To investigate the effects of valproic acid (VPA) on renal tubular function, we examined 15 ambulatory children with epilepsy who received VPA for at least 6 months. None of the patients had mental retardation. Fourteen age- and sex-matched children were used as a control group. No statistically significant differences were. found between patients and control subjects with respect to blood urea nitrogen (BUN), creatinine (Cr), uric acid, creatinine clearance (C-Cr), tubular reabsorption of phosphorus (TRP), urinary Ca:creatinine ratio, urinary pH and mean urinary beta (2)-microglobulin concentrations (P>0.05). Protein and glucose in patient urine samples were negative. Urine microscopic examinations and amino acid chromatographies of patients were also normal. However, significant differences were found between patient and control groups with respect to mean urinary N-acetyl-beta-D-glucosamine:creatinine ratio (NAG:Cr) and mean urinary malondialdehyde:creatinine (MDA:Cr) ratio (P<0.05). In conclusion, ambulatory children with epilepsy taking VPA therapy may develop proximal renal tubular dysfunction. Although this finding is clinically insignificant, it should be kept in mind during VPA therapy.