes the resident’s refusal to continue treating her as “abandonment” and states that she is contemplating self-harm because of this rejection. In her letters, Ms. I admits that she was a long-term patient in a state psychiatric hospital in her home state and suffers from persistent auditory hallucinations. She also wants a romantic relationship with the resident and repeatedly threatens the resident’s female acquaintances and former romantic partners whose relationships she had surmised from news articles available on the Internet. Ms. I also threatens to strangle the resident. The resident sends her multiple written requests that she cease contact, but they are not acknowledged.

The authors’ observations

Stalking—repeated, unwanted attention or communication that would cause a reasonable person fear—is a serious threat for many psychiatric clinicians.¹ Prevalence rates among mental health care providers range from 3% to 21%.^2,3 Most stalkers have engaged in previous stalking behavior.³

Being stalked is highly distressing,⁴ and mental health professionals often do not reveal such experiences to colleagues.⁵ Irrational feelings of guilt or embarrassment, such as being thought to have poorly managed interactions with the stalker, often motivate a self-imposed silence (Table 1).⁶ This isolation may foster anxiety, interfere with receiving problem-solving advice, and increase physical vulnerability. In the case involving Ms. I, the psychiatric resident’s primary responsibility is safeguarding his own physical and psychological welfare.

Clinicians who work in a hospital or other institutional setting who are being stalked should inform their supervisors and the facility’s security personnel. Security personnel may be able to gather data about the stalker, decrease the stalker’s ability to communicate with the victim, and reduce unwanted physical access to the victim by distributing a photo of the stalker or installing a camera or receptionist-controlled door lock in patient entryways. Security personnel also may collaborate with local law enforcement. Having a third party respond to a stalker’s aggressive behavior—rather than the victim responding directly—avoids rewarding the stalker, which may generate further unwanted contact.⁷ Any intervention by the victim may increase the risk of violence, creating an “intervention dilemma.” Resnick⁸ argues that before deciding how best to address the stalker’s behavior, a stalking victim must “first separate the risk of continued stalking from the risk that the stalker will commit a violent act.”

Mental health professionals in private practice who are being stalked should consider retaining an attorney. An attorney often can maintain privacy of communications regarding the stalker via the attorney-client and attorney-work product privileges, which may help during legal proceedings.

Table 1

Factors that can impede psychiatrists from reporting stalking

RESPONSE: Involving police

Over 2 months, Ms. I phones the resident’s home 105 times (the resident screens the calls). During 1 call, she states that she is hidden in a closet in her home and will hurt herself unless the resident “resumes” her psychiatric care. The resident contacts police in his city and Ms. I’s community, but authorities are reluctant to act when he acknowledges that he is not Ms. I’s psychiatrist and does not know her. Police officers in Ms. I’s hometown tell the resident no one answered the door when they visited her home. They state that they would enter the residence forcibly only if Ms. I’s physician or a family member asked them to do so, and because the resident admits that he is not her psychiatrist, they cannot take further action. Ms. I leaves the resident a phone message several hours later to inform him she is safe.

The authors’ observations

Stalking-induced countertransference responses may lead a psychiatrist to unwittingly place himself in harm’s way. For example, intense rage at a stalker’s request for treatment may generate guilt that motivates the psychiatrist to agree to treat the stalker. Feelings of helplessness may produce a frantic desire to do something even when such activity is ill-advised. Psychiatrists may develop a tolerance for antisocial or threatening behavior—which is common in mental health settings—and could accept unnecessary risks.

A psychiatrist who is being stalked may be able to assist a mentally ill stalker in a way that does not create a duty to treat and does not expose the psychiatrist to harm, such as contacting a mobile crisis intervention team, a mental health professional who recently treated the stalker, a family member of the stalker, or law enforcement personnel. A psychiatrist who is thrust from the role of helper to victim and must protect his or her own well-being instead of attending to a patient’s welfare is prone to suffer substantial countertransference distress.

The situation with Ms. I was particularly challenging because the resident did not know her complete history and therefore had little information to gauge how likely she was to act on her aggressive threats. Factors that predict future violence include:

• a history of violence
• significant prior criminality
• young age at first arrest
• concomitant substance abuse
• male sex.⁹

Unfortunately, other than sex, this data regarding Ms. I could not be readily obtained.

A psychiatrist’s duty

Although sympathetic to his stalker’s distress, the resident did not want to treat this woman, nor was he ethically or legally obligated to do so. An individual’s wish to be treated by a particular psychiatrist does not create a duty for the psychiatrist to satisfy this wish.¹⁰ State-based “Good Samaritan” laws encourage physicians to assist those in acute need by shielding them from liability, as long as they reasonably act within the scope of their expertise.¹¹ However, they do not require a physician to care for an individual in acute need. A delusional wish for treatment or a false belief of already being in treatment does not create a duty to care for a person.

OUTCOME: Seeking help

Ms. I’s phone calls and letters continue. The resident discusses the situation with his associate residency director, who refers him to the hospital’s legal and investigative staffs. Based on advice from the hospital’s private investigator, the resident sends Ms. I a formal “cease and desist” letter that threatens her with legal action and possible jail time. The staff at the front desk of the clinic where the resident works and the hospital’s security department are instructed to watch for a visitor with Ms. I’s name and description, although the hospital’s investigator is unable to obtain a photograph of her. Shortly after the resident sends the letter, Ms. I ceases communication.

