Purpose: Reconstruction using a revascularised fibula has advantages in terms of remodelling the transplant to the stress forces and of better resistance against infections. Osteosynthesis offers mechanical advantages (stabilisaton facilitating primary and stress fracture healing) but with the risk of preventing hypertrophy of the fibula which is not exposed to the stress forces.

Material and methods: Our series of 25 vascularised fibular transfers (maximum follow-up 15 years) concerned post-trauma tissue loss (11 cases) and tumour resections (14 cases). For this study of remodelling, we retained only oncological reconstructions because in the event of trauma sequelae, many factors can interfer (infection, preservation of the homolateral fibula with tibial tissue loss). None of the patients were lost to follow-up and remodelling was assessed at two years or more.

Results: There was one failure requiring leg amputation (infected nonunion of the distal tibia on a radiated osteo-sarcoma). All the other fibulae healed. Three metaphyseal resections of the distal femur were assembled with a lateral plate using the fibula as a medial strut under compression. This type of assembly favours remodelling and excellent results were obtained in three cases. Five arthrodeses of the knee were performed using a fibula with a femorotibial nail. Healing was slow and the fibula thickened little, particularly when it was simply apposed on the tutor (three cases) rather than encased under compression (two cases). For five proximal humeri, use of a thin plate in three cases (forearm plate) was sufficient to allow healing without inhibiting remodelling.

Discussion: For the lower limb, good remodelling is obtained with an assembly allowing compression of the fibula placed medially to the shaft alignment. For metaphyseal loss, we advise a lateral plate with a fibular strut medially. For arthrodesis, a nail is probably more prudent. The position of the lateral tutor decreases the stress on the fibula and is recommended less than the medial strut position. For the humerus, synthesis is required but may be minimal to allow optimal remodelling.

Conclusion: Vascularised fibula transfer is a reliable technique which can be recommended for major resections (mean 160 mm) in active subjects. Axial compression forces applied to the graft and use of light osteosynthesis appear to favour healing and remodelling.