The present study provides in vivo trials of silicone rubber chambers filled with different concentrations of bilobalide (0, 50, 100, 200, 400 microM) to bridge a 15 mm sciatic nerve defect in rats. Collagen was also filled in the chambers to prevent the bilobalide from leakage. Histological and electrophysiological techniques were used to evaluate the functional recovery of the nerve. At the conclusion of 8 weeks, animals from the group treated with the bilobalide, especially at the concentration of 200 microM, had a higher rate (40%) of successful regeneration across the wide gap and a significantly larger number of myelinated axons (4094 +/- 1555), compared to only 10% and 2485 in the control group. However, the high dose bilobalide (400 microM) completely reversed this positive effect of growth-promoting capability and inhibited nerve regeneration. Only 10% of the animals treated with the high dose bilobalide had regenerated cables within the silicone rubber chambers. These results indicated that bilobalide could be involved in both positive and negative effects on regenerating nerves. Therefore, whether a proper dosage of bilobalide is used or not plays a critical factor in deciding if it can sustain nerve regeneration over long gaps.