Objectives  To describe an atypical behavioral presentation of Alzheimer disease and to review the literature on the subject.

Design  Case study.

Setting  Outpatient specialty clinic.

Patient  A 70-year-old man with progressive behavioral symptoms of hyperorality, hypersexuality, hypermetamorphosis, visual agnosia, hyperphagia, and apathy who died at age 77 of asphyxiation on a foreign object.

Main Outcome Measures  Clinical symptomatology, brain imaging, and neuropathology.

Results  The pathologic diagnosis was Alzheimer disease with abundant tangles and plaques in the lateral amygdala.

Conclusions  This case represents a variant of Alzheimer disease with prominent amygdala abnormalities and a Klüver-Bucy phenotype that was misdiagnosed as frontotemporal dementia. Clinical and imaging findings that may aid in accurate diagnosis are reviewed.

REPORT OF A CASE

His medical history included asthma, hypertension, gout, and sleep apnea. He had no psychiatric history. His medications were albuterol, ipratropium bromide, cromolyn sodium, triamcinolone, theophylline, allopurinol, and indomethacin. There was no history of alcohol, tobacco, or other substance use. His sister and father had unspecified late-onset dementia.

On examination, he was observed to be obese (107.55 kg, 167.64 cm tall) and "jocular and inappropriate." His Mini-Mental State Examination score was 30 of 30. Results of his neurologic examination were normal except for symmetrical hyporeflexia, marginally impaired graphesthesia, and a right palmomental sign. Neuropsychological testing revealed impairment in object naming and mild impairment in acquisition, with otherwise normal scores (Table). Basic chemistries, complete blood cell count, thyrotropin level, vitamin B₁₂ level, and microhemagglutination Treponema pallidum level were normal. Brain magnetic resonance imaging revealed a 0.7 x 2-cm meningioma adjacent to the left sylvian fissure and very mild subcortical punctate white matter changes; otherwise, the results were interpreted as normal. Retrospective review of these images suggested bilateral parietal atrophy (Figure 1). The initial clinical impression was mild frontotemporal dementia (FTD).

View this table: [in this window] [in a new window] [as a PowerPoint slide]   Table. Neuropsychological Test Results for the Patient at Ages 70 to 74 Yearsa

View larger version (43K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 1. Brain magnetic resonance images of the patient at age 72 years showing slightly enlarged biparietal sulci and the absence of frontal or temporal lobar atrophy.

During the next 2 years, he reported that he began having more prominent sexual fantasies about other women while also becoming increasingly suspicious that his wife was cheating on him; he was prescribed paroxetine mesylate. His Mini-Mental State Examination score remained at 28 of 30; however, the neuropsychological studies began to show clear deficits in delayed recall and language with spared visuoperceptive ability (Table). Positron emission tomography (PET) demonstrated hypometabolism in the bilateral parietal and anterotemporal regions (Figure 2). The possibility of AD was considered, but the leading diagnosis remained FTD. From the ages of 74 to 77 years, he had a substantial decline in cognitive function, and his Mini-Mental State Examination score declined to 7. He displayed increasing hyperorality, and his wife noted that he would put "any" object in his mouth, including dog food, candles, adhesive bandages, and his wedding ring. His appetite seemed insatiable. He would reach into the toilet to manipulate fecal matter. When asked by his wife to get a rake, he brought a screwdriver. He became progressively withdrawn and apathetic. During the last year of his life, he required assistance with most activities of daily living. He died at age 77 years of asphyxiation on several adhesive bandages.

View larger version (43K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 2. Positron emission tomography with radiolabeled fluorodeoxyglucose in the patient at age 73 years showing hypometabolism in the bilateral parietal and anterotemporal regions.

The pathologic diagnosis was AD with abundant neurofibrillary tangles and moderate neuritic plaques throughout the brain (Braak and Braak stage VI and a CERAD [Consortium to Establish a Registry for Alzheimer's Disease] rating of "frequent"). The brain weight was 1400 g. The distribution of tangles and plaques was similar to that encountered in sporadic AD with this staging; however, the degree of amygdala abnormality was atypical. Neurofibrillary tangles were abundant in the lateral amygdala (approximately 150 per 1.6 mm²) (Figure 3 and Figure 4). This number of tangles is greater than 4+ (>50 tangles per 1.6 mm²) using the Kromer Vogt rating system for tangles in the amygdala in AD.² There were neuritic plaques in the lateral, laterobasal, and accessory basal amygdala ranging from approximately 10 to 20 per 1.6 mm² (Figure 5); 2+ Kromer Vogt rating for plaques in the amygdala.² The amygdala showed moderate neuronal depletion and gliosis. There was also evidence of grade II cerebral amyloid angiopathy. Genetic testing showed that the patient had apolipoprotein E 3/3; presenilin genotyping was not performed.

