Introduction: Bone defects are challenging to treat surgically. The primary objective of our study was to compare the union rate and time to union between the tibia and femur when using the induced membrane technique. The secondary objective was to document how failures were managed. Material and Methods: This retrospective, single-center study involved 33 patients (23 men, 10 women) who were older than 18 years of age. They were treated surgically for a leg fracture or long bone nonunion (22 tibia, 11 femur) using the induced membrane technique between January 2011 and December 2016 and had a complete follow-up. The minimum follow-up was 1 year for fractures and 2 years for non-union cases. Bone union was defined as the presence of at least two cortices with bridging on two radiographic views and return to full weight bearing. Results: The mean patient age was 38.3 ± 15.5 years (18-72). The mean bone defect size was 7.9 ± 5.0 cm (2.3-18.0). The mean follow-up was 3.3 ± 1.8 years (1-7.2). The union rate was 61% (20 patients). The mean time to union was 10 ± 6.4 months (3-23). The time to union was significantly longer in the tibia (11.6 ± 6.9 months [3-23]) than in the femur (6.3 ± 2.9 months [3.4-10.3]) (p = 0.025). The failure rate did not differ between the tibia and femur. Nine of the 13 patients (69%) in which the treatment failed were reoperated; 7 of them underwent nonunion treatment (78%) and 2 underwent amputation (22%). The other 4 patients were waiting for an infection to resolve before being reoperated. Conclusion: The induced membrane technique is an effective surgical procedure for large bone defects in both the tibia and femur. However, the time to union was shorter in the femur than the tibia in our cohort.