Background: Autograft is the best option in nerve defects when end-to-end repair can not sufficiently preserve nerve continuity. Theoretically, if the severed nerve is reversely grafted, it may prevent axonal growth into nerve branches, and larger amounts of axons will reach the target organ and more satisfactory results will be obtained. In this study we aimed to compare conventional versus reverse nerve grafting.Methods: This study was performed in Animal laboratory of Hazrat Fatemeh Hospital from April till August 2011. We randomly divided 40 Wistar rats into two groups. We excised 1.5 cm of the right sciatic nerve and anastomosed it conventionally between the proximal and distal ends of the nerve in rats in group A and in a reverse manner in rats in group B. The rats’ footprints were recorded in the first and 16th weeks after surgery. In week 16, the grafted nerves were removed under anesthesia for pathological examination and axon count. Subsequently, the results were compared clinically by sciatic functional index (SFI) through footprint analysis and paraclinically by axon count. A p-value smaller than 0.05 was considered statistically significant.Results: Conventional and reverse nerve grafting no had statistically significant differences in clinical assessment in the first and 16th weeks (P=0.87) post-surgically and also no difference in paraclinical assessment in week 16 (P=0.68). Conclusion: We had no significant clinically or para clinically differences between two approaches. It should be considered that the diameter and length of nerves and muscles in human is larger than rats, so the results of nerve repair may differ in human. We suggest a study in animal model which is anatomically more similar to human.