The authors’ observations

This case is unusual because most stalking victims know their stalkers. Identifying a stalker’s motivation can be helpful in formulating a risk assessment. One classification system recognizes 5 categories of stalkers: rejected, intimacy seeking, incompetent, resentful, and predatory (Table 2).¹ Rejected stalkers appear to pose the greatest risk of violence and homicide.⁸ However, all stalkers may pose a risk of violence and therefore all stalking behavior should be treated seriously.

Table 2

Classification of stalkers

Responding to a stalker

The approach should be tailored to the stalker’s characteristics.¹² Silence—ie, lack of acknowledgement of a stalker’s intrusions—is one tactic.¹³ Consistent and persistent lack of engagement may bore the stalker, but also may provoke frustration or narcissistic or paranoia-fueled rage, and increased efforts to interact with the mental health professional. Other responses include:

• obtaining a protection or restraining order
• promoting the stalker’s participation in adversarial civil litigation, such as a lawsuit
• issuing verbal counterthreats.

Restraining orders are controversial and assessments of their effectiveness vary.¹⁴ How well a restraining order works may depend on the stalker’s:

• ability to appreciate reality, and how likely he or she is to experience anxiety when confronted with adverse consequences of his or her actions
• how consistently, rapidly, and harshly the criminal justice system responds to violations of restraining orders.

Restraining orders also may provide the victim a false sense of security.¹⁵ One of her letters revealed that Ms. I violated a criminal plea arrangement years earlier, which suggests she was capable of violating a restraining order.

Litigation. A stalker may initiate civil litigation against the victim to feel that he or she has an impact on the victim, which may reduce the stalker’s risk of violence if he or she is emotionally engaged in the litigation. Based on the authors’ experience, as long as the stalker is talking, he or she generally is less likely to act out violently and terminate a satisfying process. Adversarial civil litigation could give a stalker the opportunity to be “close” to the victim and a means of expressing aggressive wishes. The benefit of litigation lasts only as long as the case persists and the stalker believes he or she may prevail. In one of her letters, Ms. I bragged that she had represented herself as a pro se litigant in a complex civil matter, suggesting that she might be constructively channeled into litigation.

Promoting litigation carries significant risk.¹⁶ Being a defendant in pro se litigation may be emotionally and financially stressful. This approach may be desirable if the psychiatrist’s institution is willing to offer substantial support. For example, an institution may provide legal assistance—including helping to defray the cost of litigation—and litigation-related scheduling flexibility. An attorney may serve as a boundary between the victim and the pro se litigant’s sometimes ceaseless, time-devouring, anxiety-inducing legal maneuvers.

Counterthreats. Warning a stalker that he or she will face severe civil and criminal consequences if his or her behavior continues can make clear that his or her conduct is unacceptable.¹⁷ Such warnings may be delivered verbally or in writing by a legal representative, law enforcement personnel, a private security agent, or the victim.

Issuing a counterthreat can be risky. Stalkers with antisocial or narcissistic personality features may perceive a counterthreat as narcissistically diminishing, and to save face will escalate their stalking in retaliation. Avoid counterthreats if you believe the stalker might be psychotic because destabilizing such an individual—such as by precipitating a short psychotic episode—may increase unpredictability and diminish their responsive to interventions.

Ms. I’s contact with the resident lasted approximately 20 months, slightly less than the average 26 months reported in a survey of mental health professionals.³ Because stalkers are unpredictable, the psychiatric resident remains cautious.

Related Resources

• National Center for Victims of Crime. Stalking resource center. www.victimsofcrime.org/our-programs/stalking-resource-center.
• Mullen PE, Pathé M, Purcell R. Stalkers and their victims. New York, NY: Cambridge University Press; 2009.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

CASE: Delusions and threats

For over 20 months, Ms. I, age 48, sends a psychiatric resident letters and postcards that total approximately 3,000 pages and come from dozens of return addresses. Ms. I expresses romantic feelings toward the resident and believes that he was her physician and prescribed medications, including “mood stabilizers.” The resident never treated Ms. I; to his knowledge, he has never interacted with her.

Ms. I describes the resident’s refusal to continue treating her as “abandonment” and states that she is contemplating self-harm because of this rejection. In her letters, Ms. I admits that she was a long-term patient in a state psychiatric hospital in her home state and suffers from persistent auditory hallucinations. She also wants a romantic relationship with the resident and repeatedly threatens the resident’s female acquaintances and former romantic partners whose relationships she had surmised from news articles available on the Internet. Ms. I also threatens to strangle the resident. The resident sends her multiple written requests that she cease contact, but they are not acknowledged.

The authors’ observations

Stalking—repeated, unwanted attention or communication that would cause a reasonable person fear—is a serious threat for many psychiatric clinicians.¹ Prevalence rates among mental health care providers range from 3% to 21%.^2,3 Most stalkers have engaged in previous stalking behavior.³

Being stalked is highly distressing,⁴ and mental health professionals often do not reveal such experiences to colleagues.⁵ Irrational feelings of guilt or embarrassment, such as being thought to have poorly managed interactions with the stalker, often motivate a self-imposed silence (Table 1).⁶ This isolation may foster anxiety, interfere with receiving problem-solving advice, and increase physical vulnerability. In the case involving Ms. I, the psychiatric resident’s primary responsibility is safeguarding his own physical and psychological welfare.

