[ERPs (N200, P300 and CNV) in alcoholics: relapse risk assessment]

The cognitive event-related potentials were studied in a group of 55 alcoholic patients, paired in age and sex with a group of 18 control subjects, using a protocol oddball (visual and auditory) and a protocol VCN Go/Nogo. The N200 obtained using an auditory oddball paradigm had lower amplitude in alcoholics than in controls. A significant amplitude decrease of visual P300 was also observed in alcoholic male subjects. There was however no difference in auditory P300 between alcoholics and control subjects. Using a Go/Nogo paradigm, a significant difference on the final part of the VCN appears between alcoholic and pilot subjects. In addition, the longitudinal follow-up of the same alcoholic patients showed an electrophysiological profile that allowed dividing them into two different groups. On the initial recording (17 days after weaning), the auditory oddball P300 amplitude was significantly higher at Cz and Pz among patients who relapsed during the 3 months follow-up. The same effect appeared on the CNV protocol, where the amplitude of P300 was higher in patients who subsequently relapsed than for those who remained abstinent. Cognitive ERPs may be clinically useful to improve the prediction of risk of relapse among alcoholic patients.