Conversion of external fixation to internal fixation in a non-acute, reconstructive setting: a case series

The aim of the study is to determine the outcomes in patients who underwent conversion from an external fixator to an internal fixation device. This is a retrospective review of 18 patients (24 limbs) who underwent conversion from external to internal fixation. The patients had external fixators applied for traumatic bone defects or congenital deformities. Conversion to internal fixation was performed for reasons of patient dissatisfaction with external fixation, pin track sepsis, persistent non-union or refracture. The complexity of cases was graded using Paley's level of difficulty score. Patients were either converted acutely or delayed. Internal fixation devices were either intramedullary nails or plate and screws. Outcome was regarded as excellent if the patients were fully weight-bearing and pain-free on a mechanically well-aligned limb and without need for further surgery: good if the patient required subsequent surgery to achieve union and poor if irreversible complications occurred. Acute conversions (fixator removal and introduction of internal fixation device at same surgery) were done in 19 limbs and delayed conversion (interval between fixator removal and internal fixation) in 5. In the acute group, 17 limbs (89.4 %) had at least a good outcome, 16 of these limbs had an excellent result. Two limbs (10.6 %) had a poor result and required amputation. Both cases were after acute conversion to intramedullary nails; the original presenting diagnosis was of an infected non-union of the tibia and both had Paley scores above 7. In the delayed conversion group, all limbs (100 %) had at least a good outcome, with 4 limbs (80 %) having an excellent result. The mean external fixator time was 185 days (61-370). Both the cases with poor outcomes had longer external fixation times. This series supports the practice of conversion of external fixation to internal fixation with the majority of patients attaining good results. It identifies that plate devices appear to produce fewer deep sepsis complications, as compared to intramedullary nails, particularly when the original presenting diagnosis is a septic non-union.