<< Back  

 Can't open the PDF? Click here for help.


Capgras delusion with violent behavior in Alzheimer dementia: Case analysis with literature review

Kenneth R. Kaufman, AM, MD

Departments of Psychiatry, Neurology, and Anesthesiology, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA

Neil B. Newman, BA

Department of Psychiatry, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA

Alveena Dawood, BS, MBS

Department of Psychiatry, Rutgers Robert Wood Johnson Medical School, New Brunswick, New Jersey, USA

BACKGROUND: Capgras delusion (CD) has multiple etiologies including neurodegenerative disorders and can be associated with violent behavior. CD is a common complication of Alzheimer dementia (AD); however, CD with violent behavior is uncommon in AD. We report escalating violent behavior by a patient with advanced AD and CD who presented to the emergency department (ED) and required admission to an academic medical center.

METHODS: Case analysis with PubMed literature review.

RESULTS: A 75-year-old male with a 13-year history of progressive AD, asymptomatic bipolar disorder, chronic kidney disease, hypertension, hyperlipidemia, and benign prostatic hypertrophy presented to the ED with recurrent/escalating violence toward his wife, whom he considered an impostor. His psychotropic regimen included potentially inappropriate medications (PIMs) for geriatric/AD patients—topiramate/amitriptyline/chlordiazepoxide/olanzapine—that are associated with delirium, cognitive decline, dementia, and mortality. Renal dosing for topiramate, reduction in PIMs/anticholinergic burden, and substituting haloperidol for olanzapine resolved his violent behavior and CD.

CONCLUSIONS: CD in AD is a risk factor for violent behavior. As the geriatric population in the United States grows, CD in patients with AD may present more frequently in the ED, requiring proper treatment. Pharmacovigilance is necessary to minimize PIMs in geriatric/AD patients. Clinicians and other caregivers require further education to appropriately address CD in AD.

KEYWORDS: Capgras delusion, Alzheimer dementia, violent behavior, neurodegenerative disorders, bipolar disorder, potentially inappropriate medications, anticholinergic burden, topiramate, benzodiazepines, amitriptyline, olanzapine, haloperidol, sertraline, trazodone, pharmacovigilance, prevalence, caregiver, ethics, education



The Capgras delusion (CD) is a disorder of misidentification initially described by Capgras and Reboul-Lachaux,1 in which the affected individual believes that a person, usually family member or friend, has been replaced by an impostor. Since its original description in 1923, clinicians have observed CD in multiple conditions including, but not limited to, schizophrenia, depression, dementia, delirium, multiple sclerosis, seizure disorders, central nervous system tumors, traumatic brain injury, and cerebral vascular accidents.2-14 In some cases, Capgras delusions have resolved when the primary pathology was treated.9,11,15

Four studies address prevalence rates of Capgras delusion in dementia and neurodegenerative disorders. The first study reported 10.1% of 158 outpatients with Alzheimer dementia (AD) had CD16; the second, larger study of 771 outpatients with AD reported only 4.4% had CD17; the third study, analyzing 392 patients with neurodegenerative disorders, reported a 15.8% CD prevalence in patients with AD18; the fourth study, a retrospective analysis of 47 patients with CD, noted 81% to have neurodegenerative conditions.19

Researchers have reported violent behavior in patients with CD from multiple etiologies including AD.17,20-25 The large Mizrahi study reported that violent behavior was rare during CD in AD; only 4 of the 34 patients with CD were violent toward the impostor, which represented only 0.52% prevalence of the 771 outpatients in this study.17

We report escalating and recurrent violent behavior by a patient with advanced stage AD secondary to CD, which required emergency department (ED) presentation and admission to an academic medical center.


We employed case analysis with a PubMed literature review for our study.


