Introduction

Closed intramedullary nailing of acute femoral diaphyseal fractures has become the standard of care and some studies have documented good results using this approach [2, 7]. In developed countries, patients typically present early for treatment and therefore neglected fractures seldom are seen. However, this is not always the case in underdeveloped countries, and neglected femoral fractures remain a problem in the developing world even in the new millennium. Some patients seek treatment from local bonesetters, and when treatment fails, these impoverished patients seek help after having lost their livelihoods [1, 10, 11].

Limb massage and indigenous splinting by local bonesetters affect the fractured bone and the entire limb [4, 8, 11, 12]. The limb may become edematous with the loss of thigh contour. The femur shortens and the knee becomes stiff, with obliteration of the normal parapatellar fossae. Prolonged neglect results in equinus of the ankle as a result of contractures. In addition to the fracture, knee and ankle stiffness have to be treated when patients present late with this constellation of findings.

In one study [4], two-stage intramedullary nailing and bone grafting of neglected femoral fractures showed a favorable outcome. In two-stage approaches, patients experience the morbidity of two anesthetics and two procedures, including blood loss and prolonged hospitalization. In theory, a single-stage surgical reduction and fixation should reduce some of this morbidity. Acute lengthening of a neglected, shortened femur can be difficult and has associated complications such as neurovascular compromise from acute lengthening and the possible need for bone grafting. We sought to determine the (1) likelihood of union, (2) complications and reoperations, and (3) functional results with single-stage open intramedullary nailing without bone grafting in patients with neglected femur fractures.

Patients and Methods

Between January 2003 and December 2007, 17 consecutive patients presented with neglected femoral shaft fractures. All were treated with single-stage nailing without bone grafting. The results of all patients were reviewed. The institutional review board approved the study and informed consent was obtained from each patient. The criteria for neglected femur fractures included fractures that were not treated by conventional methods or that were managed by nonmedical traditional bonesetters for a minimum of 1 month after injury (Fig. 1). Malunited fractures and nonunions after operative interventions were excluded from the study.

Fig. 1A–B
figure 1

The preoperative (A) AP and (B) lateral radiographs show a 10-week-old neglected proximal third femur shaft fracture. Overriding and varus angulation are seen on the AP radiograph and callus formation can be seen at the fracture site on the lateral radiograph.

Full size image

There were 15 men and two women with a median age of 27 years (range, 22–70 years). The mechanisms of injury were motor vehicle accidents in eight patients, fall from a height in eight, and fall of a heavy object onto the thigh at the workplace in one patient (Fig. 2). Nine patients had other system or skeletal injuries affecting the management and outcome of the femur fracture. The fractures were classified according to the AO/OTA classification [9]. Ten patients previously were treated elsewhere with massage and indigenous splints by traditional bonesetters, six patients did not seek any medical help and were lying at home, and one patient initially was treated for a head injury and presented to us late. The median time from injury to the hospital visit was 10 weeks (range, 4–44 weeks).

Fig. 2
figure 2

A preoperative AP radiograph shows a 20-week-old neglected midfemur shaft fracture sustained by Patient 8 in a fall from a height.

Full size image

Operative Technique

The operations were performed with the patients in the lateral position using a standard posterolateral approach to the femur. With the patient under anesthesia, before skin incision, the knee was manipulated gently with the thigh in its shortened position and the hip in neutral to avoid stretching of the sciatic and femoral nerves. The fracture site was exposed. The organized hematoma and the new bone formed around the fractured ends were removed for later use. The ends of the proximal and distal fragments were cleared of interposing fibrous tissue. Fibrous tissue and attachments to the linea aspera were released to mobilize the proximal and distal fragments. The medullary canals were opened with a drill and reamed with increasing diameter of reamers. The proximal and distal fragment ends were approximated with acute angulation and then levered to gradually straighten the femur, thus lengthening the limb progressively and allowing the soft tissues to adjust to the new tension. This position was maintained for some time to allow the soft tissues to stretch and a guidewire was inserted. After reduction, if the fracture pattern was rotationally and axially stable, a cloverleaf Kuntscher nail was used for stabilization, whereas if the reduction was unstable, a statically locked nail (simplified universal femur nail) was used. Proximal interlocking was done through the jig and distal interlocking was done using an image intensifier. In one patient, we used slot screws to achieve rotational stability with a Kuntscher nail (Fig. 3). The reamed material was placed in and around the fracture site before closure. The diameter of all the nails used was 11 mm.

Fig. 3A–B
figure 3

(A) A postoperative AP radiograph after Kuntscher nailing shows the slot screws that were added for rotational stability. (B) On the final AP radiograph obtained at 37 months, union and remodeling are seen.

Full size image

The mean duration of hospital stay was 7 days (range, 4–20 days). Five patients required a blood transfusion either intraoperatively or postoperatively to a maximum of 2 pints. All patients were followed until union for a minimum of 6 months (mean, 33 months; range, 6–72 months). Plain AP and lateral radiographs of the femur were taken to assess union. The patients received instructions in physiotherapy and their progress was monitored at each outpatient visit. Emphasis was placed on regaining active and passive knee movement and developing quadriceps strength. Weightbearing was dictated by the stability of the fracture and the associated injuries that the patients had sustained. With isolated femur fractures, patients were partial weightbearing for the first 2 months and full weightbearing by 3 months, although radiographic union was seen at approximately 4 months. Patients were provided with postoperative thromboembolic stockings as deep vein thrombosis prophylaxis and no chemical prophylaxis was followed. Two of the authors (PRJVCB, TJ) assessed all radiographs. The fracture was considered united if three of four cortices had bridging callus across the fracture site. At followup, another author (AZ) assessed fracture union, knee ROM using a goniometer, return to work, and clinical limb shortening was measured using a measuring tape.

