Neglected hip dislocations often occur in children and are rarely reported in adults. This study showed effective correction of limb discrepancy and significant function improvement with external fixation-assisted reduction in consecutive thirteen adult patients. Though skeletal traction is important in joint reduction, the acetabulum becomes filled with fibrous tissue in neglected dislocations would make reduction difficult by simple traction. Skin traction usually carries a weight under 5 kg. Halo-femoral and tibial tubercle traction can carry more weight but it often ends with unsatisfactory results in patients with neglected hip dislocation [9, 16]. Prolonged dislocation may cause severe joint contracture so it is difficult to achieve satisfied reduction by regular traction. Furthermore, heavy traction may lead to a high risk of neurovascular complications.
Distraction osteogenesis by external fixators has been widely used in patients with bone defects [17]. The good extension and regeneration ability of the skin and bones offers a theoretical basis for the application of external fixators in traction reduction. Screws were placed in the anterior inferior iliac spine and ipsilateral femur mid-diaphysis to assemble the external fixator. The daily lengthening was about 1–3 mm and could be suitably increased according to the patients’ tolerance. Plain radiographs were routinely taken to examine the correction of the shortening. Though shortening could be effectively corrected through traction by external fixators, it was hard to achieve accurate reduction and the associated acetabular fracture, which occurred in 10 patients who also needed surgical repair. Open reduction was performed when the femoral head was reduced underneath the articular surface. At this time, with muscle stiffness alleviated, the surgery was accomplished with minimal invasiveness and duration.
Intermittent traction by an external fixator was accomplished with a moderate force, and was more tolerable in patients compared with heavy weight traction. The external fixators made patients cast off long-stays in bed, and no complications related to prolonged immobilization such as pressure sores or venous thrombogenesis were observed. Neurovascular condition was closely monitored during the traction. Stretching would be suspended if neurological signs such as paralysis and pain occurred and continue after symptoms relief. No neurovascular injuries related to stretching occurred in our patients.
Common complications of traumatic hip dislocation include infection, sciatic nerve palsy, AVN, heterotopic ossification, and posttraumatic arthritis [18]. Previous studies have reported that the incidence rate of arthritis was 16.1–30.0%, and 8.1–10% for AVN in the mid-to-long-term follow-up [1, 19]. The blood supply to femoral head is often damaged by traumatic dislocation, especially the medial femoral circumflex artery. Vascular compromise leads to intravascular coagulation and ischemic necrosis, resulting in chondral failure and accelerated joint degeneration [20]. A high incidence of osteonecrosis has been reported from 10–25% in hip dislocations [21]. Furthermore, the severity of the injury and time to reduction are associated with increased risks of osteonecrosis.[20] The rate of osteonecrosis is only 10% in adults after simple dislocation, but reaches 70% if the dislocation is accompanied by severe proximate bony destruction such as fracture of the femoral head and acetabulum [22]. Hougaard and Thomsen reported that the rate of osteonecrosis of the femoral head can be reduced from 58% to 4.8% if hip reduction is performed within the first 6 h following injury [23]. Others have concluded a trend for decreased osteonecrosis of the femoral head when hip reduction was performed within 12 h [1]. In our patients, AVN occurred in 2 patients (15.4%), and osteoarthritis occurred in 1 patient (7.7%). All 3 patients suffered acetabular fractures with an average duration from injury to surgery of 6.3 months (range, 4 to 8 months). We agree with previous studies and think the long duration and associated injury could be critical factors. Patient 6 (Table 1) showed AVN 10 months after the surgery, and the Merle d'Aubigne score was improved from 4 to 12. Patient 7 (Table 1) showed AVN 12 months after surgery, and the Merle d'Aubigne score was improved from 5 to 13. Patient 11 (Table 1) showed osteoarthritis at the last follow-up (14 months), and the score was improved from 4 to 11. Though complications occurred, patients acquired significant function improvement. Scholars have also reported that complications usually developed within 5 years of dislocation [7, 24], though our study only reports a short-term follow-up result within 18 months. As the incidence of secondary complications may increase with time[25], an accurate complication rate should be recalculated for long-term follow-up.
THR has been recommended for hip dislocations with a duration of more than 3 months [10], but usually providing limited correction of limb discrepancy within 6 cm [11, 12] and muscle release was needed for better correction [26]. It may provide limited efficacy for patients in this case series which all suffered limb inequality of more than 5 cm. Considering all patients were under 50 years and no signs of necrosis or arthritis occurred when on admission, we decided to perform open reduction and fixation after traction. Follow-up results showed significant pain relief and function improvement. The satisfactory rate of clinical outcome was 53.8% (excellent and good results). Limb discrepancy was effectively corrected, and no patient was left with >2 cm of inequality, thereby laying the foundation for possible THR in the future. AVN and arthritis occurred in three patients. Though they showed unsatisfactory outcomes (two poor and one fair), none of them received further treatment at the last follow-up. The clinical efficacy of the open reduction and THR should be compared in a more convincing case control study, and a more accurate assessment of complications should be concluded with long-term follow-ups.
From our experiences, external fixation was effective for pre-reduction in patients suffering from neglected acetabular fractures with limb discrepancy, but still several contraindications are noteworthy. First, it should not be used in patients with heterotopic ossification, which most commonly occurs in the hip joint, and the incidence after traumatic dislocation was 32% to 37% [27, 28], because traction would be resisted by ectopic bone. As the fixation technique requires stable anchors, it is unfit for patients with an unstable pelvic ring or femur. Furthermore, patients with osteoporosis should not be considered because osteopenic bone may fail under power of traction.
There are limitations in this study including the retrospective design and a small number of patients. This study only reported short-term results and did not compare with other operation strategy. Long-term clinical outcome should be followed and a larger number of patients should be included in the future to evaluate clinical value of our method.