Figure 1. Frequency distributions of severity of macroscopic atrophy in the frontal and temporal lobes of patients with presenile-onset frontotemporal lobar degeneration (FTLDp) and elderly patients with frontotemporal lobar degeneration (FTLDe). Severity of atrophy: 0 = not visible, 1 = mild, 2 = moderate, 3 = moderately severe, 4 = very severe. n = number of cases where data available. *Statistically significant difference in temporal lobe atrophy between FTLDe and FTLDp (χ2 = 5.86; P < .02).
Figure 2. Frequency distributions of severity of microscopic atrophy (neuronal loss) in the frontal and temporal lobes and hippocampal sclerosis (HS) of patients with presenile-onset frontotemporal lobar degeneration (FTLDp) and elderly patients with frontotemporal lobar degeneration (FTLDe). Severity of atrophy: 0 = not visible, 1 = mild, 2 = moderate, 3 = moderately severe, 4 = very severe. n = number of cases where data available. *Hippocampal sclerosis was significantly different between FTLDe and FTLDp (χ2 = 4.63; P < .05). LMB indicates lower midbrain; SN, substantia nigra; UMB, upper midbrain.
Baborie A, Griffiths TD, Jaros E, et al. Frontotemporal dementia in elderly individuals. Arch Neurol. Published online April 23. doi:10.1001/archneurol.2011.3323.
eTable 1. Neuropathological features and diagnostic categories of the 19 younger onset FTLD patients (FTLD).
eTable 2. Macroscopic pattern of brain atrophy in 19 FTLDp patients (numbers 12 to 30).
Customize your JAMA Network experience by selecting one or more topics from the list below.
Baborie A, Griffiths TD, Jaros E, et al. Frontotemporal Dementia in Elderly Individuals. Arch Neurol. 2012;69(8):1052–1060. doi:10.1001/archneurol.2011.3323
Objective To determine whether cases of frontotemporal lobar degeneration (FTLD) do exist in elderly individuals and have clinical and neuropathological features distinct from those with presenile onset.
Design Retrospective matched cohort study.
Setting Regional Neuroscience Centre, North East England.
Patients We compared clinicopathological features of 11 cases of FTLD in elderly individuals with 19 cases of presenile-onset FTLD.
Results Retrospective case note analysis showed that most elderly patients with FTLD had behavioral features consistent with orbitofrontal and basofrontal involvement, similar to presenile-onset FTLD, though symptomatic memory loss was present in 91% (10 of 11) of elderly patients with FTLD compared with only 36% (7 of 19) of patients with presenile-onset FTLD. Neuropathologically, the group of elderly patients with FTLD comprised 7 with FTLD–TDP-43, 1 with ubiquitin-positive FTLD, 2 with FTLD-tau/Pick disease, and 1 with FTLD-tau/neurofibrillary tangle–predominant dementia with TDP-43, a composition similar to presenile-onset FTLD. However, hippocampal sclerosis was more common in elderly patients with FTLD than patients with presenile-onset FTLD (82% vs 37%) and more severe in elderly patients with FTLD (P < .05). By contrast, severe atrophy of the frontal and temporal lobes was less common in elderly patients with FTLD (frontal: 45%; temporal: 27%) than patients with presenile-onset FTLD (frontal: 63%; temporal: 78%). Elderly patients with FTLD represented 3.2% of all elderly patients with dementia autopsied at Newcastle General Hospital.
Conclusions Frontotemporal lobar degeneration in elderly patients does exist as a separate entity from presenile-onset FTLD. Its main features include (1) clinically frequent memory loss and behavioral change predominating over language and semantic dysfunction and (2) neuropathologically prominent hippocampal sclerosis but less pronounced cortical lobar atrophy. Clinically, FTLD in elderly patients is underrecognized and should be considered in the elderly subjects presenting with an “atypical Alzheimer disease” phenotype.
The notion of frontotemporal lobar degeneration (FTLD) as a presenile dementia is empirically grounded and based on rare epidemiological studies.1-3 Within the presenile period, FTLD may be the second most common cause of dementia after Alzheimer disease (AD).4 In elderly individuals, however, AD, dementia with Lewy bodies, and vascular dementia are recognized as the major causes of dementia.5,6 Although emphasized as a cause in the differential diagnosis of late-onset dementia, cases with FTLD with onset after age 70 years are occasionally encountered, anecdotally reported, or included in reports that also describe early-onset FTLD.7 Consequently, the epidemiological status of FTLD in elderly individuals has not been extensively studied.8,9 Current diagnostic criteria are weighted toward the younger group; indeed, the Lund and Manchester consensus for FTLD2 lists an onset age of before 65 years as a supportive diagnostic feature.
