Introduction

Primary wound closure in open tibial fractures has not been recommended. Traditionally initial debridement with fracture stabilisation and delayed wound closure was the accepted treatment. However this practice was developed before the use of prophylactic intravenous antibiotics and improved techniques for fracture stabilisation. Studies suggest that infections are not caused by the initial contamination but the organisms acquired in the hospital. Subsequent primary wound closure after adequate wound care and fracture stabilisation should be a safe concept and should not increase the rate of complications.

This prospective study included 236 open tibial fractures treated by unreamed AO nailing at three to four days after injury. Ten patients were lost to follow-up and 24 were excluded. We reviewed 202 consecutive open tibial fractures treated between January 1994 and December 2000. The mean age was 31 years (18 to 66) and 152 patients were men. There were 74 Gustilo grade-I, 66 grade-II, 32 grade-IIIA and 30 grade-IIIB fractures. The soft tissue injuries were managed by delayed primary closure, split skin graft or local flap. Most fractures were stabilised by unreamed nailing and statically locked.

Below-knee casts were applied routinely for six weeks. Full weight-bearing was permitted and maintained 6 to 12 weeks after the operation. In 16 patients (8%), union was delayed, but union occurred within 12 to 30 weeks in 194 patients (96%). Nonunion was observed in eight fractures (4%). Functional union occurred in all patients. Ten (5%) had mild knee pain. Acceptable shortening was noted in seven cases (3.5%) and varus/valgus angulations in 12 (6%). In two patients the nail was too long and required exchange. Eight locking screws bent or broke. There was no sepsis but three wounds were superficially infected. Compartment syndrome developed in one patient. Twelve additional operations (exchange nailing, bone grafting, fibular osteotomy and dynamisation) were performed for delayed unions and nonunions. Seven additional operations were undertaken for surgical errors where the nail was too long, there was early excessive rotation, or there were missed locking screws.

The delay in final treatment of open tibial fractures allows safe selection of patients for secure and cost-effective unreamed intramedullary nailing, with minor complications.

In a retrospective study, we reviewed 45 peri-trochanteric fractures treated between April 1995 and November 2002. The mean age of the 24 men and 21 women was 71 years (57 to 91). There were 34 inter-subtrochanteric, four reverse obliquity intertrochanteric fractures and nine subtrochanteric fractures. On the AO classification, there were 11 type-31A2-2, 21 type-31A2-3 and four type-31A3-1 intertrochanteric fractures. The fracture extended into the femoral neck in one case and into the diaphysis in three. Cardiopulmonary diseases were present in more than 60% of patients.

In most cases, fractures were reduced by closed reduction or reduction through a short incision. In 42 cases, a Smith and Nephew femoral reconstruction nail was used. Three fractures were stabilised with AO undreamed femoral nail with spiral blade. Distal locking screws were inserted in all cases. Progressive passive hip and knee movement was introduced from day one postoperatively. Partial weight-bearing was permitted from the onset. All but two fractures healed within 3 to 6 months. Union was delayed in two subtrochanteric fractures. Functional hip and knee movement was present in all patients. There was no sepsis. Proximal screws backed out in four cases, but this was of no functional significance. In one case the superior proximal screw was too long and required removal. In one case screws backed out of the femoral neck and further surgery was required. No breakage of screws or nail was observed. Inter-subtrochanteric and subtrochanteric fractures in the elderly may be successfully managed with intramedullary femoral reconstruction nails.

In a prospective, consecutive study we reviewed the results of 32 supracondylar femoral fractures treated by Smith & Nephew intramedullary supracondylar nailing between January 1996 and October 2002. The mean age of 23 women and nine men was 67 (58 to 89). All fractures were closed. Two patients had associated upper limb fractures. In four cases, fractures occurred around total knee prostheses. Four patients had previously undergone ipsilateral total hip replacement or had had a sliding hip screw. On the AO classification the majority of fractures were type-33A1 and A2; seven fractures were classified as C1 and C2. The patients were placed in the supine position on a radiolucent operating table with the knee in 30° of flexion.

Postoperatively a hinged knee brace was applied and worn until union. Supported, progressive knee movement was introduced from day one. Partial weight-bearing was permitted as soon as pain subsided and continued until there were radiological signs of union. Within three to six months all but two fractures united. These united after prolonged bracing. There was no sepsis or fixation failure. A functional range of knee movement was observed in all patients.

Retrograde intramedullary nailing is a safe and successful method of management of supracondylar femoral fractures in the elderly and offers a minimally invasive alternative.