Clinicians who work in a hospital or other institutional setting who are being stalked should inform their supervisors and the facility’s security personnel. Security personnel may be able to gather data about the stalker, decrease the stalker’s ability to communicate with the victim, and reduce unwanted physical access to the victim by distributing a photo of the stalker or installing a camera or receptionist-controlled door lock in patient entryways. Security personnel also may collaborate with local law enforcement. Having a third party respond to a stalker’s aggressive behavior—rather than the victim responding directly—avoids rewarding the stalker, which may generate further unwanted contact.⁷ Any intervention by the victim may increase the risk of violence, creating an “intervention dilemma.” Resnick⁸ argues that before deciding how best to address the stalker’s behavior, a stalking victim must “first separate the risk of continued stalking from the risk that the stalker will commit a violent act.”

Mental health professionals in private practice who are being stalked should consider retaining an attorney. An attorney often can maintain privacy of communications regarding the stalker via the attorney-client and attorney-work product privileges, which may help during legal proceedings.

Table 1

Factors that can impede psychiatrists from reporting stalking

RESPONSE: Involving police

Over 2 months, Ms. I phones the resident’s home 105 times (the resident screens the calls). During 1 call, she states that she is hidden in a closet in her home and will hurt herself unless the resident “resumes” her psychiatric care. The resident contacts police in his city and Ms. I’s community, but authorities are reluctant to act when he acknowledges that he is not Ms. I’s psychiatrist and does not know her. Police officers in Ms. I’s hometown tell the resident no one answered the door when they visited her home. They state that they would enter the residence forcibly only if Ms. I’s physician or a family member asked them to do so, and because the resident admits that he is not her psychiatrist, they cannot take further action. Ms. I leaves the resident a phone message several hours later to inform him she is safe.

The authors’ observations

Stalking-induced countertransference responses may lead a psychiatrist to unwittingly place himself in harm’s way. For example, intense rage at a stalker’s request for treatment may generate guilt that motivates the psychiatrist to agree to treat the stalker. Feelings of helplessness may produce a frantic desire to do something even when such activity is ill-advised. Psychiatrists may develop a tolerance for antisocial or threatening behavior—which is common in mental health settings—and could accept unnecessary risks.

A psychiatrist who is being stalked may be able to assist a mentally ill stalker in a way that does not create a duty to treat and does not expose the psychiatrist to harm, such as contacting a mobile crisis intervention team, a mental health professional who recently treated the stalker, a family member of the stalker, or law enforcement personnel. A psychiatrist who is thrust from the role of helper to victim and must protect his or her own well-being instead of attending to a patient’s welfare is prone to suffer substantial countertransference distress.

The situation with Ms. I was particularly challenging because the resident did not know her complete history and therefore had little information to gauge how likely she was to act on her aggressive threats. Factors that predict future violence include:

• a history of violence
• significant prior criminality
• young age at first arrest
• concomitant substance abuse
• male sex.⁹

Unfortunately, other than sex, this data regarding Ms. I could not be readily obtained.

A psychiatrist’s duty

Although sympathetic to his stalker’s distress, the resident did not want to treat this woman, nor was he ethically or legally obligated to do so. An individual’s wish to be treated by a particular psychiatrist does not create a duty for the psychiatrist to satisfy this wish.¹⁰ State-based “Good Samaritan” laws encourage physicians to assist those in acute need by shielding them from liability, as long as they reasonably act within the scope of their expertise.¹¹ However, they do not require a physician to care for an individual in acute need. A delusional wish for treatment or a false belief of already being in treatment does not create a duty to care for a person.

OUTCOME: Seeking help

Ms. I’s phone calls and letters continue. The resident discusses the situation with his associate residency director, who refers him to the hospital’s legal and investigative staffs. Based on advice from the hospital’s private investigator, the resident sends Ms. I a formal “cease and desist” letter that threatens her with legal action and possible jail time. The staff at the front desk of the clinic where the resident works and the hospital’s security department are instructed to watch for a visitor with Ms. I’s name and description, although the hospital’s investigator is unable to obtain a photograph of her. Shortly after the resident sends the letter, Ms. I ceases communication.

The authors’ observations

This case is unusual because most stalking victims know their stalkers. Identifying a stalker’s motivation can be helpful in formulating a risk assessment. One classification system recognizes 5 categories of stalkers: rejected, intimacy seeking, incompetent, resentful, and predatory (Table 2).¹ Rejected stalkers appear to pose the greatest risk of violence and homicide.⁸ However, all stalkers may pose a risk of violence and therefore all stalking behavior should be treated seriously.

Table 2

Classification of stalkers

Responding to a stalker

The approach should be tailored to the stalker’s characteristics.¹² Silence—ie, lack of acknowledgement of a stalker’s intrusions—is one tactic.¹³ Consistent and persistent lack of engagement may bore the stalker, but also may provoke frustration or narcissistic or paranoia-fueled rage, and increased efforts to interact with the mental health professional. Other responses include:

• obtaining a protection or restraining order
• promoting the stalker’s participation in adversarial civil litigation, such as a lawsuit
• issuing verbal counterthreats.

Restraining orders are controversial and assessments of their effectiveness vary.¹⁴ How well a restraining order works may depend on the stalker’s:

• ability to appreciate reality, and how likely he or she is to experience anxiety when confronted with adverse consequences of his or her actions
• how consistently, rapidly, and harshly the criminal justice system responds to violations of restraining orders.

