BACKGROUND: Traumatic approaches, such as sacrifice and perfusion sampling, have been used to evaluate efficiency of stem cell transplantation. However, these methods are not applicable to human studies. Cell tracing, in combination with non-invasive imaging technology, can be utilized to trace cell survival following transplantation to evaluate the efficacy of cell transplantation therapy. OBJECTIVE: To explore feasibility of magnetic resonance imaging (MRI) to observe in vivo repair of injured sciatic nerves following feridex and polylysine (FE-PLL) complex-labeled bone marrow stromal cell (BMSC) transplantation. DESIGN, TIME AND SETTING: A randomized, controlled, animal experiment was performed at the Laboratory of the Department of Neurosurgery, Zhujiang Hospital from March to December 2008.MATERIALS: Feridex was purchased from Advanced Magnetic, USA, and polylysine was purchased from Sigma, USA.METHODS: BMSCs were harvested from adult rabbit femurs and were cultured in vitro with neural stem cell culture medium, leukemia inhibitory factor, and basic fibroblast growth factor. Bone marrow stromal cell-derived neural stem cells (BMSC-D-NSCs) were obtained and labeled with FE-PLL complex. The right sciatic nerve (0.8 mm) was excised from healthy, New Zealand rabbits, aged 1.5 months, and the epineuria of distal stumps underwent tover and were anastomosed at the proximal ends. FE-PLL labeled BMSC-D-NSC suspension or culture medium was transplanted into the epineurial lumen using a microsyringe. The left sciatic nerve was left intact and served as the normal control. MAIN OUTCOME MEASURES: Cellular morphology, proliferation, and differentiation, as well as expression of nestin and neuron-specific enolase (NSE), of BMSCs-D-NSCs were observed. Efficacy of FE-PLL labeling and effects on cells were measured. In addition, neural regeneration at 2, 8, and 16 weeks following transplantation was observed by MRI. Histopathology and mean number of regenerated nerve fibers in the proximodistal-injured sciatic nerve were evaluated by hematoxylin and eosin and Bielschowsky staining. RESULTS: Results demonstrated that BMSCs expanded, proliferated, and differentiated into neural-like cells with slim, long processes. The cells expressed nestin and NSE, as detected by immunocytochemistry. BMSC-D-NSCs were effectively labeled by FE-PLL, with a labeling efficiency of 98%. In addition, cell viability was not influenced by the FE-PLL complex. MRI results revealed low signals in the FE-labeled BMSC-D-NSC-implanted region of the sciatic nerve. A low-signal region was observed at 2 weeks, which was widely spread at 8-16 weeks after cell transplantation. The regenerated nerve fibers were orderly arranged in the cell transplantation group and exhibited no significant differences compared with the normal control side (P > 0.05). CONCLUSION: BMSCs were successfully cultured in vitro, and the cells proliferated and trans-differentiated into neuronal-like cells, which expressed nestin and NSE. The FE-PLL complex effectively labeled rabbit BMSC-D-NSCs in vitro and did not affect peripheral neural regeneration following cell transplantation. Results demonstrated that MRI could be used to track FE-labeled BMSC-D-NSCs transplanted in the sciatic nerve.