The purpose of this prospective randomized clinical study was to compare the outcomes of mosaic type autologous osteochondral transplantation (OAT) and microfracture (MF) procedures for the treatment of the articular cartilage defects of the knee joint in athletes. Between 1998 and 2002, a total of 57 athletes with a mean age of 24.3 years and with a symptomatic lesion of the articular cartilage in the knee were randomized to undergo either OAT or MF procedure. There were 28 athletes in OAT group and 29 in MF group. Patients were evaluated using a modified Hospital for Special Surgery (HSS) and International Cartilage Repair Society (ICRS) scores, MRI and clinical assessment after 6, 12, 24 and 36 months after the surgery. According to the modified HSS and ICRS scores, functional and objective assessment showed that 96% had excellent or good results after OAT compared with 52% after MF procedure (P<0.001). In 12, 24 and 36 months after the operations, the HSS and ICRS showed statistically significantly better results in the OAT group (P=0.03; P=0.006; P=0.006). Twenty-six (93%) athletes following OAT and fifteen (52%) athletes following MF returned to sports activities at the preinjury level at an average of 6.5 months (range, 4-8 months) after the operations. At an average of 37.1 months follow-up, our prospective, randomized, clinical study in athletes has shown significant superiority of the OAT over MF for the repair of articular cartilage defects in the knee.