Restraining orders also may provide the victim a false sense of security.¹⁵ One of her letters revealed that Ms. I violated a criminal plea arrangement years earlier, which suggests she was capable of violating a restraining order.

Litigation. A stalker may initiate civil litigation against the victim to feel that he or she has an impact on the victim, which may reduce the stalker’s risk of violence if he or she is emotionally engaged in the litigation. Based on the authors’ experience, as long as the stalker is talking, he or she generally is less likely to act out violently and terminate a satisfying process. Adversarial civil litigation could give a stalker the opportunity to be “close” to the victim and a means of expressing aggressive wishes. The benefit of litigation lasts only as long as the case persists and the stalker believes he or she may prevail. In one of her letters, Ms. I bragged that she had represented herself as a pro se litigant in a complex civil matter, suggesting that she might be constructively channeled into litigation.

Promoting litigation carries significant risk.¹⁶ Being a defendant in pro se litigation may be emotionally and financially stressful. This approach may be desirable if the psychiatrist’s institution is willing to offer substantial support. For example, an institution may provide legal assistance—including helping to defray the cost of litigation—and litigation-related scheduling flexibility. An attorney may serve as a boundary between the victim and the pro se litigant’s sometimes ceaseless, time-devouring, anxiety-inducing legal maneuvers.

Counterthreats. Warning a stalker that he or she will face severe civil and criminal consequences if his or her behavior continues can make clear that his or her conduct is unacceptable.¹⁷ Such warnings may be delivered verbally or in writing by a legal representative, law enforcement personnel, a private security agent, or the victim.

Issuing a counterthreat can be risky. Stalkers with antisocial or narcissistic personality features may perceive a counterthreat as narcissistically diminishing, and to save face will escalate their stalking in retaliation. Avoid counterthreats if you believe the stalker might be psychotic because destabilizing such an individual—such as by precipitating a short psychotic episode—may increase unpredictability and diminish their responsive to interventions.

Ms. I’s contact with the resident lasted approximately 20 months, slightly less than the average 26 months reported in a survey of mental health professionals.³ Because stalkers are unpredictable, the psychiatric resident remains cautious.

Related Resources

• National Center for Victims of Crime. Stalking resource center. www.victimsofcrime.org/our-programs/stalking-resource-center.
• Mullen PE, Pathé M, Purcell R. Stalkers and their victims. New York, NY: Cambridge University Press; 2009.

Disclosure

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products.

Mrs. A, age 68, has a 40-year history of schizoaffective disorder with comorbid anxiety disorder not otherwise specified, type 2 diabetes mellitus, and hypertension. She takes furosemide, 40 mg/d, lisinopril, 20 mg/d, and metformin, 2,000 mg/d, for hypertension and diabetes; lorazepam, 1.5 mg/d, and paroxetine, 40 mg/d, for anxiety; and quetiapine extended release, 800 mg/d, for psychotic features and mood dysregulation with schizoaffective disorder. Mrs. A’s husband died 5 years ago and she lives alone in a senior care facility. Mrs. A uses a weekly pill reminder box because her residential facility does not monitor medication adherence. She sees her psychiatrist once a month and her primary care provider every 3 months. She has no history of illicit drug, alcohol, or tobacco use.

Two weeks ago, Mrs. A was found leaning against the wall in a hallway, complaining of dizziness and disorientation, and unable to find her way back to her apartment. In the emergency department, her serum potassium is low (3.0 mEq/L; normal range: 3.5 to 5.0), fasting glucose is elevated (110 mg/dL; range: 65 to 99), and ECG reveals a prolonged QTc interval of 530 milliseconds. Before this episode, Mrs. A had been medically stable without mood or psychotic symptoms, although her daughter reported medication self-administration was becoming difficult.

Exposure to psychotropics carries a risk of QTc prolongation. The QT interval is an ECG measure of ventricular depolarization and repolarization. The QTc designation indicates a correction for heart rate with increasing heart rate correlating with a shorter QT interval. Readings of 440 milliseconds are considered normal.¹ QTc prolongation is defined as >450 milliseconds for men and >470 milliseconds for women.² An increase in the QT interval is a predictor of serious cardiac events.³

Antidepressants and antipsychotics have been associated with QTc prolongation. When identifying agents that could disrupt cardiac conduction, clinicians need to consider whether the drug’s molecular structure, receptor affinity, or pharmacologic effects are most critical.² Although these may be important, patient-specific variables that increase the risk of QTc prolongation may have greater impact. These include:

• age >65
• female sex
• electrolyte imbalances (specifically low serum potassium and magnesium levels)
• high or toxic serum levels of the suspected drug
• preexisting cardiovascular impairment, such as bradycardia.^4,5

Other risk factors include concurrent use of an agent with similar cardiovascular effects or one that competes for metabolism (either enzymatic or at the binding site), physiologic limitations such as renal insufficiency, and medication changes that may increase or decrease psychotropic clearance.^4,6 Geriatric patients with dementia have an increased risk for cardiovascular-related death.^7,8

Antidepressants

Among tricyclic antidepressants, most reports of QTc prolongation involve amitriptyline and maprotiline.⁹ Risk factors include demographics (eg, female sex, age), personal or family history (congenital long QT syndrome, cardiovascular disease), and concurrent conditions or drug use, particularly those associated with QTc prolongation.³ Desipramine and nortriptyline also have been identified as high-risk agents.¹⁰

