This study compares the predictive value of age at surgery in high tibial osteotomy. Twenty-seven high-tibial osteotomies in patients who are 65 years or older (mean age at surgery 68+/-4 years, follow up 12+/-2 years) were compared to 67 osteotomies in patients younger than 65 years (mean age at surgery 56+/-6 years, follow up 13+/-3) with respect to the outcome by Cox regression analysis. Failure, i.e. endpoint, was defined as implantation of a knee endoprosthesis and assessed by Kaplan-Meier analysis. There is a significantly higher risk for failure of high tibial osteotomies in patients of 65 years or more compared to younger patients (failure rate 38.4+/-11.3% vs. 23.1+/-5.8%) resulting in a relative risk of 1.5 (P=0.0461). The hazard of failure increased 7.6% per year of age. We conclude that in regard to the increasing risk of failure per year of age and the higher failure rate in older patients, high-tibial osteotomy should not be performed on patients older than 65 years.