Preoperatively, the mean ROM of the knee was 28° (range, 10°–150°). The mean preoperative femoral length discrepancy measured clinically from the anterosuperior iliac spine to the medial joint line of the knee was 3.1 cm (range, 1–5 cm) (Table 1). All patients were nonambulatory and were not working.

Table 1 Summary of patients
Full size table

Results

All fractures united at a mean of 16 weeks (range, 7–32 weeks) (Fig. 4), and all patients were able to return to their original occupation.

Fig. 4A–B
figure 4

Followup (A) AP and (B) lateral radiographs obtained at 48 months show union and restoration of alignment.

Full size image

No patient experienced neurologic complications related to the single-stage nailing procedure. The total femoral length regained was 4.5 cm. Sixteen patients (94%) regained their original femoral length. The one patient with shortening of 3.5 cm opted to manage with height correction footwear.

The mean postoperative knee flexion (Table 1) was 130° (range, 0°– 60°, 0°–150°). Fourteen (82%) patients regained full ROM. Of the three patients with knee stiffness, one regained full flexion after quadricepsplasty. The second patient gained only 60° and the other gained 90° with manipulation under anesthesia. They were offered quadricepsplasty but refused because they were satisfied with their improvement in movement. No patient had any extensor lag.

Two patients had preoperative common peroneal nerve palsy, and their foot drop recovered in 3 months. Another patient had serous discharge from the incision site which resolved at 3 weeks with regular wound dressings.

Three patients underwent reoperation, including one for repositioning an interlocking screw and another for dynamization. The third patient had vascular compromise during the immediate postoperative period and underwent a thrombectomy. This patient presented 24 weeks after a midshaft fracture sustained in a motor vehicle accident, with preoperative shortening of 4.5 cm. Surgery was uneventful and the patient’s pulses were palpable in the recovery room. However, his pulses where not palpable in the ward an hour later. Color Doppler confirmed a thrombus extending from the distal external iliac artery to the superficial femoral artery, which was identified and treated by thrombectomy. The vascularity was restored and his pulses were palpable. He has no residual ischemic changes and can walk 4 km.

Discussion

Closed intramedullary nailing has become the standard of care for acute diaphyseal femur fractures [2, 7]. However, the literature is sparse for neglected femur fractures. Neglected fractures, as a problem in developing and underdeveloped countries, do not appear to be addressed sufficiently in the literature. We sought to determine the likelihood of union, complications, reoperations, and functional results with single-stage open intramedullary nailing without bone grafting in patients with neglected femur fractures.

This study has several limitations. It could be improved by comparing our patients with a cohort of patients who could be treated with staged intramedullary nailing; however, because neglected femur fractures are not common enough to do a comparative study, it would be difficult to perform.

We found that open intramedullary nailing without iliac crest bone grafting for neglected femoral shaft fractures resulted in union in all patients thus treated, and 16 patients (94%) regained their original femoral length. In a previous study [8], single-stage open intramedullary nailing showed good results in patients with a similar presentation as described in our series, however, only five of 14 patients regained their original length. Mahaisavariya and Laupattarakasem shortened the bone to achieve reduction in nine patients [8]. Similarly, in another series of 11 patients with isolated femoral fractures, with a mean shortening of 3.8 cm [4], preoperative traction was used in selected cases and the nailing procedure was staged. Eight of 11 patients underwent iliac crest bone grafting along with open nailing [4].

We found that acute correction of femoral length can be performed safely up to 4 cm. Previous studies have shown that acute lengthening minimally affects knee ROM and the results have been gratifying [3, 13, 14]. Fourteen (82%) patients regained full ROM of the knee at latest followup. Neurovascular complications can be avoided by restricting lengthening to 4 cm to 4.5 cm in adults and 3 cm to 3.5 cm in children and maintaining knee flexion during nailing [6]. Akinyoola et al. [1] described the problems and outcome of neglected femur fractures in a large series of patients. The superficial and deep wound infection rate was 23.6%, and residual shortening was present in 10% of patients. We agree that these fractures can be difficult to treat with high complication rates compared with fresh fractures. In our series, all patients were treated with a single-stage open intramedullary nailing without bone grafting and traction. The mean preoperative shortening was 3.1 cm (range, 1–5 cm). The patients who regained their femur length had no neurologic complications.

All patients returned to work during the followup period and regained functional ROM. In the series of Mahaisavariya and Laupattarakasem, 42% of patients regained full ROM, and in the other 58%, the range regained was adequate for activities of daily living [8]. Iliac crest bone grafting is associated with potential complications and morbidity [5, 16]. Bone grafting has a role when performed for one-stage lengthening [3, 13, 15] and in cases of initial open injury with bone loss, however in our series, we think the local callus and reaming were sufficient to enhance fracture healing. A Kuntscher nail was used in this particular demographic, because it is inexpensive. When the fracture showed a stable geometry, the Kuntscher nail proved to be adequate. Interlocking nails were used only when the fracture was found to be rotationally unstable.

The strengths of this study are that it is among the larger studies of neglected femur fractures; other studies have ranged from 11 to 52 patients [1, 4, 8]. All patients in this series were followed up until union and subsequently after union; none was lost to followup before union or before 6 months. Neglected femur fractures are uncommon, but they are important because of the morbidity they can cause, and because they disproportionately affect an already-disadvantaged, generally impoverished, patient population. Single-stage nailing without bone grafting showed timely union, low complication rate, and good restoration of function in our series. Although comparative controlled studies are required to prove its potential advantages, it is reasonable to hypothesize that single-stage open intramedullary nailing can be performed safely in most patients without shortening the bone and avoiding the addition of autogenous or allograft bone.