BACKGROUND Very few articles describe the compensations in gait caused by limb-length discrepancy (LLD). Song and colleagues explored kinematic and kinetic variables utilizing work equalization as a marker of successful compensation for LLD. They found no difference in strategies based on the location of pathology. The purpose of this study was to define the various gait patterns in patients with LLD and the impact of these compensations on gait kinetics. METHODS Forty-three children (mean age 12.9±3.7 y) with LLD >2 cm were evaluated in the motion lab using a VICON motion system with 2 AMTI force plates. Etiologies included Legg-Calve-Perthes, developmental hip dysplasia, growth plate damage due to infection or trauma, congenital shortening of the femur or tibia, and syndromes creating shortening of the limb. Evaluation included physical examination and 3-dimensional motion data generated using the model described by Vicon Clinical Manager (VCM). For data analysis, 3 representative trials were processed with the Plug-in Gait lower-body model using the "VCM spline" filter. Walking strategies were identified by visual review. A kinematic threshold of 2 SD away from normal values was used for inclusion in each group. Strategies included: (1) pelvic obliquity with the short side lower (<-1.5 degrees); (2) flexion of the knee of the longer leg in stance (>5.2 degrees); (3) plantar flexion of the ankle on the shorter leg through the gait cycle (<0 degrees); and (4) early plantarflexion crossover of the shorter limb (plantarflexion crossover occurred before 35% of the gait cycle). Variables were extracted into Excel using PECS (Vicon Motion Systems). The mean of the 3 trials was used for analysis. Scanograms were used to establish lengths of the femur and the lower leg including the foot. The percentage difference for the subject (%LLD) was calculated as the leg length between the 2 sides divided by the length of the long side. The total mechanical work over the stride was the sum of the positive work and the absolute value of the negative work in all planes. Paired t tests were used to analyze the work differences between the short limb versus the long limb. Unpaired t tests were used to compare between the different groups (short tibias, short femurs, and controls). RESULTS Distribution of single strategies for the group included: pelvis (11), equinis (5), vaulting (7), knee flexion (3); 17 subjects used multiple strategies. If the discrepancy was in the femur, patients chose a more distal compensation strategy, utilizing ankle movements, which resulted in more work at the ankle joint on the short limb compared with normal (P<0.0001). All subjects with tibia shortening showed pelvic obliquity (3 combined with knee flexion), which caused more work at the hip joint on the short limb compared with normal (P<0.01). Total mechanical work on the uninvolved limb was above normal for all groups (P<0.0001). CONCLUSIONS Our study contradicts previous literature that found no difference in strategy on the basis of location of the shortening and also a higher number of children with pelvic obliquity than previously described. It appears that different compensation schemes are used by patients with LLD. The increase in work may have long-term implications for management. Future studies will include changes in kinematics and work, after intervention. Better understanding of postoperative changes from different surgical methods may provide more insight for preoperative planning and may lead to a more satisfactory outcome for specific patients. METHODS Level II.