Session VII - Pelvic Trauma

Irreducible Fracture-Dislocation of the Hip: A Severe Injury with a Poor Prognosis

Michael D. McKee, MD, FRCS(C), Michael E. Garay, Research Assistant, Emil H. Schemitsch, MD, FRCS(C), David J. G. Stephen, BSc, MD, FRCS(C), Hans J. Kreder, MD, FRCS(C)

St. Michael's Hospital and the Sunnybrook Health Science Center, Toronto, Ontario, Canada

Purpose: Irreducible fracture-dislocations of the hip are a difficult management problem. We reviewed 25 cases to determine associated injuries, optimal treatment methods, prognostic factors and outcome.

Conclusion: A fracture-dislocation of the hip which is irreducible by closed means is a severe injury with a high incidence of associated sciatic nerve injury, femoral head or neck fracture and subsequent avascular necrosis. The structures typically blocking reduction include fragments of the posterior lip and/or femoral head, and posterior soft-tissue elements. Irreducible fracture-dislocations of the hip have a poor prognosis, a high incidence of associated injuries and complications, and are technically challenging to fix. Although operative repair is difficult due to distortion of the anatomy and the severe degree of injury, the best chance for success lies with prompt, definitive open reduction.

Summary of Methods, Results, and Discussion: We reviewed the trauma databases of two Level One trauma centers from 1/90 to 3/96 and identified 25 patients who had sustained an irreducible fracture dislocation of the hip. Inclusion criteria included patients with a traumatic hip dislocation, no contact between the femoral head and the dome of the acetabulum following attempts at closed reduction under general anaesthetic, and the requirement of an open procedure to obtain concentric reduction of the hip. Thus, hips that were reduced but not concentric, or remained "subluxated" were not included. There were twenty males and five females with an average age of 41.8 years (range 17 to 81 years). Seventeen patients were involved in a motor vehicle accident, three fell from a height, three were pedestrians struck by vehicles, one was struck by a subway train and one was injured water-skiing. Eighteen patients had multiple fractures and fifteen patients had sustained multi-system injury. While no patient had an associated vascular lesion, seven had an associated sciatic nerve palsy, an incidence of 28%. There were nineteen posterior and six central dislocations. All patients had suffered a fracture of some form: there were eleven fractures isolated to the posterior wall or column and ten complex fracture patterns involving both columns, transverse, or "T" type fractures. There were nine associated femoral neck or head fractures, an incidence of 36%. Open reduction was performed in less than six hours after injury in six patients, between six and twelve hours after injury in nine, and between twelve and twenty-four in five. Five patients had the open reduction performed greater than 24 hours post-injury due to severe associated injuries or transfer from a peripheral center. Sixteen hips were reduced through a posterior approach, four had combined anterior/posterior approaches and three had a triradiate approach and two an anterior approach alone. Mean operative time for the hip procedure was 3.8 hours wtih a range from 1.8 to 6.5 hours. The structure blocking the reduction of the hip was judged to be acetabular fragments in thirteen, soft tissue (capsule, piriformis, glutei, obturator muscle, sciatic nerve) in nine and femoral head fragments in seven. Four patients were thought to have multiple causes. Mean follow-up was 33 months with a range from 6 to 72 months. Overall, outcome was poor. As rated by the Thompson and Epstein scale, there were 4 excellent, 3 good, 2 fair and 14 poor results. Two patients were lost to follow-up. There were numerous reasons for the poor results (avascular necrosis - 6, post-traumatic OA - 4, persisting sciatic nerve injury - 2, ankylosis - 1, and associated ipsilateral lower extremity injuries - 1). Six patients underwent or are awaiting total hip arthroplasty. Thus, only 6 of 24 patients had a satisfactory result (Thompson/Epstein good or excellent).