Very Early Semantic Dementia With Progressive Left≫Right Temporal Lobe Atrophy: An Eight-Year Longitudinal Study

Abstract

Background

Semantic dementia (SD) is a syndrome within the spectrum of frontotemporal lobar degenerations (FTLD) characterized by fluent progressive aphasia (particularly anomia) and loss of word meaning.

Objective

To report a unique case of very early semantic dementia with slowly progressive course allowing insights into the early natural history of this disorder.

Design

Case report.

Setting

Tertiary care university hospital and academic center.

Patient

A 62-year-old female retired teacher presenting with “memory” complaints.

Main Outcome Measures

Clinical course, neuropsychological data, MRI.

Results

The patient was first evaluated when standard neuropsychological measures were normal, but subtle left anterior temporal lobe atrophy was present. Over the follow-up period of eight years, she developed profound anomia and loss of word meaning associated with progressive left anterior temporal lobe atrophy consistent with semantic dementia. In more recent years, anterograde memory impairment as well as mild prosopagnosia have evolved in association with left hippocampal atrophy and subtle atrophy in the homologous gyri of the right anterior temporal lobe. She remains functionally independent despite her current deficits.

Conclusions

Early identification of patients who will develop semantic dementia is difficult and might be missed with standard clinical, neuropsychological, and structural neuroimaging evaluations. Recognition of this relatively rare syndrome is important for early diagnosis and prognostication, and particularly for therapeutic interventions in the future.

Introduction

Mesulam first described a series of patients characterized by aphasia without dementia in 1982 and later named this syndrome “primary progressive aphasia”.¹ Patients can be distinguished based on fluency of speech into having either progressive nonfluent or fluent aphasia. The term “semantic dementia” (SD) refers to fluent aphasia with additional loss of word meaning for both spoken and written content.² Clinically, patients have impairment in language function and object representation with preservation of episodic memory and absence of deficits in other cognitive domains during the first two years of the illness.³ Imaging studies reveal distinct focal atrophy involving the inferolateral temporal lobes, usually in an asymmetric distribution.⁴ There have been advances in determining the underlying neuropathology of frontotemporal dementia, and most patients with SD have ubiquitin-positive, tau-negative inclusions as the correlate of their disease.⁵ More recent analyses have shown such inclusions immunostain with TAR-DNA binding protein-43 (TDP-43).^6,7 There also have been reports of tauopathies and Alzheimer's pathology in patients with SD⁸, although there might be differentiating features on neuropsychological testing and brain imaging.⁷ The early natural course of SD is largely unknown, and many patients only come to medical attention later in their illness when either profound anomia is already established or additional behavioral symptoms have developed. We report a unique case who presented to our clinic very early in her illness and developed slowly progressive SD with typical clinical and radiological features, allowing insights into the early natural history of the disorder.

Methods

The patient was evaluated once yearly over eight years by comprehensive neurological examination⁹, neuropsychological testing and magnetic resonance imaging (MRI) of the brain as a participant in the Mayo Alzheimer's Disease Research Center program – a Mayo Foundation Institutional Review Board-approved research program.

Neuropsychology

The patient underwent serial neuropsychological testing including assessment of global functioning [Folstein Mini-Mental State Exam (MMSE)¹⁰, Kokmen Short Test of Mental Status (STMS)¹¹], learning and memory [Logical Memory (WMS-LM) and Visual Reproductions (WMS-VR) of the Weschler Memory Scale-Revised (WMS-R), Percent Retention on the Auditory Verbal Learning Test percent retention (AVLT-PR)], executive functioning [Trail Making Test parts A and B (TMT-A, TMT-B), Digit Symbol (Dig Sym) subtest of the Weschler Adult Intelligence Scale-Revised (WAIS-R)], language functioning [Boston Naming Test (BNT), Controlled Oral Word Association Test (COWAT), category fluency (CAT FLU)], and visuospatial functioning [Block Design subtest of the WAIS-R (WAIS-BD), Rey-Osterrieth Complex Figure Test (Rey-O), Judgment of Line Orientation (JLO)]. Mayo Older American Normative Studies (MOANS) norms were used to determine scaled scores for these tests, in which 10 represents the mean and the standard deviation is 3.^12-15

Brain imaging

MRI was performed using a GE scanner at 1.5 Tesla, and images of the brain were obtained in sagittal (T1-weighted), axial [proton-density, T2-weighted, fluid attenuation inversion recovery (FLAIR)], and coronal (T1-weighted) planes.