Advanced age may therefore militate against a diagnosis of FTLD in elderly patients, leading to misdiagnosis in individual cases and underrepresentation in epidemiological studies. Although the clinical and pathological diagnosis of FTLD can be challenging at any age,10 reflecting the heterogeneous clinical4,11 and histological phenotypes,4,11-13 newer diagnostic laboratory tools, including the identification of TDP-43–, ubiquitin-, or FUS-positive inclusions,12,13 have helped to overcome many of these problems.
In the present study, we have retrospectively analyzed patients' case notes and reviewed and updated neuropathological reports according to current diagnostic criteria to determine whether the clinical and neuropathological features of FTLD in elderly patients differ from those of “typical” cases presenting before age 65 years. We have also estimated the proportion of elderly dementia cases represented by FTLD.
Over a 25-year period (1979-2004), 11 cases of FTLD in elderly patients and 19 cases of presenile-onset FTLD were neuropathologically identified (Table 1 and eTable 1) in the Newcastle Brain Tissue Resource. All cases were from the North East of England. All elderly patients with FTLD had been seen by consultants in the Old Age Psychiatry or Adult Psychiatry units during their illness, and one had been assessed by a consultant neurologist (patient 8). Seven were referred for autopsy by consultants in the Old Age Psychiatry Unit (patients 1-5, 9, and 11), 2 were referred by consultant general physicians in the northern region (patients 6 and 7), and 2 were referred by relatives as “brain donations” to Newcastle Brain Tissue Resource (patients 8 and 10). Twelve of the presenile-onset FTLD cases were referred by consultants in the Old Age Psychiatry, Neurology, and General Medicine units and 7 were brain donations referred by relatives to Newcastle Brain Tissue Resource.
Case summaries of the 11 elderly patients with FTLD were constructed from detailed medical, neurological, and psychiatric histories from psychiatry, medical, and general practitioner notes, with the relevant psychiatrists being consulted about specific presenting features. Blinded retrospective clinical impressions were achieved by an experienced cognitive neurologist (T.D.G.) who scored and analyzed a systematic checklist of cognitive and neurological features, including (1) symptomatic anterograde memory loss (AML), (2) behavioral features, (3) anomia, (4) other language deficits, (5) prosopagnosia, (6) other abnormal perception including hallucinations, (7) neurological features that could not be reasonably explained by drugs, (8) altered level of consciousness, and (9) affective disturbance.
All patients had a general medical examination and appropriate investigations (Table 2 and Table 3). In many of the cases, psychiatric assessments were directed toward establishing a diagnosis at single times and included the 37-item Mental Test Score (MTS)18 or the 30-item Mini-Mental State Examination19 scores (except in patients 5 and 6). Where possible, memory loss was confirmed using data from the MTS, including short-term memory. The history was also used in individual patients to ascertain any AML; repeated tests of mental function were not systematically performed. Brain imaging over this 25-year period was generally not available or not requested, although in 4 more recent patients (patients 6-9) computed tomography and/or magnetic resonance imaging scans were performed. Autopsy reports were obtained from the Neuropathology Department at Newcastle General Hospital.
Standard macroscopic and microscopic neuropathology examination of all relevant brain regions (including pattern and location of atrophy20) was performed in all cases as previously described.14 Where applicable, examined brain regions were semiquantitatively scored (− = nil, normal; + = a few, mild; ++ = frequent, moderate; +++ = numerous, severe; ++++ = abundant, very severe) with respect to the following features: microvacuolation, neuronal loss, ubiquitin (cytoplasmic, nuclear, or neuropil) inclusions, astrocytic gliosis, Alzheimer type pathology including amyloid plaques scores,16 neurofibrillary tangle staging,17 and vascular pathology. Blocks from the prefrontal and temporal cortices were additionally scored for ubiquitin, τ2, TDP-43, and glial fibrillary acidic protein pathology. Patients were anonymized and 2 neuropathologists (R.P. and A.B.) reviewed the neuropathology examination results in all cases independently, with an agreement rate of 90%. Differences were discussed in the remainder and agreement, reached (Table 1 and eTable 1).