In a prospective study, we reviewed 72 distal tibial fractures treated by percutaneous plating between July 1996 and June 2001. The patient’s mean age was 36 years (19 to 76). The majority of them were men. Seven fractures were open, with three of them Gustilo grade IIIA. Of 65 closed fractures, 15 were Tscherne grade III. All fractures were type 43A according to AO classification and were less than 5 cm from the ankle joint. Most of the fractures were group A3, with 22 group A3.3. Percutaneous plating was delayed for a mean of five days (2 to 15). Pre-contoured small fragment dynamic compression plate was placed on the medial aspect of the tibia under image intensifier control, through a short distal skin incision. On average, three distal and two proximal screws were inserted. Fibular fractures were stabilised in a similar fashion. Satisfactory fracture reduction was achieved in all cases. Postoperatively a below-knee cast was applied for six weeks and weight-bearing was permitted at eight weeks. Fracture healing occurred within 12 weeks (10 to 16). One patient needed bone grafting for treatment of delayed union. All patients had a functional range of ankle movement. In one patient, breakage of all screws was observed in a united fracture with shortening of the fracture. Local late infection where the skin was tented by skin screw heads occurred in eight patients and was resolved by debridement and hardware removal.

Percutaneous plating of type A43 tibial fractures is safe, reproducible and successful and has few complications. The few adverse affects may well be eliminated by the use of newly-introduced low profile plates and screws.

In a prospective study, we reviewed 23 proximal humeral fractures treated by AO/Synthes intramedullary nailing between January 1999 and December 2000.

According to Neer’s classification, there were 12 two-part fractures, eight three-part and three four-part fractures. There was anterior dislocation of the glenohumeral joint in four patients. The mean age of the 16 men and seven women was 49 years (26 to 71). More fractures occurred in patients over 55 years of age. Anteroposterior and trans-scapular radiographs were taken and CT routinely performed. Surgery was performed within 5 to 14 days of injury. In young patients with two-part fractures, we used percutaneous integrate nailing. Three and four-part fractures were reduced and fixed through a short anterolateral deltoid split approach. The nail was inserted without reaming. The fracture fragments were reduced around the exposed proximal part of the nail and reduction secured by insertion of locking screws and a tension wire band. Ruptures of the rotator cuff were repaired. The nail was locked distally in 16 fractures.

The arm was immobilised for two or three weeks but supervised shoulder movement started as early as four to five days postoperatively. All fractures healed within 12 weeks. Functional shoulder movement returned in all but two cases. In younger patients recovery was faster and a near-full range of abduction and flexion returned. No sepsis occurred. Postoperative backing-out of the nail and varus deformity of the humeral head occurred in two patients. Two patients required re-operation. Backing-out of proximal locking screws was observed but did not affect functional outcome.

This minimally-invasive method of fixation by intramedullary nail, locking screws and tension wire band through a short incision may be an alternative way of managing complex proximal humeral fractures.

In a prospective study, we reviewed 52 metaphyseal fractures of the proximal tibia treated by percutaneous plating between January 1996 and October 2000. Owing to the proximity of the fractures to the joint, intramedullary nailing was not suitable.

The mean age of the patients, most of whom were men, was 41 years (16 to 82). Five fractures were open. There were 10 comminuted fractures extending into the diaphysis and five segmental fractures. The fractures were reduced and under the image intensifier percutaneously plated through a short approach proximal to the fracture. Fracture reduction was achieved either by manipulation and traction or by use of femoral distractor and reduction clamp. Synthes tibial head buttress plates and screws were used for stabilisation. On average, three proximal and distal screws were percutaneously inserted. Satisfactory fracture reduction was achieved in the anteroposterior plane in all fractures, but in the sagittal plane tilting of the proximal fragment was observed in five cases. There were no intra-operative neuro-vascular complications. Postoperatively the leg was immobilised in a brace for 6 to 12 weeks. At a mean of six to eight weeks, when radiological signs of healing were noted, weight-bearing was permitted. The mean time to union was 12 weeks (8 to 18). There were two cases of delayed union. No patient had functional restrictions, secondary displacement or failure of fixation. In four patients the proximal screws backed out, but this did not affect functional outcome. Late sepsis, which developed at the site of the distal screws in six patients, subsided after drainage of abscesses in two patients and removal of plate and screws in four.

Percutaneous plating may be used to manage proximal tibial fractures unsuited to intramedullary nailing.