Vigabatrin (VGB) is an antiepileptic drug (AED) that acts by irreversibly inhibiting gamma-aminobutyric acid transaminase (GABA-T). To evaluate immune responses to GVG, we studied 29 idiopathic or symptomatic epileptic children and also examined a control group (n = 15). Epileptic children were tested before and after 1 and 3 months of VGB treatment. Whole blood was used to connect subsets with commercial monoclonal antibodies. Peripheral blood mononuclear cells (PBMC) were used to assess natural killer (NK) cell activity and lymphocyte response to phytohemagglutinin (PHA) and concavalin A (Con A). Immunoglobulin (Ig) levels were tested in serum. At baseline, no immunologic abnormalities were observed in either control or treated patients. During treatment, the percentage and absolute number of B lymphocytes, serum concentration of Ig, number of T total mature lymphocytes (CD3), T-rosetting lymphocytes (CD11), T-helper cells (CD4), and mitogenic response of lymphocytes remained unchanged. Several other immunologic responses showed a statistically significant increase after 1 and 3 months of VGB treatment, however, including the percentage and absolute number of T-suppressor cells (CD8) and NK cells and NK cell activity. The correlation between number of NK cells and NK cell activity was significant. Data obtained demonstrated that VGB may interfere with the modulation of the immune system, especially cytotoxic cell populations.