Background: Bone morphogenetic proteins (BMPs) induce new bone in patients with bone defects and at extraskeletal sites in animals. Standard treatment for symptomatic scaphoid non-unions is bone graft with or without internal fixation by a screw or wires. We tested the ability of human recombinant osteogenic protein-1 (OP-1, BMP-7) with compressed autologous or allogeneic bone graft to accelerate the healing of scaphoid non-union.

Study Design: Randomized and controlled pilot study in 17 patients with a scaphoid nonunion.

Methods: Patients were randomly assigned to one of three groups: (1) Autologous iliac graft (n=6), (2) Autologous iliac graft + OP-1 (n=6) and (3) Allogeneic iliac graft + OP-1 (n=5). Radiographic, scintigraphic and clinical outcomes were assessed throughout the follow-up period of 24 months.

Results: OP-1 improved the performance of both autologous and allogeneic bone implants. Three dimensional helical CT scans and scintigraphy showed that the pre-existing sclerotic bone within proximal scaphoid poles was mainly replaced in OP-1 treated patients with well vascularized new bone. Addition of OP-1 to allogeneic bone implant equalized the clinical outcome with the autologous graft procedure and enabled circumventing the second donor graft harvest procedure resulting in less blood loss, shorter anesthesia and no pain at the donor side.

Conclusion: This is the first evidence that a recombinant human BMP accelerates scaphoid bone non-union repair and resorption of sclerotic bone in this specific microenvironment.

Clinical Relevance: OP-1 might be successfully used in healing of scaphoid non-union.

A computer-assisted method of preoperative planning was used to create virtual models of the deformed distal end of the radius after malunion of a fracture. By comparison with a similar model of the uninjured wrist, values were calculated for the angles and lengths to be corrected by osteotomy. Shifts of the distal fragment were analysed for 33 deformed wrists, 27 of which underwent corrective osteotomy and bone grafting. In more than half the cases there was dorsal or volar shift of 3 mm or more. The accuracy of the correction was measured by comparing the three-dimensional models before and after osteotomy with the model of the normal wrist. The volar and ulnar inclination angles of the articular surface of the radius and the radial length were regularly restored to normal.