This paper reports our experience of revision open reduction, internal fixation and bone grafting of distal humeral fracture non-unions and in addition looks specifically at factors that may predispose to the development of non-union.

Between 1993 and 2003 18 patients with distal humeral fracture non-unions underwent revision surgery with bone grafting and rigid internal fixation. Two patients were lost to follow-up leaving a study group of 16 patients.

The patients’ age, sex, mechanism of injury, AO classification of the initial fracture and the primary treatment method were analysed with respect to possible factors predisposing to non-union.

All revision procedures were performed by the senior author. The non-union site was debrided, bone grafted and rigidly internally fixed.

Clinical assessment was performed using the Mayo Elbow Performance Score and radiographs were reviewed for evidence of bony union.

The Mayo elbow performance scores were excellent in 11, good in 2, fair in 2 and poor in 1.

Our results indicate that age, sex and mechanism of injury are not important in the development of non-union. Twelve patients (75%) however were considered to have undergone inadequate management of the original fracture.

Our experience would suggest that to reduce the risk of non-union following distal humeral fractures appropriate consideration must be given to the established and well proven surgical techniques.

If adequate fixation is considered beyond the experience of the treating surgeon we would strongly advise referral to a specialist unit.

Although it is generally accepted that revision total elbow replacement may be necessary for loosening, instability, peri-prosthetic fracture and infection there is less agreement as to whether surgery should be performed as a one or two stage procedure. This can be of vital importance since the soft tissues around the elbow are often relatively poor making a single operation desirable. However, a one stage procedure in the presence of undetected low grade infection will result in joint failure with early loosening.

In our unit we have found the use of a preliminary aspiration/drill biopsy prior to revision surgery helpful in evaluating whether a one or two stage procedure should be performed.

Over an 8 year period 18 revision total elbow replacements have been undertaken. 9 patients were revised for aseptic loosening, 4 for proven infection, 3 for instability of an unlinked implant and 2 for peri-prosthetic fracture.

With this experience we have devised the following management plan: Early instability of an unlinked implant is due to either poor implant positioning or soft tissue balancing and is suitable for a one stage revision without the need for aspiration/drill biopsy. Late instability is due to implant wear or low grade infection. In this situation we regard an aspiration/drill biopsy as necessary. A negative result allows a one stage revision whereas a positive aspiration indicates the need for a two stage revision.

In a peri-prosthetic fracture if the bone cement mantle is intact a one stage revision without aspiration/ drill biopsy can be performed. If however, there is bone cement lucency we would advise an aspiration/ drill biopsy.

We have found the aspiration/drill biopsy helpful prior to revision total elbow replacement and we have used it to guide us as to whether a one or two stage procedure should be performed.