QTc prolongation has been reported with all selective serotonin reuptake inhibitors at plasma concentrations above the therapeutic level.¹¹ Fluoxetine-associated QTc prolongation was limited to cases of overdose or when additional risk factors were reported.⁴ QTc prolongation from psychotropics could increase the risk of torsades de pointes, according to an analysis of the FDA Adverse Event Reporting System.¹² In 2011, the FDA reported an increased risk of abnormal heart rhythms—including QTc prolongation—with citalopram doses >40 mg/d.¹³ Although cases of QTc prolongation with paroxetine have not been reported,¹¹ the Arizona Center for Education and Research on Therapeutics lists paroxetine with other agents that may increase the risk for QTc prolongation with concurrent use of medications that may prolong QTc interval.¹⁴ Venlafaxine doses >300 mg/d may require additional cardiac monitoring.^5,12 Data from venlafaxine poisoning case reports found a positive correlation between dose and QTc prolongation.¹⁵ In a review of toxicology database information, Wenzel-Seifert et al⁴ found extended QT interval with citalopram, fluoxetine, and venlafaxine at toxic doses or in the presence of additional risk factors such as sex, older age, or personal or family history of congenital long QT syndrome or cardiovascular disease.

Antipsychotics

Case reports, case series, and research trials have evaluated the risk of QTc prolongation with antipsychotics (Table).^1,2,4,16,17 The first-generation antipsychotics thioridazine,^4,16,18 mesoridazine,^16,18 chlorpromazine,¹⁹ and haloperidol³ warrant cardiac monitoring. The QTc prolongation effects of thioridazine and its active metabolite mesoridazine are well-documented and thioridazine-mediated QTc prolongation increases are dose-dependent.^4,18 ECG monitoring is recommended with IV haloperidol, which is used for delirium in adults.²⁰ QTc prolongation has been associated with long-term ziprasidone use more often than with risperidone, olanzapine, or quetiapine.¹⁹ Ziprasidone prolongs the QTc interval an average of 20 milliseconds,²¹ which could represent a clinically significant change. QTc prolongation for iloperidone is comparable to ziprasidone and haloperidol.²² There is some evidence that aripiprazole may shorten, rather than prolong, the QTc interval.^4,17

Cardiovascular adverse effects associated with clozapine—including QTc prolongation—are dose-dependent.³ Olanzapine prolongs QTc interval, although the mean change is less than with other agents unless other variables were present, such as:

• concomitant use of medications that may prolong QTc interval (ie, amantadine, hydroxyzine, or tamoxifen²)
• preexisting cardiovascular conduction disorders
• higher doses (>40 mg/d).^3,23

In 17 case reports of cardiac changes associated with quetiapine use, doses ranged from 100 mg/d²⁴ to an overdose of 36 g/d.²⁵ Only 1 patient death was reported secondary to overdose and preexisting dysrhythmia and hypertension.²⁶ QTc prolongation associated with risperidone was minor¹ based on oral doses in the normal therapeutic range and incidences of overdose.¹⁰ Paliperidone²⁷ and lurasidone²⁸ are associated with clinically insignificant QTc prolongation. Changes in QTc interval were positively correlated with asenapine dose, although at the highest dose of 40 mg/d, the increase was <5 milliseconds.²⁹

Mrs. A presents with a number of risk factors for QTc prolongation, including older age, female sex, and psychiatric and medical comorbidities that require medication. A pill count revealed that she was taking more than the prescribed daily doses of her medications. During the interview, Mrs. A said that if she missed her medication time, she would take them when she remembered. If she could not remember if she took her pills, she would take them again. Her physicians will explore strategies to increase medication adherence.

Table

Examples of QTc prolongation associated with select antipsychotics^a

Related Resources

• De Hert M, Detraux J, van Winkel R, et al. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat Rev Endocrinol. 2011;8(2):114-126.
• Vieweg WV, Wood MA, Fernandez A, et al. Proarrhythmic risk with antipsychotic and antidepressant drugs: implications in the elderly. Drugs Aging. 2009;26(12):997-1012.
• Sandson NB, Armstrong SC, Cozza KL. An overview of psychotropic drug-drug interactions. Psychosomatics. 2005;46(5):464-494.

Drug Brand Names

• Amantadine • Symmetrel
• Amitriptyline • Elavil
• Aripiprazole • Abilify
• Asenapine • Saphris
• Chlorpromazine • Thorazine
• Citalopram • Celexa
• Clozapine • Clozaril
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Furosemide • Lasix
• Haloperidol • Haldol
• Hydroxyzine • Atarax, Vistaril
• Iloperidone • Fanapt
• Lisinopril • Prinivil, Zestril
• Lorazepam • Ativan
• Lurasidone • Latuda
• Maprotiline • Ludiomil
• Mesoridazine • Serentil
• Metformin • Glucophage
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Paroxetine • Paxil
• Pimozide • Orap
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Tamoxifen • Nolvadex, Soltamox
• Thioridazine • Mellaril
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. No similar work by the authors is under review or in press. No funding was requested or received in conjunction with this manuscript.