A 75-year-old male with a 13-year history of AD and asymptomatic bipolar disorder presented to an academic medical center ED after calling emergency medical services that an impostor female called “J” was in his home while his actual wife, whom he also referred to as “J,” was missing. This CD, which was present for 6 months with multiple calls to the police and associated with mild agitation/violence from the first episode, had increased both in frequency and intensity to the point of markedly agitated and significantly violent behaviors. Detailed collateral data from the wife revealed this to be his eighth episode of believing that she was missing while another woman had replaced her and that his more violent episodes were precipitated by her wearing new clothes, in this instance a green dress. All episodes occurred in the evening. During this episode, the patient for the first time attempted to hit his dog with a cane. On this occasion, he was so distressed by the presence of the impostor “J” that he called the police for assistance, which resulted in his ED presentation.

When seen in psychiatric consultation, the patient was pleasant and cooperative though responding to questions most often with 1-word answers. He was noted to have advanced AD, markedly decreased memory, and significantly impaired executive and occupational functioning. His Mini-Mental State Examination (MMSE) was 15/30 on 3 consecutive days confirming his moderately severe dementia, which showed progressive cognitive decline compared with a MMSE of 21/30 performed during an admission 4 months earlier. When asked about the precipitating incident, the patient remembered the other woman “J” who was at his home while his wife was missing and the green dress though he did not recall attacking his dog.

In addition to AD, the patient also had chronic kidney disease, hypertension, benign prostatic hypertrophy, and hyperlipidemia. A non-contrast head CT revealed age appropriate atrophy and small vessel ischemic disease unchanged from prior neuroimaging studies. Chest radiograph was consistent with an infiltrate and broad spectrum IV antibiotics for presumptive pneumonia were initiated prior to obtaining a culture. Pertinent abnormal laboratories included blood urea nitrogen, 29 mg/dL, and creatinine, 2.9 mg/dL.

At the time of the initial psychiatric consultation, the patient’s psychotropic regimen included topiramate, 100 mg twice a day; amitriptyline, 20 mg once in the morning and 30 mg nightly at bedtime; trazodone, 100 mg nightly at bedtime; chlordiazepoxide, 50 mg once in the morning, and 75 mg nightly at bedtime; olanzapine, 15 mg nightly at bedtime; and sertraline, 50 mg once in the morning. We tapered these psychotropics during his admission and his discharge regimen included topiramate, 100 mg nightly at bedtime; amitriptyline, 30 mg nightly at bedtime; trazodone, 100 mg nightly at bedtime; chlordiazepoxide, 25 mg every 6 hours as needed; haloperidol, 2 mg every 8 hours; and sertraline, 50 mg nightly at bedtime.

During this admission, with olanzapine changed to haloperidol, the patient did not display his CD and was nonviolent. After medical stabilization, the patient was transferred to an inpatient psychiatric hospital and thereafter he was lost to follow-up.


This unique case raises specific issues pertinent to this patient, CD in AD, and the treatment of patients with AD in general.

CD is common in patients with AD with several studies suggesting a prevalence of 4.4% to 15.8%.16-18 In the largest study of 771 outpatients with AD, 4 of 34 patients with CD were violent toward the impostor, or 11.8% within the CD subpopulation, but only 0.52% within the total study population.17 As such, violence in AD patients secondary to CD is an uncommon finding, which may not be expected by clinicians but requires urgent intervention.

CD is associated with progressive AD. One study, for example, found that AD patients with a mean MMSE of 10.4 +/- 4.7 had CD whereas those AD patients with a mean score of 17.4 +/- 7.1 did not.16 The development of CD with decreasing MMSE was noted in this case.

Optimally, families desire to maintain relatives with neurodegenerative disease, including AD, in the home setting for it is a non-threatening environment that minimizes confusion. However, since violence in CD is often toward the primary caregiver, this is especially problematic when the patient is residing at home; for when does the relative decide intervention is required? Our case exemplifies this dilemma, for the patient only presented after the eighth incidence of violence though he had 3 medical and 6 psychiatric admissions during the previous year associated with dementia with delirium without violent behavior. Further, the increased frequency and intensity of violence noted in this case mirrors the disinhibition associated with progressive AD.

