Objective:To investigate the repair effect of electrospun gelatin polycaprolactone (GT/PCL) nanofiber aerogels (NFA) combined with cartilage extracellular matrix (ECM) for treatment of cartilage injuries in rabbits.Methods:Firstly, the GT/PCL electrospun membrane was prepared by electrospinning and was ground into the short fiber at high speed. ECM was extracted and separated from fresh bovine articular cartilage, which mixed with the short fiber solution (10 ∶1). Subsequently, it was used to prepared GT/PCL/ECM (NFA) three-dimensional scaffold. Finally, the physical characteristics of the three different scaffolds (GT/PCL, ECM and GT/PCL/ECM) were detected by scanning electron microscope and Fourier-transform infrared (FTIR) spectrometer, including the composition, microstructure, swelling rate, porosity, compressive strength and degradation rates. And the biocompatibility research was getting on by co-culturing the scaffold with chondrocytes. Fifteen male New Zealand white rabbits were divided into blank control group (Group A, n=5), ECM group (group B, n=5) and composite scaffold(GT/PCL/ECM)group (Group C, n=5) according to the random number table. An injury model was established and three types of bio-scaffold materials were implanted into different groups. At 3 weeks, the cartilage repair was evaluated among groups by semi quantitative global MRI scoring system (WORMS). After the animals were killed, the knee joints of each group were scored by the international society for cartilage repair histological score (ICRs); the ICRs histological score was performedby HE staining and safranine green staining. Results:Three scaffolds showed a porous three-dimensional structure under the scanning electron microscope. FIRT showed that GT and PCL were introduced into the scaffolds successfully. The GT/PCL NAF was loose and unable to be characterized by materials science. The swelling rate of GT/PCL/ECM scaffold [(1, 092.0±32.2)%] was higher than that of ECM scaffold [(933.6±16.3)%] ( P<0.01). The porosity of GT/PCL/ECM scaffold [(92.3±2.3)%] was higher than that of ECM scaffold [(85.9±2.2)%] ( P<0.05). The compressive strength of ECM scaffold [(2.7±0.1)kPa] and of GT/PCL/ECM scaffold [(2.4±0.1)kPa] showed no statistical difference ( P>0.05). The degradation rate of ECM scaffold was higher than that of GT/PCL/ECM scaffold, but the difference was not statistically significant ( P>0.05). The cytotoxicity rating of GT/PCL/ECM scaffold was grade I, indicating that its biocompatibility was better. At 3 weeks, the MRI WORMS score in Group C [(49.0±11.4)points] was significantly higher than that in Group B [(40.0±6.7)points] and that in Group A [(24.0±6.5) points] ( P<0.05 or 0.01); the general ICRS score of group C was [(7.4±1.1) points], which was significantly higher than that of group B [(4.6±1.1)points] and group A [(3.0±1.2)points] ( P<0.01); The ICRS histological scores of group C and group B were [(6.8±0.8)points] and [(4.2±0.8)points] respectively compared with group A [(2.8±0.8)points] were significantly higher ( P<0.05 or 0.01). Conclusion:GT/PCL/ECM (NFA) scaffold has similar tissue structure to natural cartilage and is superior to traditional ECM scaffold in physical properties and biocompatibility, which provides a stable environment for chondrocyte adhesion and growth, promotes collagen regeneration, and thus accelerates the repair of cartilage injury.