The treatment of femoral non-union, especially femoral exchanged nailing, has had mixed results in the recent literature. A review of the literature has suggested that exchanged nailing may be the gold standard for the treatment of femoral non-union. Is femoral exchange nailing an acceptable method of treatment of femoral fracture non-union? What is the evidence? In this systematic review we compare four different methods for the treatment of femoral non-union.

English speaking literature from 1970 to 2007 was searched using Pubmed^® and OVID™ databases and a manual reference search to reveal the original research, presenting the results of the treatment of femoral non-union with exchange nailing, plating, external fixation, and isolated bone grafting. Outcome parameters chosen were union rate and speed of union. The baseline variables chosen were age, number of surgeries, infection and the type of non-union.

In total, fifty six case series were identified containing 861 patients treated with exchange nailing (31 studies), 214 patients treated with plating (11 studies), 140 patients treated with external fixation (13 studies), 81 patients treated with bone grafting (4 studies). The average union rate was 89% for exchange nailing, 93% for plating and external fixation and 62% for bone grafting. Speed of union was 7.3 months for exchange nailing, 8.6 months for plating and 9.15 months for external fixation.

Study of baseline variables showed patients to be older in plating and external fixation groups (range 32–44 years, P< 0.001). There were more previous operations performed in the external fixation and plating groups (3 and 1.8 vs. 1.2, P< 0.001), significantly more infected non-unions in the plating and external fixation groups as compared to the exchange nailing group (40% vs. 11.5% P< 0.001) and significantly more atrophic non-unions in the plating and external fixation groups as compared to the exchange nailing group (85% vs. 65%, P< 0.001).

The literature suggests that femoral exchange nailing has an equivalent or poorer outcomes when compared to external fixation and plating in spite of having been performed in potentially less complex cases. Plating of non-union in the literature has a higher rate of associated bone grafting than femoral exchange nailing which may be responsible for marginally better union rate in the external fixation and plating groups.

Isolated bone grafting without revision of fixation does not provide adequate union rate and hence may be of questionable treatment value.

Proximal ulnar fractures may be difficult to treat and may result in chronic instability, non-union heterotrophic ossification, synostosis, stiffness and post-traumatic arthritis. The aim of this study is to study success of treatment in achieving stable reduction and early rehabilitation.

Between December 2003 and January 2007 fifty patients (fifty-one elbows) which had sustained 21.A3 and 21.C3 fractures were identified and retrospectively studied.

There were two broad groups of patients. Young males following high energy injuries (average age=38+/−16) and old females with osteoporotic fractures (average age=65+/−17). Twenty-two patients had associated monteggia and seven patients had trans-olacrenon dislocation. Twenty-three patients had radial sided injury. This included two capitelar fractures, nineteen radial head and neck fractures and one radial shaft fracture. Fortyfive patients were treated with plating and two patients were treated with tension band wiring. Five different plating techniques were used to stabilize the fractures.

Eighteen patients had incongruent reduction. Eighteen patients had complications of the treatment. There were seven cases of non-union, one case of loss of fixation, three cases of heterotrophic ossification, three cases of synostosis, one case of deep infection and five complications resulting from radial head fractures.

There was no relationship between loss of fixation and plating techniques. There was a direct relation between comminution and post fixation incongruence. Heterotrophic ossification was associated with comminution, radial head fracture, monteggia fracture-dislocation and non-union. Radio-ulnar synostosis was associated with comminution of the ulnar fracture.

In conclusion, the main predictor of poor outcome is the comminution of proximal ulnar fracture and the ability to achieve congruous fracture fixation.