Hospitalized AD patients with behavioral disturbances often present with comorbid medical conditions and associated medications, which may complicate the clinical picture as there may be a delirium superimposed on the AD. Polypharmacy is common in the geriatric population resulting in an increased risk of adverse effects from medications, some of which are contraindicated in AD patents. Polypharmacy is dependent on the prescribing physician; in one large study, general practitioners had the greatest rate of prescribing psychotropics to mentally ill geriatric patients, while gerontologists and psychiatrists had the lowest rates.26

During several previous medical admissions for this patient associated with non-violent behaviors within the 9 months preceding this admission for violent behavior, the psychiatric consultation-liaison (CL) service had minimized inappropriate medications (topiramate, amitriptyline, chlordiazepoxide, and olanzapine) in the inpatient setting only to have the outpatient general practitioner resume prior psychotropics. The recurrent admissions secondary to inappropriate outpatient prescriptions might have been minimized had there been a team approach in which the CL service participated in this patient’s outpatient care.

Recent studies have reported that potentially inappropriate medications (PIMs) are prescribed frequently in both geriatric patients and patients with mild to moderate Alzheimer disease, 53.6% and 46.8% respectively.27,28 Polypharmacy, which occurred in this case report, was noted to be a significant risk factor for PIMs. A review of the admission psychotropics prescribed in this case suggests that these medications may have negatively impacted his presentation. That observation suggests several specific considerations for AD patients and geriatric patients in general.

First, though low dose topiramate has been reported to improve the behavioral disorders associated with dementia, dose-dependent cognitive deficits are significant adverse effects associated with topiramate.29-31 Further, renal dosing for topiramate is required for patients with an estimated glomerular filtration rate (GFR) below 70 mL/min/1.73m2.32 In this case, with a creatinine level of 2.9, the estimated GFR was 21 mL/min/1.73 m2.

Second, benzodiazepines are relatively contraindicated in the geriatric population secondary to adverse effects including sedation, falls, and cognitive impairment.28,33-35 A recent population-based prospective study found benzodiazepine use to be a significant risk factor for dementia in the geriatric population.36 Benzodiazepines with long half-life and active metabolites, such as chlordiazepoxide, are of particular concern. The literature recommends 5 mg/d diazepam equivalents33; in this case, the admission chlordiazepoxide, 125 mg/d, equated to approximately 10 times the recommended geriatric total daily dose.

Third, although a recent animal study suggests that amitriptyline is a procognitive agent in aged and cognitively impaired mice,37 multiple human studies have emphasized anticholinergic agents as PIMs and focused on the development of delirium, cognitive decline, dementia, and even mortality with increased anticholinergic burden.28,38-42 Amitriptyline, a tricyclic antidepressant, has a very high anticholinergic burden with severe negative cognitive effects and should be avoided in geriatric patients.42-44

Fourth, atypical antipsychotics, which are used frequently to treat behavioral and psychological symptoms of dementia, caused increased cognitive decline in the Clinical Antipsychotic Trials of Intervention Effectiveness-Alzheimer’s Disease study (CATIE-AD) comparing the pooled treatment group (olanzapine, quetiapine, and risperidone) to placebo in AD patients with psychosis or agitated/aggressive behaviors.45 CATIE-AD also revealed that olanzapine AD patients had a statically greater decline in MMSE and cognitive summary compared with placebo.45 A study that addressed AD patients without psychosis, depression, or agitated/aggressive behaviors reported greater cognitive impairment (ADAS-Cog and MMSE) in AD patients treated with olanzapine, when compared with placebo.46 Olanzapine has a severe anticholinergic burden and should be used with caution in geriatric patients and patients with dementia.42-44 Pertinent to this case is the increased risk of pneumonia associated with atypical antipsychotics compared with typical antipsychotics in the geriatric population.47

Best practice pharmaceutical treatment maximizes therapeutic response while minimizing adverse drug effects. In this case, PIMs were either decreased in dosage or discontinued with an alternative psychotropic prescribed. Specifically, topiramate, chlordiazepoxide, and amitriptyline were decreased in total daily dose and haloperidol was substituted for olanzapine. Haloperidol has a low anticholinergic burden,38,42-44 but may increase the risk of mortality when compared with other antipsychotics.48 Haloperidol was used specifically for immediate stabilization preparatory to inpatient psychiatric transfer. Clinicians need to consider the patient’s psychiatric symptoms and diagnoses for appropriate psychotropic management; an understanding of renal and hepatic status is important for appropriate dosing; finally, medical diagnoses and/or medications used to treat such diagnoses may either preclude specific psychotropics or result in significant drug-drug interactions. Clearly, pharmacovigilance is important to the treatment of geriatric patients and patients with AD.49

