Introduction and Aims: A retrospective review of the medical records and radiographs of patients treated with operative fixation of two-part proximal humerus fractures was undertaken to compare the results of different modes of fixation. Comparison was made between ORIF with a blade plate, percutaneous Kirschner wire fixation, and intramedullary nail fixation utilising a Polarus nail.

Method: Thirty-six patients were treated with intramedullary fixation, 11 with blade plate fixation and 10 with percutaneous fixation, utilising Kirschner wires. Union rates were 34/36 (94%) for the Polarus nail, 9/10 (90%) for K-wire fixation and 9/11 (82%) for blade plate fixation. Time to union averaged 12.4 weeks for the Polarus nail, 11 weeks for K-wire fixation and 21 weeks for blade plate fixation. Average active shoulder range of motion in forward flexion/abduction were 125/118 degrees for intramedullary nail, 132/132 degrees for the blade plate and 112/111 degrees for patients treated with K-wire fixation.

Results: The major discrepancy in comparison of the different modes of fixation was in the complication rate. Ten complications occurred in the group treated with intramedullary fixation. These included one non-union, one painless fibrous non-union and back out of the proximal interlocking screw in eight patients, five of which required screw removal. Seven of 11 patients treated with blade plate fixation experienced complications, including two non-unions, two malunions, two with functionally limiting heterotopic ossification, one arthrofibrosis and one with persistent pain. The complications associated with percutaneous Kirschner wire fixation included one non-union, two malunions, four developed functionally limiting heterotopic ossification, five incidences of early pin removal secondary to pin migration, one arthrofibrosis requiring surgical intervention and one infection requiring surgical irrigation and debridement.

Conclusion: Results comparable in all groups. Fewer complications seen with intramedullary fixation. Majority of complications with Polarus nail related to backing out of proximal interlocking screw. Modification of implant to include end cap, which locks the proximal screw seems to eliminate complication. Results indicate that intramedullary nail fixation is superior to blade plate fixation or pecutaneous Kirschner wire fixation for two-part proximal humerus fractures.

Introduction and Aims: A non-union of the scaphoid has traditionally been managed by open reduction and internal fixation with bone grafting. The purpose of this study was to evaluate the healing rate of selected scaphoid non-unions managed by arthroscopic assisted fixation alone without supplemental bone grafting.

Method: Fifteen patients underwent arthroscopic fixation of a scaphoid nonunion without bone grafting. There were 14 males and one female. Average age was 20 years (range 17–28 years). A history of a scaphoid fracture was present an average of eight months prior to fixation (range 4–15 months). The lunate was neutral in all pre-operative radiographs without carpal collapse in order to be included in the study. There were 12 horizontal oblique fractures involving the middle third, one transverse fracture and two proximal third fractures. A guide wire was placed under fluoroscopic guidance and anatomic reduction of the fracture was assessed arthroscopically from the mid-carpal space. A headless cannulated compression screw was placed dorsal to volar. No patient underwent bone grafting. Fracture union was evaluated by CT scan and plain radiographs.

Results: All patients healed their fractures. Average time to union was 12 weeks (range 8–18 weeks). Average wrist extension was 50 degrees, flexion was 60 degrees, radial deviation 20 degrees and ulnar deviation 25 degrees. Utilising the Mayo Modified Wrist Score, there were 12 excellent and three good results. Arthroscopic assisted fixation of selected scaphoid non-unions without bone grafting yielded a 100 percent union rate with minimal morbidity in this series.

Conclusion: Placement of a cannulated screw under arthroscopic guidance avoided soft tissue stripping, preserved the blood supply to the fracture fragments, and yielded excellent range of motion in this series. This technique would not be recommended for patients with a humpback deformity or signs of carpal collapse where bone graft would be required to correct the flexion deformity of the scaphoid.