Mrs. A, age 68, has a 40-year history of schizoaffective disorder with comorbid anxiety disorder not otherwise specified, type 2 diabetes mellitus, and hypertension. She takes furosemide, 40 mg/d, lisinopril, 20 mg/d, and metformin, 2,000 mg/d, for hypertension and diabetes; lorazepam, 1.5 mg/d, and paroxetine, 40 mg/d, for anxiety; and quetiapine extended release, 800 mg/d, for psychotic features and mood dysregulation with schizoaffective disorder. Mrs. A’s husband died 5 years ago and she lives alone in a senior care facility. Mrs. A uses a weekly pill reminder box because her residential facility does not monitor medication adherence. She sees her psychiatrist once a month and her primary care provider every 3 months. She has no history of illicit drug, alcohol, or tobacco use.

Two weeks ago, Mrs. A was found leaning against the wall in a hallway, complaining of dizziness and disorientation, and unable to find her way back to her apartment. In the emergency department, her serum potassium is low (3.0 mEq/L; normal range: 3.5 to 5.0), fasting glucose is elevated (110 mg/dL; range: 65 to 99), and ECG reveals a prolonged QTc interval of 530 milliseconds. Before this episode, Mrs. A had been medically stable without mood or psychotic symptoms, although her daughter reported medication self-administration was becoming difficult.

Exposure to psychotropics carries a risk of QTc prolongation. The QT interval is an ECG measure of ventricular depolarization and repolarization. The QTc designation indicates a correction for heart rate with increasing heart rate correlating with a shorter QT interval. Readings of 440 milliseconds are considered normal.¹ QTc prolongation is defined as >450 milliseconds for men and >470 milliseconds for women.² An increase in the QT interval is a predictor of serious cardiac events.³

Antidepressants and antipsychotics have been associated with QTc prolongation. When identifying agents that could disrupt cardiac conduction, clinicians need to consider whether the drug’s molecular structure, receptor affinity, or pharmacologic effects are most critical.² Although these may be important, patient-specific variables that increase the risk of QTc prolongation may have greater impact. These include:

• age >65
• female sex
• electrolyte imbalances (specifically low serum potassium and magnesium levels)
• high or toxic serum levels of the suspected drug
• preexisting cardiovascular impairment, such as bradycardia.^4,5

Other risk factors include concurrent use of an agent with similar cardiovascular effects or one that competes for metabolism (either enzymatic or at the binding site), physiologic limitations such as renal insufficiency, and medication changes that may increase or decrease psychotropic clearance.^4,6 Geriatric patients with dementia have an increased risk for cardiovascular-related death.^7,8

Antidepressants

Among tricyclic antidepressants, most reports of QTc prolongation involve amitriptyline and maprotiline.⁹ Risk factors include demographics (eg, female sex, age), personal or family history (congenital long QT syndrome, cardiovascular disease), and concurrent conditions or drug use, particularly those associated with QTc prolongation.³ Desipramine and nortriptyline also have been identified as high-risk agents.¹⁰

QTc prolongation has been reported with all selective serotonin reuptake inhibitors at plasma concentrations above the therapeutic level.¹¹ Fluoxetine-associated QTc prolongation was limited to cases of overdose or when additional risk factors were reported.⁴ QTc prolongation from psychotropics could increase the risk of torsades de pointes, according to an analysis of the FDA Adverse Event Reporting System.¹² In 2011, the FDA reported an increased risk of abnormal heart rhythms—including QTc prolongation—with citalopram doses >40 mg/d.¹³ Although cases of QTc prolongation with paroxetine have not been reported,¹¹ the Arizona Center for Education and Research on Therapeutics lists paroxetine with other agents that may increase the risk for QTc prolongation with concurrent use of medications that may prolong QTc interval.¹⁴ Venlafaxine doses >300 mg/d may require additional cardiac monitoring.^5,12 Data from venlafaxine poisoning case reports found a positive correlation between dose and QTc prolongation.¹⁵ In a review of toxicology database information, Wenzel-Seifert et al⁴ found extended QT interval with citalopram, fluoxetine, and venlafaxine at toxic doses or in the presence of additional risk factors such as sex, older age, or personal or family history of congenital long QT syndrome or cardiovascular disease.

Antipsychotics

Case reports, case series, and research trials have evaluated the risk of QTc prolongation with antipsychotics (Table).^1,2,4,16,17 The first-generation antipsychotics thioridazine,^4,16,18 mesoridazine,^16,18 chlorpromazine,¹⁹ and haloperidol³ warrant cardiac monitoring. The QTc prolongation effects of thioridazine and its active metabolite mesoridazine are well-documented and thioridazine-mediated QTc prolongation increases are dose-dependent.^4,18 ECG monitoring is recommended with IV haloperidol, which is used for delirium in adults.²⁰ QTc prolongation has been associated with long-term ziprasidone use more often than with risperidone, olanzapine, or quetiapine.¹⁹ Ziprasidone prolongs the QTc interval an average of 20 milliseconds,²¹ which could represent a clinically significant change. QTc prolongation for iloperidone is comparable to ziprasidone and haloperidol.²² There is some evidence that aripiprazole may shorten, rather than prolong, the QTc interval.^4,17

Cardiovascular adverse effects associated with clozapine—including QTc prolongation—are dose-dependent.³ Olanzapine prolongs QTc interval, although the mean change is less than with other agents unless other variables were present, such as:

• concomitant use of medications that may prolong QTc interval (ie, amantadine, hydroxyzine, or tamoxifen²)
• preexisting cardiovascular conduction disorders
• higher doses (>40 mg/d).^3,23

