Electrophysiological and other auxiliary investigations in patients with Parkinson's disease dementia
Yasuyuki Okuma and Yoshikuni Mizuno
Patients with Parkinson’s disease (PD) often show cognitive deficits. Impairments have been documented in almost all areas of cognition, including general intellectual functions, visuo- spatial functions, attention, memory, and executive functions [1-4]. Various electrophysiological investigations have been used to study these cognitive dysfunctions in patients with PD. In this chapter, the results of electrophysiological investigations in PD patients with and without dementia are reviewed. In addition, data on 123I-metaiodobenzylguanidine ([123I]-MIBG) myocardial scintigraphy, which has been found to be very useful for differentiating Lewy body diseases such as PD and dementia with Lewy bodies (DLB) from Alzheimer’s disease (AD) [5, 6], are also reviewed.
Early studies showed that EEG findings are abnormal in around 30-40% of PD patients. The slowing of the alpha rhythm and a generalized or focal increase in slow-wave activities are the common findings, particularly in patients with dementia [7-13]. Tanaka et al.  conducted EEG power spectral analyses and their results agreed with the previous findings that delta and theta power increased in PD patients with dementia (PD-D), but they did not see a decreased alpha power compared with controls. Recently, Kamei and colleagues [15-18] made an extensive quantitative EEG (qEEG) study by employing multiple logistic regression analysis, as well as estimating the distribution of frequency changes, in a large number of PD patients without brain ischaemia. They showed diffuse slowing except for the frontal pole in the qEEG compared with age-matched healthy controls  and a decreased spectral ratio at all electrode locations with progression of the disease . They also studied qEEG alterations in PD patients with executive dysfunction (ExD) and cognitive impairment; they found an increase in slow-wave activity in frontal and frontal- pole locations in ExD  and at all brain areas in patients with dementia . Klassen et al.  reported qEEG findings such as decreased frequency of background rhythm and relative power in the theta band being predictive biomarkers for incident PD-D. The underlying pathophysiology of EEG changes in PD is difficult to elucidate. Both subcortical and cortical involvement may account for the EEG slowing, Soikkeli et al.  suggested that deficits in cholinergic transmission could in part explain the slowing in EEG, particularly in PD-D patients.
To explore whether EEG abnormalities can discriminate between DLB, AD, and PD-D in the early stages, Bonanni et al.  studied EEG in 50 DLB, 50 AD, and 40 PD-D patients with earliest stage of dementia (MMSE 20 <) at first visit. Dominant mean frequencies were 8.3 Hz for the AD group and 7.4 Hz for DLB. The variability of the dominant frequency also differed between the AD (1.1 Hz) and DLB groups (1.8 Hz). By comparison, fewer than half of the patients with PD-D exhibited the EEG abnormalities seen in those with DLB. The authors concluded that if revised consensus criteria for DLB diagnosis  are properly applied, EEG may be helpful for discriminating between AD and DLB in the early stages of dementia. The difference in qEEG findings between PD-D and AD was also investigated by Fonseca et al. . Differences in beta coherence were found between AD and PD-D, with an increase in PD-D and a decrease in AD, along with a greater increase in the slow-wave absolute powers (delta and theta) in PD-D. These neurophysiological differences are likely to be related to distinct mechanisms involved in the pathogenesis of dementia in AD and PD-D.