Case report

A 62-year-old married Caucasian female was referred to our Behavioral Neurology Clinic with a six-month history of “memory loss,” specified as problems recalling names of people and objects. She had mild difficulties in organizational skills and complex decision making, but was highly functional in all activities of daily living and engaged in various social activities. Her past medical history was significant for treated pernicious anemia and hypothyroidism. The patient had recently retired from teaching and led a health-conscious lifestyle. She reported several second degree family members diagnosed with late onset Alzheimer's disease. General neurological examination and standard laboratory work-up were unremarkable. Her mental status exam revealed a score of 33/38 on the STMS, recalling 3 out of 4 items correctly on delay. On detailed neuropsychological testing, performance on learning and memory, language, and other cognitive domains was within low average to average range (Figure 1). Initial MRI of the brain was considered normal for age, although in retrospect subtle widening of the collateral sulcus and thinning of the superior temporal gyrus in the left anterior temporal lobe was present (Figure 2). The patient was seen on return visits for the following eight years. Two years after onset of her complaints, her husband reported substituting words, circumlocution, semantic errors and frequent spelling errors. Formal speech evaluation by an experienced speech pathologist demonstrated normal motor speech, verbal comprehension and retention, verbal expression, reading and writing skills, but a marked deficit in word retrieval (<5^th percentile on the BNT). Furthermore, she did not recognize missed target words when they were provided to her. Over the following years, the patient had progression of anomia as documented by decline on BNT and category fluency. In contrast, scores remained average to above average for attention/executive functioning and visuospatial tasks. She performed near or above generally accepted cut-off scores on screening measures (eg, ≥24 on MMSE and ≥29 on STMS). Anterograde memory performance was variable across follow-up with generally impaired WMS-VR and RAVLT performance for the last 3 to 5 years. In the most recent evaluation, all measures of anterograde memory were impaired (Figure 1). On serial language examinations, she developed surface dyslexia, loss of word meaning, and difficulties in recognizing famous faces and clearly met criteria for semantic dementia.³ Serial MRI scans document progressive left anterior inferolateral temporal lobe atrophy with relative sparing of the hippocampus until age 68 (Figure 2). On the most recent scans, there is evidence of right temporal lobe atrophy in a strikingly similar pattern of topography compared to the left side. From a behavioral standpoint, the patient has become slightly more mentally rigid and parsimonious, but remains independent in her simple and complex activities of daily living.

Figure 1.

Longitudinal performance in screening mental status examinations and key cognitive domains for specific neuropsychological tests. All raw scores were converted to scaled scores (SS) based on MOANS norms (mean = 10; standard deviation = 3). Shaded areas represent scores in the abnormal range. Note that impaired performance on the BNT preceded impairment on semantic fluency, and impairment in verbal and visual memory coincided with impaired naming. By history, a decline in the patient's episodic memory was only noted at evaluation 7. See text for abbreviations.

Figure 2.

Serial coronal T1-weighted (left column) and axial FLAIR (right column) MR images showing progressive left anterior temporal lobe atrophy. Note subtle widening of the left collateral sulcus and thinning of the superior temporal gyrus on early scans with clear progression to atrophy; development of hazy increased signal in the left mesial temporal lobe on FLAIR images; and development of atrophy in homologous regions in the right anterior temporal lobe over more recent scans. Numbers on the left represent the patient's age (in years) corresponding to that row of images.

Comment

We report a case of very early SD with longitudinal neuropsychological and radiological follow-up allowing insights into the early natural course of this syndrome. Our patient exemplifies how subjective “memory” complaints precede objective evidence of cognitive impairment by standard testing methods. Her course over eight years was remarkably slowly progressive, compared to previously published data with an estimated mean survival time of 8 years after diagnosis.¹⁶ On first evaluation, both her neuropsychological assessment and imaging findings were felt to be normal, although in hindsight subtle atrophy of the left collateral sulcus and superior temporal gyrus was present on MRI. The striking asymmetry of temporal lobe atrophy with predilection for the dominant hemisphere has been well described,^4,17,18 and relative sparing of the mesial temporal lobe structures seems to account for preservation of episodic memory, at least during the initial years. Later in the course, the opposite homologous temporal lobe regions tend to become atrophic in the exact same topography, as seen in our patient. The underlying mechanisms of focal atrophy and eventual homologous opposite hemisphere involvement in SD are unknown, but one could hypothesize apoptosis and/or transcollosal degeneration is/are at play.

Making an early diagnosis and differentiating between the various dementia syndromes remains challenging even for behavioral neurologists, stressing the need for diagnostic biomarkers. Furthermore, correlating the clinical dementia syndrome with underlying pathophysiological changes has proven to be difficult in FTLD.¹⁹ In a post-mortem study, SD has been shown to be associated with tau-negative, ubiquitin-positive inclusions in 13/18 cases.⁵ The ubiquitinated protein has recently been identified as TAR DNA-binding protein 43 (TDP-43), raising hope for future treatment strategies targeting TDP-43 pathophysiology.⁶ Treatment is most likely to be beneficial the earlier it is instituted, but as exemplified by this case, early diagnosis will be challenging using standard neuropsychologic measures and MRI.