In 17 case reports of cardiac changes associated with quetiapine use, doses ranged from 100 mg/d²⁴ to an overdose of 36 g/d.²⁵ Only 1 patient death was reported secondary to overdose and preexisting dysrhythmia and hypertension.²⁶ QTc prolongation associated with risperidone was minor¹ based on oral doses in the normal therapeutic range and incidences of overdose.¹⁰ Paliperidone²⁷ and lurasidone²⁸ are associated with clinically insignificant QTc prolongation. Changes in QTc interval were positively correlated with asenapine dose, although at the highest dose of 40 mg/d, the increase was <5 milliseconds.²⁹

Mrs. A presents with a number of risk factors for QTc prolongation, including older age, female sex, and psychiatric and medical comorbidities that require medication. A pill count revealed that she was taking more than the prescribed daily doses of her medications. During the interview, Mrs. A said that if she missed her medication time, she would take them when she remembered. If she could not remember if she took her pills, she would take them again. Her physicians will explore strategies to increase medication adherence.

Table

Examples of QTc prolongation associated with select antipsychotics^a

Related Resources

• De Hert M, Detraux J, van Winkel R, et al. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat Rev Endocrinol. 2011;8(2):114-126.
• Vieweg WV, Wood MA, Fernandez A, et al. Proarrhythmic risk with antipsychotic and antidepressant drugs: implications in the elderly. Drugs Aging. 2009;26(12):997-1012.
• Sandson NB, Armstrong SC, Cozza KL. An overview of psychotropic drug-drug interactions. Psychosomatics. 2005;46(5):464-494.

Drug Brand Names

• Amantadine • Symmetrel
• Amitriptyline • Elavil
• Aripiprazole • Abilify
• Asenapine • Saphris
• Chlorpromazine • Thorazine
• Citalopram • Celexa
• Clozapine • Clozaril
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Furosemide • Lasix
• Haloperidol • Haldol
• Hydroxyzine • Atarax, Vistaril
• Iloperidone • Fanapt
• Lisinopril • Prinivil, Zestril
• Lorazepam • Ativan
• Lurasidone • Latuda
• Maprotiline • Ludiomil
• Mesoridazine • Serentil
• Metformin • Glucophage
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Paroxetine • Paxil
• Pimozide • Orap
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Tamoxifen • Nolvadex, Soltamox
• Thioridazine • Mellaril
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. No similar work by the authors is under review or in press. No funding was requested or received in conjunction with this manuscript.

Mrs. A, age 68, has a 40-year history of schizoaffective disorder with comorbid anxiety disorder not otherwise specified, type 2 diabetes mellitus, and hypertension. She takes furosemide, 40 mg/d, lisinopril, 20 mg/d, and metformin, 2,000 mg/d, for hypertension and diabetes; lorazepam, 1.5 mg/d, and paroxetine, 40 mg/d, for anxiety; and quetiapine extended release, 800 mg/d, for psychotic features and mood dysregulation with schizoaffective disorder. Mrs. A’s husband died 5 years ago and she lives alone in a senior care facility. Mrs. A uses a weekly pill reminder box because her residential facility does not monitor medication adherence. She sees her psychiatrist once a month and her primary care provider every 3 months. She has no history of illicit drug, alcohol, or tobacco use.

Two weeks ago, Mrs. A was found leaning against the wall in a hallway, complaining of dizziness and disorientation, and unable to find her way back to her apartment. In the emergency department, her serum potassium is low (3.0 mEq/L; normal range: 3.5 to 5.0), fasting glucose is elevated (110 mg/dL; range: 65 to 99), and ECG reveals a prolonged QTc interval of 530 milliseconds. Before this episode, Mrs. A had been medically stable without mood or psychotic symptoms, although her daughter reported medication self-administration was becoming difficult.

Exposure to psychotropics carries a risk of QTc prolongation. The QT interval is an ECG measure of ventricular depolarization and repolarization. The QTc designation indicates a correction for heart rate with increasing heart rate correlating with a shorter QT interval. Readings of 440 milliseconds are considered normal.¹ QTc prolongation is defined as >450 milliseconds for men and >470 milliseconds for women.² An increase in the QT interval is a predictor of serious cardiac events.³

Antidepressants and antipsychotics have been associated with QTc prolongation. When identifying agents that could disrupt cardiac conduction, clinicians need to consider whether the drug’s molecular structure, receptor affinity, or pharmacologic effects are most critical.² Although these may be important, patient-specific variables that increase the risk of QTc prolongation may have greater impact. These include:

• age >65
• female sex
• electrolyte imbalances (specifically low serum potassium and magnesium levels)
• high or toxic serum levels of the suspected drug
• preexisting cardiovascular impairment, such as bradycardia.^4,5

Other risk factors include concurrent use of an agent with similar cardiovascular effects or one that competes for metabolism (either enzymatic or at the binding site), physiologic limitations such as renal insufficiency, and medication changes that may increase or decrease psychotropic clearance.^4,6 Geriatric patients with dementia have an increased risk for cardiovascular-related death.^7,8

Antidepressants

Among tricyclic antidepressants, most reports of QTc prolongation involve amitriptyline and maprotiline.⁹ Risk factors include demographics (eg, female sex, age), personal or family history (congenital long QT syndrome, cardiovascular disease), and concurrent conditions or drug use, particularly those associated with QTc prolongation.³ Desipramine and nortriptyline also have been identified as high-risk agents.¹⁰

