No abstract

We had previously observed that in groups of children with low socioeconomic and cultural level the EEG spectral parameters did not correlate with age due to the great variance in the distribution of such EEG parameters. In this paper we have made several statistical analyses in order to look for other factors that may explain the great variance observed. EEG during rest was recorded in 14 monopolar leads from 119 children. After FFT the following measures were calculated: absolute power, relative power, dominant frequency, coherence and left power/right power ratio for the delta, theta, alpha and beta bands in each derivation. For each measure, in each derivation, ANOVA analyses were performed taking age as covariable and sex, presence or absence of antecedents of risk factors and of learning disorders (LD) as independent variables. Age had its main effect on absolute power in the different bands and in the dominant frequency in the alpha band. Sex affected absolute and relative power in the alpha band. Risk factors per se had no effect. Presence of LD showed its main effect on absolute power. Many interactions between risk and LD and sex and LD were observed. Other analyses, such as MANCOVA, corroborated these results: significant differences between girls and boys, significant differences between children with and without LD and no differences between children with and without antecedents of risk factors. Keywrds: EEG. socioeconomic factors, power spectra. learning disorderReprints requests: Thalia Harmony, ENEP-UNAM-lztacala, Apartado 314,54030 Tlalnepantla, Estado de Mexico. Mexico. 123 Int J Neurosci Downloaded from informahealthcare.com by McMaster University on 11/03/14 For personal use only.

It has been shown that EEG maturation can be described in terms of regression equations on age of broad band EEG frequency parameters. In the present paper six groups of children with different economic and psychosocial characteristics of three countries were studied. Regression equations on age of the EEG relative power (expressed in percentages of the total EEG activity) in the delta, theta, alpha and beta bands in 8 bipolar derivations were computed in each group of children and the slopes compared with those previously published by John et al. (1980). Those children who grew up with adequate nutritional, sanitary and cultural environmental conditions showed the same slopes as U.S. and Swedish children selected with strict criteria of normality. Children nourished in poor socioeconomic and sanitary environments and who frequently had pathological personal antecedents with risk factors associated with brain damage showed either a slow maturation of the EEG characterized by smaller slopes of theta relative power or a great variance of EEG parameters and no relation of these parameters to age.

For the past decade the Cuban Neuroscience Center has organized on behalf of the Ministry of Public Health of the Republic, a nationwide Program for the introduction of quantitative EEG (qEEG). This Program has involved a) development of standardized equipment for "paperless" EEG, qEEG and brain topography; b) establishment of a network of 21 laboratories of clinical neurophysiology; c) creation of the specialty of clinical neurophysiology which trains physicians from all provinces in both traditional and quantitative electrophysiological methods; d) introduction of standardized protocols for the collection of clinical and electrophysiological information; e) organization of a national normative and neuropsychiatric database; f) establishment of normative regression equations. Among the special issues discussed are: 1) relationship between traditional and quantitative methods; 2) evaluation of the effectiveness of the technology introduced; 3) use of qEEG in the early detection of brain dysfunction.

To evaluate the hypothesis that quantitative EEG (qEEG) analysis is susceptible to detect early functional changes in familial Alzheimer's disease (AD) preclinical stages. Three groups of subjects were selected from five extended families with hereditary AD: a Probable AD group (18 subjects), an asymptomatic carrier (ACr) group (21 subjects), with the mutation but without any clinical symptoms of dementia, and a normal group of 18 healthy subjects. In order to reveal significant differences in the spectral parameter, the Mahalanobis distance (D 2) was calculated between groups. To evaluate the diagnostic efficiency of this statistic D 2, the ROC models were used. The ROC curve was summarized by accuracy index and standard deviation. The D 2 using the parameters of the energy in the fast frequency bands shows accurate discrimination between normal and ACr groups (area ROC = 0.89) and between AD probable and ACr groups (area ROC = 0.91). This is more significant in temporal regions. Theses parameters could be affected before the onset of the disease, even when cognitive disturbance is not clinically evident. Spectral EEG parameter could be firstly used to evaluate subjects with E280A Presenilin-1 mutation without impairment in cognitive function.