OBJECTIVE: Implantation of encapsulated glial cell line-derived neurotrophic factor-secreting cells into brain parenchyma reduces histological brain damage following hypoxic-ischemic stress in neonatal rats. We examined the effect of glial cell line-derived neurotrophic factors on long-term learning and memory impairment and morphological changes up to 18 weeks after hypoxic-ischemic stress in neonatal rats. STUDY DESIGN: Baby hamster kidney cells were transfected with expression vector either including (glial cell line-derived neurotrophic factor-hypoxic-ischemic group; n = 10) or not including (control-hypoxic-ischemic group; n = 8) human glial cell line-derived neurotrophic factor cDNA, encapsulated in semipermeable hollow fibers, and implanted into the left brain parenchyma of 7-day-old Wistar rats. Two days after implantation the rats received hypoxic-ischemic stress, and their behavior was then examined in several learning tasks: the 8-arm radial maze, choice reaction time, and water maze tasks, which examine short-term working memory, attention process, and long-term reference memory, respectively. The rats were killed 18 weeks after the hypoxic-ischemic insult for evaluation of brain damage. Two additional control groups were used: the control group (n = 15), which underwent no treatment, and the glial cell line-derived neurotrophic factor group (n = 6), which underwent implantation of the glial cell line-derived neurotrophic factor capsule but did not undergo hypoxic-ischemic stress. RESULTS: The decrease in the size of the cerebral hemisphere was significantly less in the glial cell line-derived neurotrophic factor-hypoxic-ischemic group, compared with the control-hypoxic-ischemic group, and improved performance was observed in all three tasks for the glial cell line-derived neurotrophic factor-hypoxic-ischemic group: for the control-hypoxic-ischemic group versus the glial cell line-derived neurotrophic factor-hypoxic-ischemic group, respectively, in the 8-arm radial maze test, average number of correct choices was 6.2 +/- 0.1 versus 6.9 +/- 0.1 ( P < .01); in the choice reaction time test, average reaction time for a correct response was 2.35 +/- 0.1 seconds versus 1.97 +/- 0.09 seconds ( P < .01); in the water maze test, average swimming length was 1120.0 +/- 95.2 cm versus 841.6 +/- 92.1 cm ( P < .01). All results for the glial cell line-derived neurotrophic factor group were similar to those for the control group. CONCLUSION: Glial cell line-derived neurotrophic factor treatment is effective in not only reducing brain damage but also inhibiting learning and memory impairment, following hypoxic-ischemic insult in neonatal rats. No adverse effects in learning and memory tests were observed in the glial cell line-derived neurotrophic factor group.