Neurocognitive performance, alcohol withdrawal, and effects of a combination of flumazenil and gabapentin in alcohol dependence

Abstract

Background: Among some alcohol-dependent individuals, early alcohol abstinence is marked by alcohol withdrawal (AW), a phenomenon mediated by GABA and glutamate signaling. We previously reported that a combination of 2 medications that affect GABA and glutamate tone, gabapentin and flumazenil, more effectively reduced drinking among individuals with higher pretreatment AW (Anton et al., 2009). This study evaluated whether this finding is related to changes in neurocognitive performance, which is also affected by cortical GABA and glutamate tone.

Methods: Neurocognitive performance was assessed at baseline and twice during the first week of treatment among 60 alcohol-dependent participants in the previously published clinical trial.

Results: AW was associated with poorer baseline performance on 4 of 8 measures, and individuals with higher baseline AW who received the gabapentin and flumazenil combination demonstrated greater improvement on a measure of response inhibition than those with lower AW or those who received a combination of placebos. Improvement in response inhibition during the first week and medication group interacted in their effect on subsequent drinking, such that improvement predicted greater abstinence only among individuals who received gabapentin and flumazenil. Improvement on other neurocognitive measures was neither differentially impacted by medication or baseline AW nor related to subsequent drinking.

Conclusions: Taken together, these data suggest that acute AW accounts for a small proportion of variance in neurocognitive performance, that gabapentin and flumazenil slightly improve response inhibition during early abstinence, and that such improvement may somewhat reduce later drinking. However, these medications may not affect other neurocognitive domains.

Copyright © 2011 by the Research Society on Alcoholism.

Figures

Figure 1

T-scores for the Stroop interference trial on Days 3 and 7 of the first week of treatment for participants treated with flumazenil and gabapetin (A: FMZ/GBP) and with placebos (B: PLA). Participants who were treated with active medications and who had higher baseline AW demonstrated greater improvement between Days 3 and 7 than those in the other groups. Figures are least-squares means (± standard errors) from the linear mixed model.

Figure 2

Percent days abstinent during the 6-week treatment trial for participants treated with flumazenil and gabapentin (A: FMZ/GBP) and with placebos (B: PLA). Participants who were treated with active medications who had greater than average improvement on the Stroop interference trial between Days 3 and 7 had more abstinence throughout the trial, while the opposite was true for participants treated with placebos. Figures are least-squares means (± standard errors) from the linear mixed model.