View larger version (197K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 3. Bielschowsky silver impregnation staining showing an overview of large neuritic plaques (arrows) and smaller neurofibrillary tangles in the lateral amygdala adjacent to the inferior horn of the lateral ventricle. Bar = 0.5 mm.

View larger version (152K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 4. Bielschowsky silver impregnation staining showing a higher-power view of neurofibrillary tangles (arrowheads) in the lateral amygdala (approximately 150 tangles per 1.6 mm2). Bar = 0.1 mm.

View larger version (174K): [in this window] [in a new window] [as a PowerPoint slide]   Figure 5. Bielschowsky silver impregnation staining showing a higher-power view of neuritic plaques, amyloid core, and surrounding tangles (approximately 10-20 neuritic plaques per 1.6 mm2). Bar = 0.1 mm.

COMMENT

Sourander and Sjogren¹⁰ described KB syndrome symptoms in patients with AD. Morris et al¹¹ suggested that excessive eating in some patients with dementia may be a form of KB syndrome. Burns and colleagues¹² examined 178 patients with AD to quantify the number of KB syndrome symptoms: 29.3% had 1, 24.7% had 2, 17.2% had 3, 8.6% had 4, and 0.6% had all the KB syndrome symptoms.

Neuronal loss in the medial, central, and cortical nuclei of the amygdala has been observed in AD.^13-14 Amyloid plaques and neurofibrillary tangles are consistently present in the amygdala in AD.^3-5 The present patient had 150 tangles per 1.6 mm² in the lateral amygdala. By comparison, 14 other AD brains with Braak and Braak stage VI were studied and were found to have 5 to 90 tangles per 1.6 mm² in the amygdala. Kromer Vogt et al² investigated the amygdala abnormalities of 20 patients with AD and demonstrated a significant amount of tangles and neuritic plaques in the cortical nucleus and accessory basal nucleus of the amygdala, which have projections with the hippocampus; however, the lateral nuclei of the amygdala were relatively spared (1 patient had a 4+ tangle score and 1 had a 3+ plaque score). Compared with patients with typical AD described by Kromer Vogt et al,² the patient described herein had significant tangle and plaque density in the lateral amygdala. Hayman and colleagues¹⁵ also described a patient with KB syndrome after lateral amygdala lesions.

This is the first published study of this amygdaloid variant of AD with serial neuropsychological testing, magnetic resonance imaging, PET, and pathologic analysis. A frontal variant of AD with prominent behavioral symptoms has been described.¹⁶ The frontal variant is characterized by an atypical distribution of AD abnormalities in the frontal lobes that may be more difficult to distinguish from FTD. In contrast, PET of this patient revealed the typical pattern of posterior hypometabolism. This tool might be a useful diagnostic aid. Behavioral variants of AD are primarily a diagnostic challenge because several of the symptoms overlap with FTD.¹⁷ Cognitive-behavioral symptoms in AD often respond to acetylcholinesterase inhibitors, and misdiagnosis of FTD might preclude this intervention and other potential disease-modifying agents for AD that are currently being investigated. Also, FTD has more rapid progression, with mean survival of 4.2 years from initial evaluation compared with 6 years for AD.¹⁸

This patient's prominent behavioral symptoms led to the incorrect diagnosis of FTD. There were, however, a variety of noteworthy signs and imaging clues that pointed toward the correct diagnosis of AD. The patient noted memory dysfunction, and delayed memory scores were very poor relatively early in the illness. Age at presentation was late, and the course was relatively slow. Magnetic resonance imaging showed evidence of biparietal atrophy. The PET demonstrated biparietal hypometabolism consistent with AD. The specificity and sensitivity of PET, used to differentiate between AD and FTD, were recently shown to be superior to those of clinical assessment alone.¹⁹ Furthermore, the positive finding of temporoparietal hypometabolism is strongly associated with AD abnormalities even in the presence of atypical features.²⁰

AUTHOR INFORMATION

Accepted for Publication: March 15, 2008.

Author Contributions: All authors had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Kile, Olichney, and DeCarli. Acquisition of data: Kile, Ellis, and Farias. Analysis and interpretation of data: Kile, Olichney, and Farias. Drafting of the manuscript: Kile and Farias. Critical revision of the manuscript for important intellectual content: Kile, Ellis, Olichney, and DeCarli. Administrative, technical, and material support: Kile and Farias. Study supervision: Ellis, Olichney, Farias, and DeCarli.

Financial Disclosure: None reported.

Funding/Support: This work was funded by grant P30 AG10129 from the National Institutes of Health.

Additional Contributions: We thank the subject volunteers and their caregivers for their generous contributions.