This study compared the healing of articular cartilage and the clinical outcome after osteotomy with or without marrow stimulation microfracture or abrasion arthroplasty for osteoarthritis of the knee. Patients with osteoarthritis of the medial compartment of the knee were divided into a group undergoing high tibial osteotomy alone (HTO group: 37 knees), a group undergoing osteotomy plus microfracture (MF group: 26 knees), and a group undergoing osteotomy plus abrasion arthroplasty (AA group: 51 knees). The extent of cartilage repair was compared at 1 year after surgery by arthroscopy with reference to Outerbridge's classification, while the clinical outcome was compared at 1, 3, and 5 years postoperatively. Second-look arthroscopy revealed better repair of the femoral condylar cartilage in the AA group than the HTO group (p<0.0005) or MF group (p<0.01), with no difference between the HTO and MF groups. Repair of the tibial condylar cartilage was also better in the AA group than the HTO group (p<0.005), but there was no difference between the AA and MF groups or the MF and HTO groups. There were no differences of the clinical outcome between the three groups. In conclusion, repair of articular cartilage at 1 year postoperatively was accelerated by abrasion arthroplasty, but not by microfracture. However, there was no difference of the clinical outcome within 5 years after surgery, so the clinical utility of marrow stimulation techniques was not apparent in this study.