BACKGROUND: Internal tibial torsion (ITT) is a common boney deformity in children with cerebral palsy (CP). The current recommended treatment is tibial derotation osteotomy (TDO) to improve gait biomechanics. Satisfactory short-term results after TDO have been reported but long-term results have not been studied. The purpose of this study was to evaluate the long-term outcome after external TDO performed to correct ITT in ambulatory children with CP. METHODS: Following IRB approval, gait kinematics and passive range of motion measurements were retrospectively evaluated in children with spastic CP who underwent TDO due to ITT comparing preoperative (E0), short-term postoperative (E1; 1 to 3 y post), and long-term postoperative (E2; >5 y post) results. Limbs were categorized as corrected, undercorrected, or overcorrected at both E1 and E2, by comparing the subjects mean tibial rotation (MTR) in gait to a group of typically developing children. Age at surgery, preop MTR (at E0), preop gait velocity (at E0), gross motor function classification system score, and foot deformity were evaluated to determine their influence on long-term results. RESULTS: The study sample consisted of 36 legs (with E0 and E2) and 17 legs (with E0, E1, and E2). The mean age at surgery was 7.4±2.8 (range, 4 to 16.6) years. Comparing the changes over time, kinematic MTR improved from 17±11 degrees initially (E0) to -10±14 degrees short term (E1) and progressed to -23±13 degrees long term (E2) (P<0.05 E0/E1/E2; internal rotation is positive). At E2, 16 legs (44%) were found to be in the kinematic corrected group and 20 legs (56%) in the kinematic overcorrected group. There were no significant differences between the corrected and overcorrected groups of children in respect to age of surgery, gross motor function classification system, E0 MTR, gait velocity, or foot deformity. CONCLUSIONS: Although external TDO is an accepted form of treatment in children with CP, in the long term a tendency to move into external tibial torsion is common. Therefore, caution is warranted with children who initially present with ITT to avoid overcorrection. LEVEL OF EVIDENCE: Level IV-therapeutic study.