To investigate the early effects of acellular xenogeneic nerve combined with adipose-derived stem cells (ADSCs) and platelet rich plasma (PRP) in repairing facial nerve injury in rabbits. The bilateral sciatic nerves of 15 3-month-old male Sprague-Dawley rats were harvested and decellularized as xenografts. The allogeneic ADSCs were extracted from the neck and back fat pad of healthy adult New Zealand rabbits with a method of digestion by collagenase type Ⅰ and the autologous PRP was prepared by two step centrifugation. The 3rd generation ADSCs with good growth were labelled with CM-Dil living cell stain, and the labelling and fluorescence attenuation of the cells were observed by fluorescence microscope. Another 32 New Zealand rabbits were randomly divided into 4 groups and established the left facial nerve defect in length of 1 cm ( n=8). The nerve defects of groups A, B, C, and D were repaired with CM-Dil-ADSCs composite xenogeneic nerve+autologous PRP, CM-Dil-ADSCs composite xenogeneic nerve, xenogeneic nerve, and autologous nerve, respectively. At 1 and 8 weeks after operation, the angle between the upper lip and the median line of the face (angle θ) was measured. At 4 and 8 weeks after operation, the nerve conduction velocity was recorded by electrophysiological examination. At 8 weeks after operation, the CM-Dil-ADSCs at the distal and proximal ends of regenerative nerve graft segment in groups A and B were observed by fluorescence microscopy; after toluidine blue staining, the number of myelinated nerve fibers in regenerated nerve was calculated; the structure of regenerated nerve fibers was observed by transmission electron microscope. ADSCs labelled by CM-Dil showed that the labelling rate of cells was more than 90% under fluorescence microscope, and the labelled cells proliferated well, and the fluorescence attenuated slightly after passage. All the animals survived after operation, the incision healed well and no infection occurred. At 1 week after operation, all the animals in each group had different degrees of dysfunction. The angle θ of the left side in groups A, B, C, and D were (53.4±2.5), (54.0±2.6), (53.7±2.4), and (53.0±2.1)°, respectively; showing significant differences when compared with the healthy sides ( P<0.05). At 8 weeks after operation, the angle θ of the left side in groups A, B, C, and D were (61.9±4.7), (56.8±4.2), (54.6±3.8), and (63.8±5.8)°, respectively; showing significant differences when compared with the healthy sides and with the values at 1 week ( P<0.05). Gross observation showed that the integrity and continuity of regenerated nerve in 4 groups were good, and no neuroma and obvious enlargement was found. At 4 and 8 weeks after operation, the electrophysiological examination results showed that the nerve conduction velocity was significantly faster in groups A and D than in groups B and C ( P<0.05), and in group B than in group C ( P<0.05); no significant difference was found between groups A and D ( P>0.05). At 8 weeks after operation, the fluorescence microscopy observation showed a large number of CM-Dil-ADSCs passing through the distal and proximal transplants in group A, and relatively few cells passing in group B. Toluidine blue staining showed that the density of myelinated nerve fibers in groups A and D were significantly higher than those in groups B and C ( P<0.05), and in group B than in group C ( P<0.05); no significant difference was found between groups A and D ( P>0.05). Transmission electron microscope observation showed that the myelinated nerve sheath in group D was large in diameter and thickness in wall. The morphology of myelin sheath in group A was irregular and smaller than that in group D, and there was no significant difference between groups B and C. ADSCs can survive as a seed cell in vivo, and can be differentiated into Schwann-like cells under PRP induction. It can achieve better results when combined with acellular xenogeneic nerve to repair peripheral nerve injury in rabbits.