DNA was extracted from frozen brain tissue by routine phenol chloroform methods and samples were tested for mutations in MAPT (exons 1 and 9-13)21 and in CHMP2B, TARDBP, FUS, and PGRN15,22 by direct sequencing using the BigDye Terminator kit (Applied Biosystems), as recommended by the manufacturer. The sequencing reactions were analyzed in ABI 3730 DNA analyzer using Sequencer software version 4.8 (Gene Codes Corporation). MAPT haplotype was determined as previously described.23APOE genotype was determined by genotyping single-nucleotide polymorphisms rs429358 and rs7412 using TaqMan assays (Applied Biosystems). The variants were compared with the National Center for Biotechnology Information Single Nucleotide Polymorphism Database and http://www.molgen.ua.ac.be/ADMutations/.
The nonparametric median test was used to obtain differences in medians between elderly patients with FTLD and presenile-onset FTLD. Differences were considered significant for P < .05. For brain weight, the 2 groups were tested using the t test. Analyses were done using Microsoft Excel 2010.
The elderly patients with FTLD differed from the patients with presenile-onset FTLD in the following: the mean age at onset (73.5 years) was 24 years later, the mean disease duration (57.7 months) was 26 months shorter, and the mean age at death (78.4 years) was 21 years later (Table 4). Three of the 11 elderly patients with FTLD (patients 1, 5, and 7) were classified as having “dementia of unknown etiology” among 93 elderly patients with dementia neuropathologically diagnosed between 1982 and 1987 with other forms of dementia.6 Hence, the prevalence of elderly patients with FTLD in Newcastle General Hospital autopsies amounts to 3.2%.
During life, only 1 elderly patient with FTLD had been formally clinically diagnosed with frontotemporal dementia (FTD), the others having been referred with diagnoses of senile dementia/nonspecific dementia syndrome, AD, vascular dementia, and/or depression (Table 3). The 11 elderly patients with FTLD presented with noncognitive and cognitive features that persisted during their illness (Table 2 and Table 3). All elderly patients with FTLD (except patient 6) had documented behavioral changes at some point during their illness similar to, but less prominent than, those in the patients with presenile-onset FTLD.24-26The changes included disinhibition (7 patients), progressive loss of insight (5 patients), emotional lability (5 patients), repetitive behavior and purposeless acts (5 patients), poor personal hygiene (4 patients), withdrawal (2 patients), stubborn personality (2 patients), hyperactivity, fatuous behavior, self-neglect, silly euphoria, and giggling inappropriately (single patients).
Memory loss, including AML, was one of the presenting symptoms in 7 of 11 elderly patients with FTLD (patients 2, 4, and 6-10) (Table 2 and Table 3) and was mentioned by relatives in most patients. Later, a severe degree of memory loss was present in 10 of 11 patients, with mild memory loss in 1 (patient 5). Scores for the MTS, Mini-Mental State Examination, and short-term memory were documented in 9 of 11 patients (Table 2), reflecting a degree of AML in the majority. An isolated AML was a prominent presenting feature in 4 patients (patients 2, 4, 6, and 10), which, in the absence of other psychiatric features in patient 4, raised the question of an isolated dysmnestic (Korsakoff type) syndrome. Anterograde memory loss with behavioral features (self-neglect, inappropriate giggling, stubbornness, and lack of concentration) was a presenting feature in patient 7. Anterograde memory loss associated with confusion was a prominent presenting feature in patient 2.
Retrospective analysis of the clinical features in the elderly patients with FTLD by a cognitive neurologist (T.D.G.) resulted in a clinical impression of FTD in 6 of 11 of the patients (Table 3). Although this compares favorably with the FTD diagnosis by nonspecialists in 1 of 11 patients (patient 3) during life, it suggests that other reasons may underlie the clinical diagnostic difficulties.
Five patients had been treated for depression at some stage in their illness (patients 3, 4, 7, 8, and 10). Antipsychotics were part of the treatment in 7 patients (patients 2, 4-6, and 8-10). Cholinesterase inhibitors were administered in patient 8 about 2½ years before death, leading to a “brighter and more settled appearance.”
