Purpose: The diversity of treatments proposed for septic nonunion of the femur demonstrates the lack of consensus. Treatment modalities validated for the leg appear to be transposable to the femur. The purpose of this work was to compare different treatments used in our centre and identify optimal management practices. Material and methods: We report a retrospective series of eleven patients (nine men and two women) who developed septic nonunion of the femur subsequent to trauma (n=9) or tumour (n=2). Sepsis developed early in seven cases and late in four. Mean time to treatment was 34.8 months. We based our strategy on a succession of steps starting with cure of the soft tissue and bone infection, before attempting reconstruction and consolidation.The first step involved fixation, antibiotic therapy and interposition of an acrylic spacer. The second step involved bone reconstruction, removal of the spacer, vascularised fibular graft associated with a cancellous bone graft (n=4) or massive cancellous graft inserted into the pseudomembrane created by the spacer (n=7). Results: Mean time to resolution of the infection was 10.9 months. Cure could not be achieved in three patients. Bone continuity was achieved in 8.8 months on average. The time to bone healing (i.e. duration of external fixation) was 22 months. Refracture occurred in four patients. Consolidation was not achieved in two patients. Discussion: During the second step, we preferred massive cancellous bone reconstruction due to easier technique, shorter healing time, and better adaptation of the reconstruction volume. Optimal time for the first step is about six months in order to avoid recurrent infection. Our healing times are similar to those reported by others: the healing index (time to healing divided by gap length) was close to that obtained with the compression-distraction technique.
Number one in frequency of all fractures in children is the distal forearm fracture. The most common green-stick fracture with minor or no dislocation is treated by short or long arm cast. Depending on the age 4 or six weeks of immobilization is sufficient. Displaced fractures of the distal radius and ulna are treated by closed reduction under general anaesthesia or lighter forms of analgesia. Reduction is followed by fixation in the “Schede position” (flexion, ulnar abduction) with obligatory change of cast after 10–14 days. Healing can be expected after 6 weeks. As an alternative percutaneous pinning of the reduced fracture allows immobilization in a short arm cast and without the the unpleasant flexion and ulnar abduction (. Voto et al 1990. , . Mani et al 1993. , . Gibbons et al 1994. , Choi et al 1959). There is currently a prospective randomised study running organized by Mr Clarke from Southampton, to the advantages and disadvantages after use of pins or abstaining from them apparent risks. For midshaft forearm fractures closed reduction and long arm cast immobilisation had been treatment of choice in the past. Remanipulation under anaesthesia because of lack of retention of both bone fractures have been common. Concerns came up mainly in the age group above 10 years with a high rate of unsatisfactory results (. Kay et al 1986. ). Plate fixation of both bones is a difficult procedure and causes damage to the interosseous membrane and can enhance rotatory deficits. In addition ugly scars are not unusual. Intramedullary nails seemed advantageous. (. Amit et al 1985. ). J.L. Morote and the Spanish school of Sevilla were the first to use a minimal invasive method of reduction and K-wire fixation of midshaft and proximal forearm fractures. (Perez-Sicilia et al 1977). The French group in Nancy and Metz had the some years later and developed their elastic stable intramedullary system for forearm fractures . Metaizeau 1988. , . Lascombes et al. 1990. ). A high rate of excellent outcomes and hardly any complications were observed. Intramedullary fixation with elastic stable nails even permits immediate motion (. Verstreken et al 1988. ). The surgical technique of Morote using blunt-ended 1,6 to 1,8 mm K-wires is described in “Operative Technique in Orthopaedics and Trauma” (. Parsch 1990. ) The results were confirmed by Kaye Wilkins (1996), . Luhmann et al 1998. , and . Richter et al 1998. An unacceptable high rate of complications was seen in groups, who used pins, which were not buried, who removed pins to early and before consolidation or who had fixed only one bone (. Cullen et al 1998. , . Shoemaker et al 1999. ). We recommend the intramedullary system for all displaced forearm fractures of children above 6 years until closure of the growth plate.(. Parsch 1990. ). The learning curve is short, the time of surgery an average of 40 minutes. The radiation exposure can be limited by the use of short impulse image intensifier. There is virtually no blood loss. With the learning curve more than 80 % can be fixed by closed means. Open reduction might be necessary in adolescents, or in delayed fracture care. Postoperative immobilisation is a plaster shell or brace is used for 2 weeks, this is not obligatory. Postoperative infections have not been observed after this minimal invasive method. Skin irritations can be avoided by complete bending of the K-wire ends. In unacceptable malunion after conservative treatment closed or open realignment of the fractures followed by intramedullary Morote pinning is the treatment of choice.