QTc prolongation has been reported with all selective serotonin reuptake inhibitors at plasma concentrations above the therapeutic level.¹¹ Fluoxetine-associated QTc prolongation was limited to cases of overdose or when additional risk factors were reported.⁴ QTc prolongation from psychotropics could increase the risk of torsades de pointes, according to an analysis of the FDA Adverse Event Reporting System.¹² In 2011, the FDA reported an increased risk of abnormal heart rhythms—including QTc prolongation—with citalopram doses >40 mg/d.¹³ Although cases of QTc prolongation with paroxetine have not been reported,¹¹ the Arizona Center for Education and Research on Therapeutics lists paroxetine with other agents that may increase the risk for QTc prolongation with concurrent use of medications that may prolong QTc interval.¹⁴ Venlafaxine doses >300 mg/d may require additional cardiac monitoring.^5,12 Data from venlafaxine poisoning case reports found a positive correlation between dose and QTc prolongation.¹⁵ In a review of toxicology database information, Wenzel-Seifert et al⁴ found extended QT interval with citalopram, fluoxetine, and venlafaxine at toxic doses or in the presence of additional risk factors such as sex, older age, or personal or family history of congenital long QT syndrome or cardiovascular disease.

Antipsychotics

Case reports, case series, and research trials have evaluated the risk of QTc prolongation with antipsychotics (Table).^1,2,4,16,17 The first-generation antipsychotics thioridazine,^4,16,18 mesoridazine,^16,18 chlorpromazine,¹⁹ and haloperidol³ warrant cardiac monitoring. The QTc prolongation effects of thioridazine and its active metabolite mesoridazine are well-documented and thioridazine-mediated QTc prolongation increases are dose-dependent.^4,18 ECG monitoring is recommended with IV haloperidol, which is used for delirium in adults.²⁰ QTc prolongation has been associated with long-term ziprasidone use more often than with risperidone, olanzapine, or quetiapine.¹⁹ Ziprasidone prolongs the QTc interval an average of 20 milliseconds,²¹ which could represent a clinically significant change. QTc prolongation for iloperidone is comparable to ziprasidone and haloperidol.²² There is some evidence that aripiprazole may shorten, rather than prolong, the QTc interval.^4,17

Cardiovascular adverse effects associated with clozapine—including QTc prolongation—are dose-dependent.³ Olanzapine prolongs QTc interval, although the mean change is less than with other agents unless other variables were present, such as:

• concomitant use of medications that may prolong QTc interval (ie, amantadine, hydroxyzine, or tamoxifen²)
• preexisting cardiovascular conduction disorders
• higher doses (>40 mg/d).^3,23

In 17 case reports of cardiac changes associated with quetiapine use, doses ranged from 100 mg/d²⁴ to an overdose of 36 g/d.²⁵ Only 1 patient death was reported secondary to overdose and preexisting dysrhythmia and hypertension.²⁶ QTc prolongation associated with risperidone was minor¹ based on oral doses in the normal therapeutic range and incidences of overdose.¹⁰ Paliperidone²⁷ and lurasidone²⁸ are associated with clinically insignificant QTc prolongation. Changes in QTc interval were positively correlated with asenapine dose, although at the highest dose of 40 mg/d, the increase was <5 milliseconds.²⁹

Mrs. A presents with a number of risk factors for QTc prolongation, including older age, female sex, and psychiatric and medical comorbidities that require medication. A pill count revealed that she was taking more than the prescribed daily doses of her medications. During the interview, Mrs. A said that if she missed her medication time, she would take them when she remembered. If she could not remember if she took her pills, she would take them again. Her physicians will explore strategies to increase medication adherence.

Table

Examples of QTc prolongation associated with select antipsychotics^a

Related Resources

• De Hert M, Detraux J, van Winkel R, et al. Metabolic and cardiovascular adverse effects associated with antipsychotic drugs. Nat Rev Endocrinol. 2011;8(2):114-126.
• Vieweg WV, Wood MA, Fernandez A, et al. Proarrhythmic risk with antipsychotic and antidepressant drugs: implications in the elderly. Drugs Aging. 2009;26(12):997-1012.
• Sandson NB, Armstrong SC, Cozza KL. An overview of psychotropic drug-drug interactions. Psychosomatics. 2005;46(5):464-494.

Drug Brand Names

• Amantadine • Symmetrel
• Amitriptyline • Elavil
• Aripiprazole • Abilify
• Asenapine • Saphris
• Chlorpromazine • Thorazine
• Citalopram • Celexa
• Clozapine • Clozaril
• Desipramine • Norpramin
• Fluoxetine • Prozac
• Furosemide • Lasix
• Haloperidol • Haldol
• Hydroxyzine • Atarax, Vistaril
• Iloperidone • Fanapt
• Lisinopril • Prinivil, Zestril
• Lorazepam • Ativan
• Lurasidone • Latuda
• Maprotiline • Ludiomil
• Mesoridazine • Serentil
• Metformin • Glucophage
• Nortriptyline • Pamelor
• Olanzapine • Zyprexa
• Paliperidone • Invega
• Paroxetine • Paxil
• Pimozide • Orap
• Quetiapine • Seroquel
• Risperidone • Risperdal
• Tamoxifen • Nolvadex, Soltamox
• Thioridazine • Mellaril
• Venlafaxine • Effexor
• Ziprasidone • Geodon

Disclosures

The authors report no financial relationship with any company whose products are mentioned in this article or with manufacturers of competing products. No similar work by the authors is under review or in press. No funding was requested or received in conjunction with this manuscript.