Alzheimer disease had been clinically suspected, either alone or as a component of the illness, in 6 elderly patients with FTLD (patients 4 and 6-10) (Table 3), but none had sufficient ATP to fulfill diagnostic criteria for AD, being consistent with age-associated ATP (Table 1). Diffuse amyloid plaques were focally moderate in 2 patients (patients 9 and 10), sparse in 4 (patients 3-5 and 8), and absent from 5 (patients 1, 2, 6, 7, and 11). Patients 5 and 11 had Braak stage I disease, but the likelihood of AD was low.16,17,27 No patient showed α-synuclein pathology.
Vascular dementia had been clinically suspected, at least in the differential diagnosis, in 4 patients (patients 2, 5, 6, and 9), but none of these cases had pathological evidence of infarcts or ischemia. Patient 9 showed bilateral moderate atherosclerosis affecting the internal carotid and middle cerebral arteries, but cerebral arteries appeared normal in all other patients. Small vascular lesions without any clinical correlates were identified in 3 other patients: patient 3 (a small ischemic area with neuronal loss was identified in the inferior cerebellum), patient 7 (small, 2-mm-diameter, cystic infarcts in the hypothalamus and the upper mid pons with a focus of ischemia in the right cerebellar hemisphere), and patient 11 (small, 10-mm-diameter, cystic infarcts in the left inferior occipital lobe).
The retrospective clinical impression of FTD bore out the neuropathological diagnosis of FTLD in 6 of 11 patients (patients 1, 3, 5, 7, 10, and 11) (Table 3). However, in the remaining 5 patients (patients 2, 4, 6, 8, and 9), the retrospective clinical diagnosis of AD/atypical AD did not accord with the neuropathological diagnosis of FTLD. In most cases, hippocampal sclerosis (HS) correlated with AML at presentation and with low MTS scores, and the absence of HS correlated with mildly impaired memory in patient 5 (Table 1 and Table 2). However, patient 7, who presented with AML, had an MTS score of 12 of 37 nine months before death (Table 2) but no HS (Table 1). This patient was 1 of only 2 with prominent caudate atrophy (Table 5). Patient 3, on the other hand, had severe HS (score ≥4) (Table 1), an intact short-term memory (score 5 of 5), and an MTS score of 34 of 37 twenty-seven months before death (Table 2). In this instance, HS may have been unilateral, but unfortunately, the other cerebral hemisphere was frozen and not available for histological examination.
The mean fresh brain weight was not significantly different in elderly patients with FTLD from that in presenile-onset FTLD (P = .22) (Table 5 and eTable 2). Atrophy of the frontal and temporal lobes was present in all elderly patients with FTLD but less severe than in patients with presenile-onset FTLD (Figure 1), with a smaller proportion of the elderly patients with FTLD (45% of frontal and 27% of temporal lobes) scoring 4 or more and 3 or more compared with the patients with presenile-onset FTLD (63% of frontal and 78% of temporal lobes). Frontal lobe atrophy was not significantly different between elderly patients with FTLD and patients with presenile-onset FTLD (χ2 = 1.39; P < .30) but temporal lobe atrophy was significantly less in elderly patients with FTLD (χ2 = 5.86; P < .02). Caudate nucleus atrophy also appeared less severe in elderly patients with FTLD than in patients with presenile-onset FTLD, with only 2 of 11 elderly patients with FTLD scoring 3 or more compared with 4 of 13 of the patients with presenile-onset FTLD. In contrast, parietal and occipital lobar atrophy predominated in elderly patients with FTLD (5 of 11) compared with patients with presenile-onset FTLD (1 of 15).
Cortex. Mild to moderately severe microvacuolation and neuron loss in superficial cortical layers of the frontal and temporal lobes was identified in all the elderly patients with FTLD, but generally, neuron loss appeared less severe than in patients with presenile-onset FTLD, with a smaller proportion of the elderly patients with FTLD (27% of frontal and 27% of temporal cortices) scoring 4 or more or 3 or more compared with patients with presenile-onset FTLD (58% of frontal and 72% of temporal cortices), although the differences were not significant (frontal, .05 < P < .10; temporal, .10 < P < .90) (Figure 2, Table 1, and eTable 1).