We hypothesize that the severity of violence in this episode of CD was secondary to the progression of his AD with superimposed delirium (comorbid medical conditions and anticholinergic burden). In the geriatric population, it is difficult to determine what causes violent behaviors, and specifically to differentiate among CD, Alzheimer dementia, and delirium as possible etiologies.

There are specific limitations to this paper. As a case report (N = 1), the findings cannot be generalized. This case was complicated by the presence of medical comorbid conditions—chronic kidney disease and presumptive pneumonia—which prevented direct comparison to all CD cases. We did not have longitudinal outpatient neuropsychiatric assessments, which would have helped address his advancing AD (MMSE 15/30) with the development of CD and increasing frequency and intensity of violence. The patient had limited neuroimaging (CT scan without contrast); more relevant neuroimaging (MRI, SPECT, and PET) was not performed. The patient did not have serial routine/quantitative EEGs, which would have assisted in determining both baseline neurophysiologic functioning and whether the patient was having a delirium that cleared over the course of his admission. Polypharmacy with use of multiple PIMs confounded interpretation of this case. The patient was lost to clinical follow-up.


CD is commonly found in patients with AD and other neurodegenerative disorders. CD in AD is a risk factor for violent behavior toward the impostor, who frequently is the caregiver. Even in the presence of violent behavior, the caregiver is often reluctant to present and appropriately address this behavior in the ED, as was demonstrated in this case. As the nation’s geriatric population grows, CD in patients with AD and other neurodegenerative disorders may present more frequently in the ED requiring increased vigilance and a team approach that includes the ED attending, psychiatric consultation, neurologic consultation, and social services. As CD has multiple etiologies, including delirium, healthcare providers need to avoid PIMs, which is a significant problem in geriatric patients and patients with AD. Increased clinical education is required for multiple specialists and caregivers to appropriately address CD in AD and to minimize associated violent behavior.

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

ACKNOWLEDGMENTS: This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.


