The purpose of this study is to evaluate the results of bone graft or cement fill for tibial bone defect in total knee arthroplasty. 51 knee in 45 patients which had bone graft or cement fill for tibial bone defect of more than 1cm3 in total knee arthroplasty, between March, 1990 and July, 1994 at the department of Orthopedic Surgery in Hanyang University Hospital were followed for average 30 months. Bone grafts were performed in 33 knees and cement fills in 18 knees. The size of tibial defect was measured after removal of sclerotic and dense bone in the tibial plateau. The average size of tibial defect treated by bone graft was 4.9 cm3 in volume, 11 mm in height and the average size of tibial defects treated by cement fills was 1.5cm3 in volume, 6.8 mm in height. The results were as follows ; 1. In true A-P and lateral x-ray checked by fluoroscopic guide, trabecular union was seen in all 33 knees of bone grafts without collapse, loosening and bone resorption. In one of 18 knees of cement fill, radiolucent line was detected 2 mm in width between cement mantle and the adjacent tibial bone. In other 17 knees, radiolucent was seen 1 mm or less in width. 2. According to the system of Knee Society Score, average knee rating score was improved from 23.2 points preoperatively to 90 points postoperatively in bone graft cases, from 23.4 points preoperatively to 93.2 points postoperatively in cement fill cases. Average functional score was improved from 14.8 points preoperatively to 81.5 points postoperatively in bone graft cases, from 13.3 points preoperatively to 81.4 points postoperatively in cement fill cases. 3. Finally, we obtained the good results by cement fill during cement fixation of tibial component or bone graft without cement fixation of tibial component for small tibial bone defect and by bone graft for large tibial bone defect.