Seven of the elderly patients with FTLD had ubiquitin-positive, TDP-43–positive neuronal cytoplasmic inclusions, intranuclear inclusions, and/or dystrophic neurites in the cortex and the dentate fascia granule cells (patients 1-7) and were classed as having FTLD–TDP-43 (Table 1)12 categorized into FTLD–TDP-43 subtypes 1, 2, and 3.12 Patient 11 (patient N30 in Baborie et al14) was classified as having FTLD-UPS because of the presence of ubiquitin-positive, TDP-43–, FUS-, or tau-negative neuronal cytoplasmic inclusions as well as dystrophic neurites in the frontal and temporal lobes.12 Three patients showed FTLD-tau: patient 8 (patient N29 in Baborie et al14) had FTLD-tau/neurofibrillary tangle–predominant dementia with TDP-43 pathology12 and patients 9 and 10 had FTLD-tau/Pick disease.12 Similarly, of the 19 patients with presenile-onset FTLD, 8 had FTLD–TDP-43, 4 had FTLD-UPS (negative for TDP-43, FUS, and tau), 6 had FLTD-tau/Pick disease, and 1 had FTLD-FUS (eTable 1).
Hippocampal Sclerosis. Strikingly, 9 of 11 elderly patients with FTLD (82%) showed selective loss of pyramidal neurons in the hippocampus (Table 1), involving the cornus ammonis 1 (CA1) and subiculum, characteristic of HS. Patient 5 had only focal neuron loss and mild gliosis in CA1 but many cytoplasmic TDP-43–positive preinclusions28 in CA1. By contrast, 7 of 19 patients with presenile-onset FTLD (37%) were affected by HS (eTable 1). Hippocampal sclerosis was more severe in elderly patients with FTLD than in patients with presenile-onset FTLD (χ2 = 4.63; P < .05), with none of the patients with presenile-onset FTLD showing the highest score (≥4) for HS (Figure 2 and Table 1).
Substantia Nigra. Loss of substantia nigra neurons was present at upper and lower midbrain levels in 9 of 11 elderly patients with FTLD and 12 of 19 patients with presenile-onset FTLD, but there was no obvious difference in the severity of loss between the 2 groups (Table 1 and eTable 1).
Genetic analysis did not reveal any pathogenic variants for MAPT, PGRN (patient 3 was homozygous for the T allele at rs5848), CHMP2B, TARDBP, or FUS in either group. APOE alleles and MAPT haplotypes were unremarkable (Table 1 and eTable 1).
In the present study, we have investigated a cohort of elderly patients with FTLD from the North East of England. We retrospectively analyzed the patients' case notes because only 1 of the patients had been formally diagnosed with FTD during life. We recognize that such retrospective interpretations of notes need to be treated with caution, but nonetheless, the clinical features of the elderly patients with FTLD differed from those cases of “typical” presenile-onset FTLD reported in the literature2,24-26 and encountered in daily clinical practice. Neuropathologically, the elderly patients with FTLD mainly had non–tau FTLD, with milder frontal atrophy and significantly more HS. Clinically, the elderly patients with FTLD presented mainly with AML, with less prominent behavioral symptoms and naming deficits.
Behavioral features in elderly patients with FTLD were not as striking as in patients with presenile-onset FTLD,24-26 though the evolving behavioral aspects of the presentations (progressive loss of insight, repetitive behavior, and poor hygiene) would militate against the diagnosis of other dementia syndromes.29 The presence of milder behavioral features in elderly patients with FTLD appears to correlate with the finding of milder lobar atrophy compared with patients with presenile-onset FTLD.
More than half of the elderly patients with FTLD had a clinical phenotype that was difficult to distinguish from AD, presenting with prominent AML. Alzheimer type pathology in elderly patients with FTLD, however, did not exceed that found in the “normal” elderly population.30 The memory problems correlated with the severity of HS, apart from patient 3. It is possible, though uncertain, that hemispheric lateralization and/or the rate of progression of HS during the course of illness had an impact in this case.
Similar to data published for an FTLD cohort in Manchester, England,14 lobar atrophy was milder and HS was more severe in elderly patients with FTLD than in patients with presenile-onset FTLD. In our Newcastle General Hospital cohort, HS was about twice as common in elderly patients with FTLD than in patients with presenile-onset FTLD.