  1. Capgras J, Reboul-Lachaux J. L’illusion des ‘sosies’ dans un délire systématisé chronique. Bulletin de la Société Clinique de Médecine Mentale. 1923;11:6–16.
  2. Morgado P, Freitas D, Carvalho J, et al. Capgras syndrome in a three-decades non-treated schizophrenia: a case report. Eur Psychiatry. 2013;28(suppl 1):1.
  3. Silva JA, Leong GB. The Capgras syndrome in paranoid schizophrenia. Psychopathology. 1992;25:147–153.
  4. Sharma I, Varma SL, Ancharaj V, et al. Capgras syndrome in depression. Indian J Psychiatry. 1989;31:93–94.
  5. Fischer C, Keeler A, Fornazzari L, et al. A rare variant of Capgras syndrome in Alzheimer’s disease. Can J Neurol Sci. 2009;36:509–511.
  6. Pies R. Capgras phenomenon delirium, and transient hepatic dysfunction. Hosp Community Psychiatry. 1982;33:382–383.
  7. Joshi D, Sharad K, Lamichhane S, et al. Capgras syndrome in postictal delirium. Psychiatry. 2010;7:37–39.
  8. Sharma A, Garuba M, Egbert M. Capgras syndrome in a patient with multiple sclerosis: a case report. Prim Care Companion J Clin Psychiatry. 2009;11:274.
  9. Kanemoto K. Periictal Capgras syndrome after clustered ictal fear: depth-electroencephalogram study. Epilepsia. 1997;38:847–850.
  10. Horikawa H, Monji A, Sasaki M, et al. Different SPECT findings before and after Capgras’ syndrome in interictal psychosis. Epilepsy Behav. 2006;9:189–192.
  11. Fennig S, Naisberg-Fennig S, Bromet E. Capgras’ syndrome with right meningioma [in Hebrew]. Harefuah. 1994;126:320–321.
  12. Mattioli F, Miozzo A, Vignolo LA. Confabulation and delusional misidentification: a four year follow-up study. Cortex. 1999;35:413–422.
  13. Dejode JM, Antonini F, Lagier P, et al. Capgras syndrome: a clinical manifestation of watershed infarct complicating the use of extracorporeal membrane oxygenation. Crit Care. 2001;5:232–235.
  14. Carota A, Dieguez S, Bogousslavsky J. Psychopathologie des accidents vasculaires cerebraux. Psychol Neuropsychiatr Vieil. 2005;3:235–249.
  15. Salviati M, Bersani FS, Macri F, et al. Capgras-like syndrome in a patient with an acute urinary tract infection. Neuropsychiatr Dis Treat. 2013;9:139–142.
  16. Harwood D, Barker WW, Ownby RL, et al. Prevalence and correlates of Capgras syndrome in Alzheimer’s disease. Int J Geriatr Psychiatry. 1999;14:415–420.
  17. Mizrahi R, Starkstein SE, Jorge R, et al. Phenomenology and clinical correlates of delusions in Alzheimer disease. Am J Geriatr Psychiatry. 2006;14:573–581.
  18. Harciarek M, Kertesz A. The prevalence of misidentification syndromes in neurodegenerative diseases. Alzheimer Dis Assoc Disord. 2008;22:163–169.
  19. Josephs KA. Capgras syndrome and its relationship to neurodegenerative disease. Arch Neurol. 2007;64:1762–1766.
  20. Casu G, Cascella N, Maggini C. Homicide in Capgras’ syndrome. Psychopathology. 1994;27:281–284.
  21. Silva JA, Leong GB, Weinstock R, et al. Delusional misidentification syndromes and dangerousness. Psychopathology. 1994;27:215–219.
  22. Nestor PG, Haycock J, Doiron S, et al. Lethal violence and psychosis: a clinical profile. Bull Am Acad Psychiatry Law. 1995;23:331–341.
  23. Silva JA, Leong GB, Weinstock R, et al. Delusional misidentification and aggression in Alzheimer’s disease. J Forensic Sci. 2001;46:581–585.
  24. Bourget D, Whitehurst L. Capgras syndrome: a review of the neurophysiological correlates and presenting clinical features in cases involving physical violence. Can J Psychiatry. 2004;49:719–725.
  25. Ahn BH, Kim JH, Oh S, et al. Clinical features of parricide in patients with schizophrenia. Aust N Z J Psychiatry. 2012;46:621–629.
  26. Martinsson G, Fagerberg I, Wiklund-Gustin L, et al. Specialist prescribing of psychotropic drugs to older persons in Sweden—a register-based study of 188024 older persons. BMC Psychiatry. 2012;12:197.
  27. Bongue B, Laroche ML, Gutton S, et al. Potentially inappropriate drug prescriptions in the elderly in France: a population-based study from the French National Insurance Healthcare system. Eur J Clin Pharmacol. 2011;67:1291–1299.
