The objective of this study was to evaluate a non-resorbable implant for the focal repair of chondral defects in eight adult horses with 12-month follow-up. The bi-layered construct composed of a polycarbonate-urethane-urea biomaterial which was printed in 3D fashion onto a bone anchor was implanted into surgically created osteochondral defects into the femoropatellar joints of eight horses. The analysis of post-mortem outcomes were compared to defects treated with microfracture in the same animal on the contralateral femoropatellar jointfemoropatellar joint. The overall macroscopic scoring after 12 months yielded higher scores in the OCI-treated stifles compared to MF treatment (p = 0.09) with better quality and filling of the defect. Histology revealed good anchorage of repair tissue growing into the 3D structure of the implant and histopathology scoring for adjacent native cartilage showed no difference between groups. MRI and micro-CT showed overall less sclerotic reactions in the surrounding bone in the implant group and no foreign body reaction was detected. Biomechanical analysis of the repair tissue revealed a significantly higher peak modulus (p < 0.05) in the implant group (0.74 ± 0.45) compared to the microfracture control group (0.15 ± 0.11). Dynamic loading yielded higher values for the repair tissue overgrowing the implant group (0.23 ± 0.17) compared to the microfracture control (0.06 ± 0.06) (p < 0.05). The bi-layered osteochondral implant provided a safe implant for focal repair of full-thickness osteochondral defects, as no adverse reaction was seen within the joints and the level of degeneration of adjacent cartilage to the repair site was not different compared to that seen in defects treated with microfracture after 12 months.