Short disease duration in the 2 elderly patients with FTLD without HS (patients 5 and 7) may explain the focal neuronal loss in CA1 rather than clear HS. Like all patients with FTLD–TDP-43 subtypes, both had pathological FTLD features, ubiquitin- and TDP-43–positive neuronal cytoplasmic inclusions, and dystrophic neurites in the frontal and temporal lobes and dentate gyrus, fulfilling criteria for FTLD–TDP-43, rather than HS with TDP-43 pathology. Patient 5 had additionally many cytoplasmic TDP-43–positive preinclusions28 in CA1, which may precede a true HS. The more severe HS in elderly patients with FTLD may therefore be more likely associated with a more severe FTLD pathology in the hippocampus rather than with a superimposed HS dementia of the very old (≥80 years31). Although HS is not common in patients with presenile-onset FTLD,14 it is being increasingly recognized as a pathological feature of FTLD–TDP-4332 and FTLD-FUS33,34 and can be present in FTLD-UPS14,35-38 regardless of subtype.13,35,36 A recent study has claimed a common variant of PGRN (rs5848)as a risk factor for HS in elderly individuals,39 though Josephs et al40 did not find any difference in the frequency of HS between patients with or without PGRN mutations. We found homozygosity for the T allele at this locus in only 1 patient (patient 3).
Based on this series, FTLD in elderly patients does exist as a separate entity, though it is comparatively rare, amounting to 3.2% of dementia cases in the elderly individuals in the Newcastle General Hospital cohort. The clinical features of the elderly patients with FTLD differed from those of the patients with “typical” presenile-onset FTLD in several respects. First, the majority of the elderly patients with FTLD presented with or developed AML and had low scores on memory tests. Frontotemporal lobar degeneration with TDP-43 molecular pathology was prominent in the cohort, which is possibly relevant to the high prevalence of HS and AML. Second, many of the elderly patients with FTLD had frontal lobe syndromes but these were not as striking in severity or as rapidly progressive as those seen in patients with presenile-onset FTLD. Third, neuropsychological features of semantic dementia were absent or rare in elderly patients with FTLD. Frontotemporal lobar degeneration in elderly patients may therefore be clinically underrecognized. This diagnosis should be considered in elderly patients with “atypical AD” phenotypes.41
Correspondence: Atik Baborie, MD, Department of Neuropathology, Walton Centre for Neurology and Neurosurgery, Lower Lane, L9 7LJ Liverpool, England (firstname.lastname@example.org).
Accepted for Publication: November 25, 2011.
Published Online: April 23, 2012. doi:10.1001/archneurol.2011.3323
Author Contributions:Study concept and design: Baborie, McKeith, Mann, and Perry. Acquisition of data: Baborie, Jaros, Momeni, McKeith, Burn, and Perry. Analysis and interpretation of data: Baborie, Griffiths, Jaros, Momeni, McKeith, Keir, Larner, Mann, and Perry. Drafting of the manuscript: Baborie, Griffiths, Momeni, Mann, and Perry. Critical revision of the manuscript for important intellectual content: Baborie, Griffiths, Jaros, McKeith, Burn, Keir, Larner, Mann, and Perry. Statistical analysis: Keir and Perry. Obtained funding: McKeith and Perry. Administrative, technical, and material support: Baborie, Griffiths, Momeni, Larner, Mann, and Perry. Study supervision: Baborie, Jaros, and Perry.
Financial Disclosure: None reported.
Funding/Support: Dr Jaros, Andrew Brown, senior BMS, and Lynne Ramsay, senior BMS, have been in part supported by the UK National Institute for Health Research Biomedical Research Centre for Ageing and Age-Related Disease Award to the Newcastle upon Tyne Hospitals NHS Foundation Trust. The Newcastle Brain Tissue Resource is funded in part by grants G0400074 and G1100540 from the UK Medical Research Council, the Newcastle NIHR Biomedical Research Centre in Ageing and Age Related Diseases award to the Newcastle upon Tyne Hospitals NHS Foundation Trust, and a grant from the Alzheimer’s Society and Alzheimer’s Research Trust as part of the Brains for Dementia Research Project.
Additional Contributions: We gratefully acknowledge the Newcastle Brain Tissue Resource, which manages all the brain tissue from the patients with FTLD used in this study. We thank all staff at the Neuropathology Department, Newcastle General Hospital, for technical assistance: Andrew Brown and Janet Thompson, senior BMS, for cutting and staining of the sections and Lynne Ramsay for the TDP-43 staining. We thank our patients, their relatives, and our colleagues for the diagnostic referrals.
Create a personal account or sign in to: