ABSTRACT
An extensive animal literature suggests that excessive corticosteroid exposure is associated with changes in memory and the hippocampus. Agents that decrease glutamate attenuate corticosteroid effects on the hippocampus. Minimal data are available on preventing or reversing corticosteroid effects on the human hippocampus. We previously reported that open-label lamotrigine was associated with significant improvement in declarative memory in corticosteroid-treated patients. We now examine the impact of 24 weeks of randomized, placebo-controlled lamotrigine therapy on declarative memory (primary aim) and hippocampal volume (secondary aim) in 28 patients (n=16 for lamotrigine, n=12 for placebo) taking prescription corticosteroids. All participants with data from at least one postbaseline assessment (n=9 for lamotrigine, n=11 for placebo) were included in the analysis. Declarative memory was assessed with the Rey Auditory Verbal Learning Test (RAVLT) at baseline and weeks 12 and 24. Hippocampal and total brain volumes were manually traced from MRI scans obtained at baseline and week 24. On the basis of an ANCOVA analysis, total words learned on the RAVLT at exit were significantly greater in the lamotrigine group (n=8, missing data or dropouts n=8) compared to the placebo group (n=11, dropout n=1). RAVLT scores in the lamotrigine group increased from mildly impaired to average range. Hippocampal volume changes were small in both lamotrigine (n=7) and placebo (n=7) groups during the 24-week assessment period and between-group differences were not significant. Results suggest that lamotrigine may improve declarative memory in patients taking prescription corticosteroids although differential dropout rate in the two groups is a concern.