The article reviewed here is ‘Distinct MRI atrophy patterns in autopsy-proven Alzheimer’s Disease and Frontotemporal Lobar Degeneration’ and is freely available here. In the abstract, the conclusion reads
‘Compared with FTLD, AD patients had decreased gray matter in posterior parietal and occipital cortex, whereas FTLD patients had selective atrophy in anterior cingulate, frontal insula, subcallosal gyrus, and striatum (p<.001, uncorrected). These findings suggest that AD and FTLD are anatomically distinct, with degeneration of a posterior parietal network in AD and degeneration of a paralimbic fronto-insular-striatal network in FTLD‘
There were a lot of findings in this study which obviously generated a lot of data. However the conclusions above focus on answering a specific question from the data set – is there an anatomical difference between the brain atrophy that is seen in Alzheimer’s Disease (AD) and that seen in Frontotemporal Lobar Degeneration (or FTLD for short)? The researchers are quite clear in their conclusions. In AD there is degeneration in a posterior parietal network and in FTLD there is degeneration in a fronto-insular-striatal network. The temporal lobes are presumably being omitted from the conclusions because the temporal lobes are affected in both conditions.
In terms of selection of subjects, the researchers have excluded Corticobasal Degeneration (CBD) and Progressive Supranuclear Palsy (PSP) due to the pattern of anatomical involvement (see (Sha et al, 2006) for a discussion both for and against the amalgamation of FTLD, PSP and Corticobasal degeneration). This exclusion would be expected to reduce the parietal involvement in the FTLD group however as the researchers note that CBD has an asymmetrical parietal lobe involvement. The researchers then sample a subgroup of subjects that had undergone postmortem at the University of California San Francisco Memory and Aging Centre with confirmed diagnosis of AD or FTLD. I wasn’t entirely clear on how the sampling was done (e.g. was it consecutive post-mortems?) and so there may be another source of selection bias. A number of subjects were excluded for a variety of reasons including movement artefact during the antemortem MRI studies. Antemortem dementia severity was assessed using the Clinical Dementia Rating Scale. Diagnosis was obtained antemortem using the NINCDS-ADRDA criteria for AD which I understand are quite robust and the Neary consensus criteria for ‘FTD clinical syndromes’ (other criteria for FTD have been developed subsequently).
A 1.5 Tesla MRI scanner was used and the earliest scans for subjects were included in the analysis. The authors used a previously described protocol for the Voxel Based Morphometry. Multiple comparisons were made of the voxels – AD with FTLD, AD with controls, FTLD with controls and so on. The researchers corrected for multiple comparisons by using family-wise error correction and the analysis was undertaken using SPM2. A protocol was also used for the post-mortems and the researchers specify NIA-Reagan criteria for AD and McKhann work group diagnostic algorithm for FTLD. SPSS was used for the statistical analysis and ANOVA, t-tests or Chi-squared tests were used depending on the data type.
The results of the VBM are displayed very succinctly. Comparison of AD and FTLD were shown very clearly. Both AD and FTLD had regions where the grey matter volume was on average greater than in the other group. However I was surprised to see what was written about the correction for multiple comparisons. After the correction for multiple comparisons the only significant finding was that the grey matter volume in the striatum was larger in AD than in FTLD.
So does this negate the conclusions drawn in the abstract quoted above? It seemed so to me. I double checked the results and then looked at the discussion. In the discussion, the researchers actually comment on the correction for multiple comparisons. They write that the uncorrected anatomical differences between AD and FTLD fitted with their a priori hypotheses. These same conclusions though surely go against the assumptions in the methodology. If the correction for multiple comparisons is unimportant, why include them?
I wasn’t convinced and came away thinking that the main finding was that striatal degeneration may distinguish AD and FTLD. There are clauses though, even with this conclusion. Ultimately there was a small sample size. The confidence interval would have been wide for the volume differences which may have explained the lack of significance after multiple comparisons. The researchers give nice explanations for their hypothesised differences between AD and FTLD and it may well turn out that they are right but I don’t think this is the study to show that. Maybe this data can be aggregated with other data in a later meta-analysis (a case for an open-source imaging database for dementia studies (as in the Alzheimer’s Disease Neuroimaging Initiative).
Sha S, Hou C, Viskontas IV, Miller BL. Nat Clin Pract Neurol. 2006 Dec;2(12):658-65. Are frontotemporal lobar degeneration, progressive supranuclear palsy and corticobasal degeneration distinct diseases?
You can follow ‘The Amazing World of Psychiatry’ Twitter by clicking on this link
You can listen to this post on Odiogo by clicking on this link (there may be a small delay between publishing of the blog article and the availability of the podcast).
If you have any comments, you can leave them below or alternatively e-mail firstname.lastname@example.org
The comments made here represent the opinions of the author and do not represent the profession or any body/organisation. The comments made here are not meant as a source of medical advice and those seeking medical advice are advised to consult with their own doctor. The author is not responsible for the contents of any external sites that are linked to in this blog