The rate of abnormality (83%) in this study indicates that prevalence of EEG abnormalities in cognitively and physically healthy oldest-old is very high. In younger healthy elderly both background alpha rhythm slowing and temporal intermittent polymorphic slowing have been observed in varying portions of participants (for a comprehensive review, see Klass & Brenner, 1995
).The few previous EEG studies with non-symptomatic participants aged 90 and older found widely varying levels of EEG abnormalities (Obrist, 1954
; Hubbard et al., 1976
; Oken & Kaye, 1992
). For example, a study with extremely healthy subjects over 84 (taking no medications) found that 52% of the participants had intermittent temporal slowing (Oken & Kaye, 1992
). A study with less healthy seniors aged 80-94 found that only 17% had delta activity (Obrist, 1954
). However, this study only had three participants aged 90 and older so the lower rate of abnormalities could be due to the younger age of the participants. The rates of abnormalities in the current study are higher than those found in previous studies, consistent with the extremely advanced age of the participants and the lack of stringent health requirements.
Temporal slowing, a relatively common abnormality in the elderly (e.g., Arenas, Brenner, & Reynolds, 1986
; Shigeta et al., 1995
), was the most frequent in the present study. The cause of temporal slowing is still unknown but one study found that it was related to deep cerebral white matter hyperintensities in participants over 84 (Oken & Kaye, 1992
). No relationship between temporal slowing and history of hypertension or any other cardiovascular factor was found in the present study. However, the small sample size used in this study makes this examination difficult. For example, only one participant had history of congestive heart failure and only one had history of atrial fibrillation. Larger sample sizes will ultimately be necessary to determine whether a relationship between cardiovascular risk factors and temporal slowing exists in the oldest-old. A disproportionate number of the temporal slowing abnormalities found in this study were lateralized, particularly the left side, despite the fact that these participants did not manifest focal neurological or neuropsychological abnormalities. Other studies have reported similar findings with left lateralization (Hubbard et al., 1976
; Torres et al., 1983
; Arenas, Brenner, & Reynolds, 1986
), however, the reason remains uncertain.
The significance of background alpha slowing (<8Hz) in the oldest-old remains in dispute. Some researchers have proposed that background alpha decreases during normal aging but most agree that below 8Hz is abnormal (Klass & Brenner, 1995
). Background frequency has been shown to decrease in people with dementia and depression. Interestingly, participants in the current study with background slowing were older than those without suggesting an association with age; no difference in depression or cognitive scores was found. A previous EEG study in healthy participants over 84 found 23% to have alpha slowing (Oken & Kaye, 1992
). The slightly higher rate in the current study (33%) is in keeping with the older age of the participants.
The most unexpected finding in this study, TIRDA, is of uncertain significance. While rhythmic EEG patterns have been related to drowsiness in older adults (frontal rhythmic delta activity, Santamaria & Chiappa, 1987
), the TIRDA recorded in the current study was present during the awake state. TIRDA has generally been examined as interictal activity present in some people with a history of temporal lobe epilepsy(Normand, Wszolek, & Klass, 1995
). More specifically, one study found that TIRDA was related to complex partial seizures; TIRDA was present in 27% to 35% of recordings (awake and sleep, respectively) in younger people with partial epilepsy (Reiher et al.,1989
). The incidence of epilepsy, particularly partial epilepsy, is known to increase exponentially in the elderly (Hauser et al., 1993
; Olafsson et al., 2005
). There are two possible explanations for the TIRDA in the current study. First, in the oldest-old, brain areas predisposed for seizures may manifest subtle epileptiform EEG abnormalities (such as TIRDA) rather than the usual spikes and sharp waves. The participants in the current study had no history of seizures or spells but they may be at a higher risk than those without TIRDA. Second, in this age group, TIRDA may be an age-related EEG finding without epileptiform significance. Future research in this group of oldest-old and others will be needed to elucidate these issues.
This study indicates that caution is required when using EEG to diagnose seizures and encephalopathies in the elderly given the evidence of a high background rate of EEG abnormalities in healthy elderly. As the population of extremely aged persons grows, more research about this special group will be needed. Additional studies with larger groups of elderly individuals and longer follow up are required to corroborate the findings of this study, and to elucidate the clinical significance.