Diagnosis and management of dementia with Lewy bodies
Third report of the DLB consortium

In the last revision of criteria for the diagnosis of dementia with Lewy bodies (LBD) [1], temporal lobe transient slow waves (in the text) or sharp waves (in table 1) on EEG was reported as a supportive feature for the diagnosis of LBD.

The only quoted reference is a review [2] reporting that "standard EEG may show early slowing, epoch-by-epoch fluctuation, and transient temporal slow-wave activity." The first reference quoted in the review [3] described EEG recordings in 11 Alzheimer disease (AD) and in 14 LBD patients. All but two patients showed "abnormal EEG with alpha activity decreased outside the normal frequency." Preservation of alpha activity was more common in AD, 55% than in LBD, 36%, p=0.45. 55% AD and 79% LBD patients had delta components. Temporal lobe transients were seen more often in LBD patients (50%) than in AD patients (18%) (p=0.11).

The second reference [4] described normal EEG patterns in three AD patients and in no LBD patients, mild but excessive theta activity, bitemporal spikes, or both in one AD patient and in one LBD patient, moderate excessive theta activity diffusely or bursts of bisynchronous, diffuse rhythmic 2 to 3Hz waves in 14 AD and 13 LBD patients and one or more persistent diffuse delta, bisynchronous spikes or sharp waves, and triphasic waves in two AD and six LBD patients. There was no significant difference in the EEG patterns between the two groups.

The authors stated that EEG is not diagnostic of LBD, failing to differentiate it from AD. The third reference was a single case report. Only one Quantitative EEG study is quoted in the review [5], describing variability of mean frequencies in LBD, but only 15 LBD and 15 AD were studied: frequency variability was expressed as difference in standard deviations of the mean frequencies, making the real epoch-to-epoch variations unclear. This difference was 2.91 LBD, 2.42 AD, 2.06 controls, meaning that the frequency variability in controls was about 3Hz. This is not surprising if the variability represents intra-group differences, but is unacceptable if it represents variations in the single controls.

The only figure shows EEG map plots with variability in the 5-22Hz range, incongruent with descriptions of theta-delta activity. Therefore, literature on EEG recordings in LBD includes anecdotal reports or inconclusive findings. It could be more appropriate, for a consensus statement, to conclude that finding EEG abnormalities does not exclude the diagnosis of LBD, instead of asserting that EEG abnormalities support LBD diagnosis.

References

1. McKeith IG, Dickson DW, Lowe J et al. Diagnosis and management of dementia with Lewy bodies. Third report of the LBD consortium. Neurology 2005 October 19, as doi:10.1212/01.wnl.0000187889.17253.b1.

2. McKeith I, Mintzer J, Aarlsand D, et al, Dementia with Lewy bodies. Lancet Neurol 2004; 3: 19-28.

3. Briel RCG, McKeith IG, Barker WA, et al. EEG findings in dementia with Lewy bodies and Alzheimer's disease. J Neurol Neurosurg Psychiatry 1999; 66: 401-403.

4. Barber PA, Varma AR, Lloyd JJ, Haworth B, Snowden JS, Neary D. The electroencephalogram in dementia with Lewy bodies. Acta Neurol Scand 2000; 101: 53-56.

5. Walker MP, Ayre GA, Cummings JL, Wesnes K, McKeith IG, O'Brien JT, Ballard CG. Quantifying fluctuation in dementia with Lewy bodies, Alzheimer's disease, and vascular dementia. Neurology 2000; 54:1616-1624.

The authors report no conflicts of interest.

We thank Bonanni et al for their interest in our article. Features listed as "supportive" of a clinical diagnosis of DLB in the revised clinical diagnostic criteria are defined in the text of the paper as being "commonly present in DLB but (which) lack sufficient diagnostic specificity to be categorised as core or suggestive". [1]

The evidence regarding EEG abnormalities in DLB is limited. Published reports, which were previously summarized, [2] suggest that the standard EEG is often abnormal with slowing into theta and delta ranges and with a tendency to temporal lobe abnormalities including slow wave transients, and bisynchronous spikes, sharp and triphasic waves. No single pattern is typical. Our inclusion of such features as supportive of a DLB diagnosis does not imply that their presence can reliably differentiate DLB from AD or any other dementia. We merely suggest that their presence is consistent with, i.e. supportive of, a clinical diagnosis of DLB.

Dr Bonanni's suggestion that we might better have worded our recommendation as "finding EEG abnormalities does not exclude the diagnosis of LBD" is a more conservative way of saying the same thing but one that would probably be of less benefit in helping clinicians to improve their detection of DLB.

More research is needed, although it seems unlikely that the standard EEG will play a major part in DLB diagnosis. Other investigative techniques including functional neuro-imaging and cardiac scintigraphy appear to have high potential.

The authors report no conflicts of interest.