EEG abnormalities in frontotemporal lobar degeneration

Methods.

Patients were recruited from the Specialist Cognitive Disorders Clinic, National Hospital for Neurology and Neurosurgery, London, UK. Sixty-four patients were diagnosed with FTLD in accordance with the diagnostic guidelines for FTLD.5 Twenty patients were diagnosed with probable AD according to National Institute of Neurological and Communication Disorders and Stroke/Alzheimer’s Disease and Related Disorders Association criteria.6 AD patients were matched with FTLD patients in terms of age, gender, and disease severity.

The presence of generalized or focal frontotemporal atrophy on MRI was considered supportive of the diagnoses of AD and FTLD, respectively, but the clinical diagnosis itself was not amended as a consequence of the neuroimaging appearances.

FTLD patients were subdivided into frontal and temporal clinical subgroups. Forty-eight patients (75%) presented with symptoms of frontal lobe dysfunction or with progressive nonfluent aphasia and were classified as having frontal-variant FTLD. Sixteen patients (25%) presented with semantic dementia or with right temporal lobe atrophy and were considered to have temporal-variant FTLD.

This study was performed according to local ethics committee guidelines.

EEG analysis.

EEG recordings were of 20 to 30 minutes’ duration. The EEG were analyzed by two independent neurophysiologists, blinded to the clinical features and diagnosis. Three grades of EEG appearance were specified, ranging from normal (grade 0) to severely abnormal (grade 2), using the appearance of the alpha rhythm as the primary determinant (figure). This reflects the fact that disruption of alpha rhythm is a more sensitive marker of neocortical dysfunction than slow wave changes and thus of greater practical utility in the evaluation of diseases such as FTLD and AD that primarily affect the cerebral cortex.

• Download figure
• Open in new tab
• Download powerpoint

Figure. Examples of differing grades of EEG abnormality. Grade 0 (normal EEG): well-preserved alpha rhythm of normal frequency for age and no excess theta or delta activity. Grade 1 (mildly abnormal EEG): alpha rhythm present but reduced in frequency, or there is an excess of theta or delta activity, or both. Grade 2 (severely abnormal EEG): alpha rhythm sparse or absent; the EEG is dominated by theta or delta activity or both.

The Clinical Dementia Rating Scale (CDR)7 was used to grade dementia severity at the time of the EEG and was determined by retrospective analysis of the case notes.

The Mann–Whitney test was used to compare EEG grades between patient groups. The Spearman rho test was used for correlational analyses.

Results.

The table summarizes the EEG abnormalities in the FTLD and AD patient groups. A comparison of the EEG appearances between the FTLD and AD patient groups did not reveal any significant differences (z = −1.09, p = 0.28).

Similar proportions of frontal- and temporal-variant FTLD cases were associated with normal EEG. There was a greater percentage of severely abnormal EEG in the temporal group (25 vs 19% in the frontal FTLD group), but this difference did not reach significance (p = 0.62). Of the 22 patients with disorders of speech and language (i.e., progressive nonfluent aphasia or semantic dementia), only 4 (18%) showed significant left hemisphere lateralization of EEG abnormalities.

Seven FTLD patients were on medications that could potentially affect the EEG. However, three had normal EEG, and the remaining four patients had preserved alpha rhythm and minor EEG abnormalities, of a degree insufficient to affect the EEG grading.

Discussion.

EEG abnormalities were observed in the majority of patients with FTLD, with the degree of EEG abnormality correlating with dementia severity. There was no significant difference in the EEG appearances observed in FTLD patients and in AD patients of similar dementia severity. Normal EEG were present in a similar proportion of patients presenting with frontal and temporal symptoms. In those FTLD cases with a pathologic diagnosis, normal EEG were approximately twice as commonly observed in those patients with atrophy primarily affecting the frontal lobes as in those with predominant temporal lobe damage.

These study data complement those obtained in a previous study using quantitative EEG techniques to compare FTLD and AD patients.8 The results of this study conflict with the view that focal degenerative disease of the frontal and temporal lobes is associated with a normal EEG. Two important observations may serve to explain this apparent discrepancy. First, a review of the initial studies of the EEG appearance in Pick disease reveals that EEG abnormalities were, in fact, present in a proportion of patients.1,4⇓ Second, and more pertinently, subsequent advances made in identifying the various clinical syndromes associated with frontotemporal degeneration mean that the modern concept of FTLD exceeds the historical diagnosis of “Pick disease” not only in the range of histopathologic diagnoses but also in the extent of associated cortical damage, particularly with regard to the degree of temporal lobe involvement, as exemplified by the syndrome of semantic dementia.9 This study has therefore provided a reappraisal of the EEG appearances in FTLD to correspond with these changes in the clinical characterization of FTLD.

The degree of EEG abnormality has been demonstrated to correlate with dementia severity in the FTLD patients. Given that the degree of EEG disturbance in AD also correlates with disease severity,10 this suggests that the EEG appearance may be a poor discriminator of the two disorders at similar stages of dementia severity. In this study, most patients had dementia of mild to moderate severity, and therefore the possibility remains that the EEG does differ in severe FTLD and AD. However, any diagnostic tool is of greatest utility in the earliest stages of disease when clinical diagnosis may be difficult, and at this stage of diagnostic appraisal, the EEG does not readily differentiate FTLD from AD.