  28. Montastruc F, Garadette V, Cantet C, et al. REAL FR Group. Potentially inappropriate medication use among patients with Alzheimer disease in the REAL.FR cohort: be aware of atropinic and benzodiazepine drugs! Eur J Clin Pharmacol. 2013;69:1589–1597.
  29. Mowla A, Pani A. Comparison of topiramate and risperidone for the treatment of behavioral disturbances of patients with Alzheimer disease: a double-blind randomized clinical trial. J Clin Psychopharmacol. 2010;30:40–43.
  30. Lee HW, Jung DK, Suh CK, et al. Cognitive effects of low dose topiramate monotherapy in epilepsy patients: A 1-year follow-up. Epilepsy Behav. 2006;8:736–741.
  31. Loring DW, Williamson DJ, Meador KJ, et al. Topiramate dose effects on cognition. Neurology. 2011; 76:131–137.
  32. Rosenfeld WE. Topiramate: a review of preclinical pharmacokinetic, and clinical data. Clin Ther. 1997;19:1294–1308.
  33. Wang PS, Bohn RL, Glynn RJ, et al. Hazardous benzodiazepine regimens in the elderly: effects of half-life, dosage, and duration on risk of hip fracture. Am J Psychiatry. 2001;158:892–898.
  34. Tannenbaum C, Paquette A, Hilmer S, et al. A systematic review of amnestic and non-amnestic mild cognitive impairment induced by anticholinergic, antihistamine, GABAergic and opioid drugs. Drugs Aging. 2012;29:639–658.
  35. Dell’osso B, Lader M. Do benzodiazepines still deserve a major role in the treatment of psychiatric disorders? A critical reappraisal. Eur Psychiatry. 2013;28:7–20.
  36. Billioti de Gage S, Begaud B, Bazin F, et al. Benzodiazepine use and risk of dementia: prospective population based study. BMJ. 2012;345:e6231.
  37. Chadwick W, Mitchell N, Caroll J, et al. Amitriptyline mediated cognitive enhancement in aged 3xTg Alzheimer disease mice is associated with neurogenesis and neurotrophic activity. PLoS One. 2011;6:e21660.
  38. Gareri P, De Fazio P, Cotroneo A, et al. Anticholinergic drug-induced delirium in an elderly Alzheimer’s patient. Arch Gerontol Geriatr. 2007;44(suppl 1):199–206.
  39. Campbell N, Boustani M, Limil T, et al. The cognitive impact of anticholinergics: a clinical review. Clin Interv Aging. 2009;4:225–233.
  40. Carrière I, Fourrier-Regiat A, Dartiques JF, et al. Drugs with anticholinergic properties, cognitive decline, and dementia in an elderly general population: 3-city study. Arch Intern Med. 2009;169:1317–1324.
  41. Fox C, Richardson K, Maidment ID, et al. Anticholinergic medication use and cognitive impairment in the older population: the Medical Research Council cognitive function and ageing study. J Am Geriatr Soc. 2011;59:1477–1483.
  42. Cai X, Campbell N, Khan B, et al. Long-term anticholinergic use and the aging brain. Alzheimers Dement. 2013;9:377–385.
  43. Chew ML, Mulsant BH, Pollock BG, et al. Anticholinergic activity of 107 medications commonly used by older adults. J Am Geriatr Soc. 2008;56:1333–1341.
  44. Boustani M, Campbell N, Munger S, et al. Impact of anticholinergics on the aging brain: a review and practical applications. Aging Health. 2008;4:311–320.
  45. Vigen CL, Mack WJ, Keefe R, et al. Cognitive effects of atypical antipsychotic medications in patients with Alzheimer’s dementia: outcomes from CATIE-AD. Am J Psychiatry. 2011;168:831–839.
  46. Kennedy J, Deberdt W, Siegal A, et al. Olanzapine does not enhance cognition in non-agitated and non-psychotic patients with mild to moderate Alzheimer’s dementia. Int J Geriatr Psychiatry. 2005;20:1020–1027.
  47. Knol W, van Marum RJ, Jansen PA, et al. Antipsychotic drug use and risk of pneumonia in elderly patients. J Am Geriatr Soc. 2008;56:661–666.
  48. Kales HC, Kim HM, Zivin K, et al. Risk of mortality among individual antipsychotics in patients with dementia. Am J Psychiatry. 2012;169:71–79.
  49. Laroche ML, Perault-Pochat MC, Ingrand I, et al. Adverse drug reactions in patients with Alzheimer’s disease and related dementia in France: a national multicentre cross-sectional study. Pharmacoepidemiol Drug Saf. 2013;22:952–960.

CORRESPONDENCE: Kenneth R. Kaufman, MD, MRCPsych, Departments of Psychiatry, Neurology, and Anesthesiology, Rutgers Robert Wood Johnson Medical School, 125 Paterson Street, Suite #2200, New Brunswick, NJ 08901 USA E-MAIL: kaufmakr@rwjms.rutgers.edu; adamskaufman@verizon.net