Prefrontal EEG slowing, synchronization, and ERP peak latency in association with predementia stages of Alzheimer’s disease

Background: Early screening of elderly individuals who are at risk of dementia allows timely medical interventions to prevent disease progression. The portable and low-cost electroencephalography (EEG) technique has the potential to serve it.

Objective: We examined prefrontal EEG and event-related potential (ERP) variables in association with the predementia stages of Alzheimer’s disease (AD).

Methods: One hundred elderly individuals were recruited from the GARD cohort. The participants were classified into four groups according to their amyloid beta deposition (A+ or A−) and neurodegeneration status (N+ or N−): cognitively normal (CN; A−N−, n = 27), asymptomatic AD (aAD; A + N−, n = 15), mild cognitive impairment (MCI) with AD pathology (pAD; A+N+, n = 16), and MCI with non-AD pathology (MCI(−); A−N+, n = 42). Prefrontal resting-state eyes-closed EEG measurements were recorded for five minutes and auditory ERP measurements were recorded for 8 min. Three variables of median frequency (MDF), spectrum triangular index (STI), and positive-peak latency (PPL) were employed to reflect EEG slowing, temporal synchrony, and ERP latency, respectively.

Results: Decreasing prefrontal MDF and increasing PPL were observed in the MCI with AD pathology. Interestingly, after controlling for age, sex, and education, we found a significant negative association between MDF and the aAD and pAD stages with an odds ratio (OR) of 0.58. Similarly, PPL exhibited a significant positive association with these AD stages with an OR of 2.36. Additionally, compared with the MCI(-) group, significant negative associations were demonstrated by the aAD group with STI and those in the pAD group with MDF with ORs of 0.30 and 0.42, respectively.

Conclusion: Slow intrinsic EEG oscillation is associated with MCI due to AD, and a delayed ERP peak latency is likely associated with general cognitive impairment. MCI individuals without AD pathology exhibited better cortical temporal synchronization and faster EEG oscillations than those with aAD or pAD.

Significance: The EEG/ERP variables obtained from prefrontal EEG techniques are associated with early cognitive impairment due to AD and non-AD pathology. This result suggests that prefrontal EEG/ERP metrics may serve as useful indicators to screen elderly individuals’ early stages on the AD continuum as well as overall cognitive impairment.