Session VIII - Pelvis

Nonoperative Treatment of Fractures of the Acetabulum

Michael Stover, MD; Matthew J. Weresh, MD; Michael J. Bosse, MD; Stephen H. Sims, MD; James F. Kellam, MD, Carolinas Medical Center, Charlotte, North Carolina

Purpose: To determine if radiographic displacement of fractures of the acetabulum occurs prior to radiographic union in patients meeting current recommendations for nonoperative care treated with an aggressive approach to patient mobilization.

Current recommendations for nonoperative treatment of fractures of the acetabulum are based upon observations of patients treated with and without open reduction and internal fixation. These include displacement of > 2 mm involving the superior articular surface and maintenance of congruence between the femoral head and acetabulum. Additional contraindications to surgery may include age, infection, pre-existing arthrosis, severe comminution, osteoporotic bone and systemic illness. Methods of nonoperative treatment may include traction or bed rest along with joint mobilization. Concerns over the risks of prolonged bed rest and the possibility of fracture displacement have prompted some to recommend percutaneous fixation of non- or minimally displaced fractures. Documentation of the natural history of these injuries until union in a subset of patients using an aggressive approach to nonoperative care may help define an appropriate care regimen.

Methods: All patients with fractures of the acetabulum admitted to the investigating institution were identified from a trauma registry. Of 338 patients identified from 1992-1997, 48 fractures in 47 patients (14.4%) were treated without surgery. All fractures were initially evaluated with the use of three standard radiographs: anterior-posterior (AP) two 45° oblique views, and computed tomography (CT). Fractures were classified according to the criteria of Judet and Letournel and the OTA fracture classification system. The greatest displacement of an anatomic landmark was determined by direct measurement without adjustment for magnification. Non-displaced fractures were defined as a visible fracture line with <1mm of displacement. The roof arc angle was determined for each radiograph, and involvement of the superior articular surface was defined as a roof arc of > 45 on either of the plain radiographs or a fracture line crossing the joint within 10 mm of the subchondral arc on CT. Joint congruency was determined by direct measurement of the radiographic joint space at three positions along the radiographic roof and directly compared to the contralateral normal hip joint. Articular surface impaction was noted or confirmed by CT scan. Review of hospital records provided information on associated injuries and specifics of the patient's nonoperative care. Each subsequent set of radiographs was used to determine if any displacement or loss of congruence occurred prior to fracture union. Inclusion in the final analysis required that the patient had not been treated in traction for more than one week, adequate radiographic follow-up was available until radiological evidence of union, and the patient had been advanced to full weight bearing. Thirty-three fractures in 32 patients, 21 males and 11 females, fit the criteria for inclusion. A motor vehicle accident was the mechanism of injury in 28 patients. The average age on admission was 32.8 years (range 13-76). Twenty-four fractures (72.7%) were simple fracture types, the majority of which were classified as a transverse pattern. Twenty-two patients (66%) had a fracture involving the superior articular surface. Twenty-nine fractures (88%) maintained congruence between the femoral head and acetabulum on admission radiographs. Twenty-four fractures (73%) had displacement of 2 mm or less. Twenty-nine met criteria for nonoperative treatment, three had relative contraindications to surgery, and one refused surgery. Twenty-two patients had associated orthopaedic injuries, including nine with concomitant pelvic ring injuries. Follow-up ranged from 2-36 months (aver. 6.7 mos.) Two groups of patients were retrospectively recognized: patients mobilized in less than one week with instructions for touch-down weight bearing vs. those requiring bed rest. Joint mobilization was allowed as tolerated. Patients designated as "bed rest" were allowed up to an upright chest position (bed to chair or head of bed up) during their treatment period, if medically able.

Results: Thirty-one of 33 fractures healed without change in position. None of the 13 fractures that were non-displaced on admission radiographs displaced, and 7 were mobilized early in their treatment course. No patient in the group who was mobilized to assisted ambulation lost position of their fracture. Of those patients in the "bed rest" group, 13 of 19 patients had associated contralateral lower extremity or pelvic ring injuries that did not allow them to undergo gait training and early mobilization regardless of the injury to the hip joint. Two had traumatic brain injuries and were unable to cooperate with physical therapy. Only one patient with an isolated non-displaced anterior column fracture was treated without early-restricted ambulation. Two fractures from Group II displaced. Each patient had involvement of the superior articular surface. Both had evidence of articular surface impaction on admission radiographs or CT. One had an associated pubic symphysis dislocation, while the other was positioned on the involved hip for treatment of a deep infection on the contralateral hip.

Discussion: In a subset of nonoperatively treated fractures of the acetabulum, with attempts to mobilize the patients out of bed early in their treatment course, 93% of the fractures healed without a change in their position. Once the decision is made to treat the patient without surgery, evaluation of the patient's associated injuries and physical capabilities may allow early mobilization of the patient to an upright position or with instructions for touch-down weight bearing of the affected lower extremity with little fear of further displacement.

Conclusion: Close radiographic follow-up of nonoperatively treated intra-articular fractures is essential, and, if displacement occurs, the treatment course should be reevaluated. Our study does not appear to support the need for routine use of traction or for operative intervention to maintain alignment of non- or minimally displaced fractures of the acetabulum.