EEG phenotype in alcoholism: increased coherence in the depressive subtype.

OBJECTIVE: Electroencephalography (EEG) power and coherence changes may be trait markers for alcoholism providing clues to brain mechanisms of vulnerability. However, it is unclear whether alpha power and coherence differences reflect reversible toxic or withdrawal effects of alcohol. METHOD: The EEGs of 10 non-abstinent and 16 long-term abstinent alcoholics (7.7 +/- 5.8 years) and 25 controls were analyzed. Levels of anxiety and depression were assessed by questionnaire. RESULTS: No statistically significant EEG power differences were observed between groups, although the numerical difference between alcoholics and controls was similar to that previously reported. Bilateral, intrahemispheric, posterior coherences were significantly increased in the alpha and beta frequency bands both in long-term abstinent and non-abstinent alcohol-dependent subjects - particularly when depressiveness was included as a covariate. CONCLUSION: These results suggest that increased EEG-coherence (cortical synchronization) may serve as endophenotype for alcoholism in conjunction with increased depressiveness and point to a possible involvement of GABAergic and/or glutamatergic neurotransmission.

Alcohol use disorders and anxiety disorders: relation to the P300 event-related potential.

BACKGROUND: The robust association of alcoholism with reduced P300 event-related potential amplitude has been largely established in severely affected alcoholics and their offspring. Few studies have examined the relationship of increased arousal, anxiety, and P300. In this study, we sought to determine whether P300 group differences could be discerned in well functioning individuals with less severe forms of alcohol use disorders and anxiety disorders. We were particularly interested in looking at the subgroup of alcohol use disorders accompanied by anxiety disorders. This subgroup has previously been found to have diminished alpha amplitude in the resting EEG. METHODS: Male and female community volunteers (99 unrelated index participants and 78 relatives) and 21 unrelated volunteers from an anxiety disorder clinic were interviewed by using the Schedule for Affective Disorders and Schizophrenia, Lifetime version. Blind-rated lifetime psychiatric diagnoses were assigned according to DSM-III-R criteria. Auditory and visual P300 event-related potentials were elicited with an oddball paradigm and were recorded at the midparietal (Pz) site. RESULTS: As expected, auditory P300 amplitudes were significantly reduced in participants with alcohol use disorders and significantly increased in participants with lifetime anxiety disorders. However, more detailed analysis revealed that, in an apparent paradox, auditory P300 amplitudes were lowest in individuals with comorbid alcohol use and anxiety disorders and highest in individuals with anxiety disorders alone. Visual P300 amplitudes followed the same trends but were generally not significant. CONCLUSIONS: Even in a sample of largely community-ascertained individuals, auditory P300 amplitude is reduced in alcoholics, particularly those with anxiety disorders, and is highest in nonalcoholics with anxiety disorders.

Association of low-voltage alpha EEG with a subtype of alcohol use disorders.

BACKGROUND: Neurophysiological traits may identify more homogeneous subgroups of alcoholics. Such discoveries could yield information regarding pathophysiological development, leading to more specific preventive measures and treatments. In an earlier study of 127 individuals, 59 of whom were unrelated, we found that a heritable resting Electroencephalographic (EEG) phenotype, i.e., the low-voltage alpha (LVA) trait, was associated with alcohol use disorders and anxiety disorders. METHODS: We evaluated these findings using an independent, similarly established, dataset of 120 subjects. We also extended the study to a larger set of 149 unrelated individuals from a total sample of 247 subjects for whom psychiatric diagnoses and resting EEG phenotypes were available. Blind-rated psychiatric diagnoses were formulated according to DSM-III-R criteria. RESULTS: In the replication sample, the LVA trait was again more common among subjects with anxiety disorders than among those without. In the total group of unrelated individuals, alcoholics were significantly (3 times) more likely to show the LVA trait than were nonalcoholics. Again, individuals with anxiety disorders were significantly (3 times) more likely to exhibit the LVA trait than were those without anxiety disorders. Of 11 unrelated alcoholics with anxiety disorders, seven showed the LVA trait. It was specifically the LVA trait and not low-amplitude alpha activity that was associated with alcohol use disorders. CONCLUSIONS: The results of this replication study and the analysis of the total sample of unrelated individuals support an association between LVA EEG and the subtype of alcohol use disorders associated with anxiety disorders. The LVA phenotype may be a vulnerability factor for alcohol use disorders and anxiety disorders.

Effects of alcohol use and gender on the dynamics of EKG time-series data.

Hurst analysis of EKG data obtained from a population of alcoholic (n = 13) and nonalcoholic (n = 48) subjects was undertaken. Potential subjects (n = 120) were screened using the Schedule for Affective Disorders and Schizophrenia and Structured Clinical Interview for DSM-III instruments. Data from subjects with a diagnosis of current alcohol dependence were analyzed. Subjects with diagnoses such as major depression, bipolar disorder or schizophrenia (Axis I diagnoses), or personality disorders (Axis II diagnoses) were excluded from analysis. Subjects undergoing testing were free of alcohol and illicit drugs. Alcoholic subjects had no clinical evidence of alcohol withdrawal symptoms at the time of testing. EKG data were obtained with eyes open or with eyes closed. Approximately 3.5 min of data were obtained for each condition. Alcoholic subjects had less complex heart rate dynamics as evidenced by higher values of H = 0.18 +/- 0.05 (mean +/- SEM), compared with healthy comparison subjects with H = 0.09 +/- 0.02, p < 0.014 for the eyes closed condition, and H = 0.17 +/- 0.05 (mean +/- SEM) compared with healthy comparison subjects with H = 0.07 +/- 0.02,p < 0.011 for the eyes open condition. A gender effect was seen, with female subjects showing evidence of more complex heart rate dynamics than male subjects.