Eleven cases with a clinical diagnosis of bvFTD
 and FTLD pathology
, 9 cases with clinical and pathological AD
, and 14 controls without dementia or significant neuropathological abnormalities were selected from a neuropathological series of cases collected by the Sydney Brain Bank through a regional brain donor program. Patients were diagnosed by experienced clinicians using standard diagnostic criteria
[17, 18] following a medical interview and an informant history. The bvFTD all had changes in personality or social behaviour (eg. apathy, disinhibition, stereotypic behaviours, alterations in food preference, or poor self-care) and executive dysfunction (eg. poor planning, forethought, reasoning or organisation). The program holds approval from the Human Research Ethics Committee of The University of New South Wales and complies with the statement on human experimentation issued by the National Health and Medical Research Council of Australia. This research project was approved by the Human Research Ethics Committees of the Universities of Sydney and New South Wales. Some of the cases included in this study have been reported in previous publications
[3, 19, 20]. The research program used standardized tests following the patients longitudinally with the last assessments performed within 14 months of death. All dementia cases had Clinical Dementia Rating (CDR) scores between 1 and 3 while controls had scores of <0.5. The postmortem interval was 16 hours on average (range: 2–45 h; mean ± SD for bvFTD = 9 ± 12, for AD = 14 ± 10, and for control = 24 ± 12, ANOVA p > 0.05). Cases with bvFTD with TDP pathology (FTLD-TDP, n = 6, 5 males; age range 53–72 years; 2 with type A, 2 with type B and 2 with type C)
 and tau pathology (FTLD-tau, n = 5, 1 male; age range 65–79 years; 5 with Pick’s disease) were selected for this study. Cases with all types of TDP-43 pathologies were selected, as our previous study showed no differences between different TDP-43 pathology types at end-stage
. Case details are shown in Table
Tissue preparation and volume measurements
Tissue preparation and volumetric methods used in this study have been published in detail elsewhere
[19, 20]. Briefly, brains were collected at autopsy, fixed in 15% neutral buffered formalin for two weeks, the fixed brain weighed and the volume determined by fluid displacement. The cerebellum and brainstem were separated from the cerebrum by sectioning through the cerebral peduncles. Each cerebrum was embedded in 3% agarose, sectioned in 3 mm coronal slices using a rotary slicer, photographed and printed at 1× magnification. The average slice thickness for each brain was determined by dividing the hemisphere length by the total number of slices. Diagnostic neuropathological screening was conducted using standardized blocks taken from the frontal, parietal, temporal and occipital cortices, amygdala, hippocampus, basal ganglia, diencephalon, midbrain, pons, medulla oblongata and cerebellum. Examination of sections stained with haematoxylin and eosin and modified Bielschowsky silver stain, as well as immunohistochemically for tau, ubiquitin, α-synuclein and TDP-43, confirmed pathological diagnoses.
Anatomical structures and gyral boundaries most consistently associated with cytoarchitectonic boundaries were used to identify the AC and PC for all cases, as previously described
[19, 20]. The AC was divided from the PC at the central lobule of the central sulcus. The volumes of these regions were determined by a point counting procedure on the brain slice photographs as previously described
[19, 20]. Briefly, the areas corresponding to each region were identified on the brain slice photographs and were randomly overlaid with a grid of 3848 points (each separated by 4 mm). The total number of points falling on each region of interest was counted and the volumes calculated by multiplying the sum of points falling on a given region by the volume represented by each point (volume/point = 16 mm2 × mean slice thickness; average of 50 mm3). This method is routinely used in our laboratory to measure regional volumes postmortem
[20, 22, 23] and approximates current point counting procedures used in MRI studies of brain volumes.
Quantitation of neuron numbers and inclusion pathologies
Standardised tissue samples of the cingulate cortices were sampled (AC just posterior to the genu of the corpus collosum, PC just anterior to the splenium of the corpus callosum) and embedded in paraffin wax using routine procedures. Ten μm serial sections were cut from each block and immunohistochemically stained with antibodies to tau (mouse anti-human tau diluted 1:1000; Thermo Scientific) and TDP-43 (rabbit anti-human TDP-43 diluted 1:800; ProteinTech) using routine techniques
. An additional section was stained with cresyl violet (0.5%). Quantitation of cortical neuronal populations was performed as previously described and validated
. Briefly, 2 × 500 μm wide strips through the entire cortical thickness from the pial surface to white matter were randomly sampled by marking perpendicular lines on the coverslip of each slide. Strips were not selected if the cortex was cut obliquely at the marked line to ensure even representation of all cortical laminae. Neurons were counted at 200× magnification using a 10 × 10 eyepiece graticule (500 μm × 500 μm) with standard inclusion (lower and left) and exclusion (upper and right) borders in contiguous, non-overlapping fields. The density of neurons within each region was calculated for each case, and the total number of neurons estimated by multiplying neuronal density by the calculated volume for the AC and PC, as appropriate. The percentage of neurons with cytoplasmic inclusions in the AC and PC was expressed as a percentage of total remaining neurons. Quantitation was performed by two raters blind to case details with an inter-rater variance of 2.4% and intra-rater variability of 2.2%.
Data were analysed using SPSS19.0 (IBM Corp., Chicago, Ill., USA). Group differences in age and postmortem delay across all groups were investigated using analysis of variance (ANOVA), and where these variables showed significant differences, they were entered as co-variates in the multivariate ANOVA. Post hoc analyses using the Bonferroni correction for multiple comparisons were performed to identify specific group differences in regional neuronal loss. Any associations between neuronal number and the percentage of neurons with pathological depositions (tau, TDP-43) or measures of disease severity (disease duration and CDR score) were identified with Spearman rank correlations.