Clinical and radiographic data were collected in 139 patients with 195 posterior cruciate retaining total condylar knee prostheses to evaluate the relationship of the proximal tibial resection level with long term results. Among the 139 patients were 75 patients with 106 total knee replacements observed for more than 8 years. All patients underwent biyearly routine examinations, including radiographs and clinical evaluations. The average medial tibial resection for the 139 patients with 195 total knee replacements was 2.95 mm, and in the subset of 75 patients (106 knees) observed for more than 8 years, it was 3.3 mm; both groups had a maximum of 14 mm. Sixty-three percent or 67 knees had medial resection levels of 3 mm or less. The average lateral tibial resection for the 195 knees was 5.48 mm and in the 106 knees was 5.71 mm, with a maximum of 22 mm. Fifty-one percent of 104 knees had lateral resection levels of 5 mm or less. Statistical analysis showed that there was no significant correlation between the level of proximal tibial resection and Knee Society knee score, range of motion, radiolucencies, or loosening or revision. These long term results suggest that minimal proximal tibial resection is not necessary for a successful arthroplasty, and problems associated with minimal resection, such as joint line elevation and thin polyethylene inserts, can be avoided.