A large number of metabolic alterations are increasingly being treated with growth hormone. Despite the fact that growth hormone is known to be the main regulator of several hepatic drug metabolizing enzymes in rodents, few studies deal with the effect of growth hormone on hepatic enzyme activities in human beings. The aim of this study was to determine the effects of growth hormone replacement therapy for 4 weeks on CYP2A6 activity in children, because changes in this enzyme activity may have important therapeutic and toxic consequences. A total of 31 growth hormone-deficient children (age range 4.1-13.1 years; mean age 9.88 +/- 2.89 years) participated. The genotypes of CYP2A6 gene, CYP2A6*1A, CYP2A6*1B, CYP2A6*4, CYP2A6*1x2 and CYP2A6*9, were determined by polymerase chain reaction. To assess the enzyme activity, we used caffeine as a probe drug at two points in time: before starting growth hormone therapy (Day 0) and after 4 weeks of growth hormone therapy (Day A). Caffeine and metabolite concentrations in urine were assayed by high-pressure liquid chromatography. The metabolite ratio 1,7-dimethilxanthine to 1,7-dimethylurate (17U/17X) served to indicate CYP2A6 activity. Median value and 95% confidence interval at baseline was 1.08 (0.98-1.24). The value after treatment was 1.08 (0.86-1.21). Data comparison between periods showed lack of statistically significant differences (P > 0.05). The relative change, measured by the ratio of medians and 90% confidence interval, was 1.02 (0.84-1.19). There were no significant differences when the ratio between genotype groups were compared. These results indicate that growth hormone replacement therapy of growth hormone-deficient children for 4 weeks does not modify the CYP2A6 activity and hence the efficacy or toxicity of the CYP2A6 substrate compounds.