Even with modern methods of fracture care, posttraumatic arthritis still occurs in more than 20% of patients after acetabular fracture. 8 When arthritis becomes severe, causing marked pain and limited function, total hip arthroplasty usually is the only viable salvage procedure. The majority of studies that have evaluated total hip arthroplasty after acetabular fracture have reported it successfully improves function and reduces pain. 4–6,9–12,14 However, compared with hip arthroplasty for most other diagnoses, cemented acetabular component durability in these patients is poorer. 12 One reason for high implant failure rates is the low mean age of patients with acetabular fractures. 14 Other important reasons for implant failure include the problems of acetabular bone loss and deformity, and compromised bone quality, all of which may be adverse for cemented acetabular component fixation.
Two series have suggested that compared with cemented components, uncemented sockets may improve the results of arthroplasty after previous acetabular fracture. 1,14 To the current authors’ knowledge, no series have reported on the long-term results of uncemented sockets for this diagnosis. The purpose of the current series was to evaluate the results of uncemented acetabular components used to treat posttraumatic arthritis after an acetabular fracture at a minimum of 10 years.
MATERIALS AND METHODS
All patients with a previous acetabular fracture treated at the authors’ institution in 1990 or earlier with an uncemented porous-coated Ti acetabular component were included in the study. Patients were identified using an institutional computerized database. For the parameters noted above the series is consecutive. All patients were followed up prospectively and this specific patient cohort was evaluated retrospectively. Patients were followed up routinely at 1 year, 2 years, 5 years, and every 5 years thereafter (or more frequently) after joint arthroplasty. Patients were contacted and asked to return for a clinical examination and radiographs. When this was not possible, patients were asked to complete a letter or telephone questionnaire and forward radiographs for review.
From 1984 through 1990, 33 patients (34 hips) met the inclusion criteria for the study. Twenty-eight patients were men and five patients were women. There were 14 left total hip arthroplasties and 20 right total hip arthroplasties. One patient had bilateral total hip arthroplasties. Fifteen patients (15 hips) had previous internal fixation of the acetabular fracture and 18 patients (19 hips) had been treated nonoperatively. The mean patient age was 49.8 years (range, 19–78 years). Five patients were younger than 30 years, five patients were between 31 and 40 years, four patients were between 41 and 50 years, nine patients were between 51 and 60 years, eight patients were between 61 and 70 years, and two patients were 71 years or older at the time of arthroplasty.
The acetabular component used was a Harris-Galante I or II acetabular component (Zimmer, Warsaw, IN) in 28 hips, an Osteonics Spherical Porous Coated acetabular component (Stryker-Howmedica Osteonics, Allendale, NJ) in four hips, an Omnifit cup (Micro, Stryker-Howmedica Osteonics) in one hip, and an Osteonics Porous Coated Dual Geometry acetabular component (Stryker-Howmedica-Osteonics) femoral in one hip. There were 32 proximally porous-coated uncemented femoral implants and two cemented femoral implants (13 Harris-Galante [Zimmer], five Anatomic [Zimmer], four Osteonics Omnifit Porous [Stryker-Howmedica Osteonics], seven Omniflex Porous [Stryker-Howmedica Osteonics], two Bias [Zimmer], two Mayo [Zimmer], and one Harris Design-2 [Stryker-Howmedica Osteonics]).
A femoral head size of 32 mm was used in three hips, a femoral head size of 28 mm was used in 26 hips, a femoral head size of 26 mm was used in two hips, and a femoral head size of 22 mm was used in three hips. The operative approach was anterolateral in 25 hips, posterolateral in five hips, and transtrochanteric in four hips.
Nine patients had acetabular bone grafts: five had particulate autogenous femoral head grafts to cavitary acetabular bone deficiencies, three had bulk autogenous femoral head grafts to segmental acetabular lesions (two medial wall, one superior wall), and one had a bulk femoral head allograft to a medial segmental wall deficiency.
Of the 33 patients (34 hips), four were lost to clinical followup at less than 10 years, and four died at less than 10 years. For patients not lost to followup, the latest evaluation was by clinical examination in 15 and letter or telephone questionnaire in 14. Of the patients who died, none was known to have a loose or failed acetabular component.
Nine hips had the acetabular metal shell (four hips) and/or acetabular polyethylene liner (five hips) revised. One hip had the acetabular component (shell and polyethylene) revised for acetabular loosening and osteolysis, two hips had the acetabular component (shell and polyethylene) revised for pelvic osteolysis without acetabular loosening, and one hip had the acetabular component (shell and liner) revised for acetabular loosening and instability. One of the loose shells was revised in a patient who was treated with a superior wall acetabular bulk femoral head autograft. Five patients had the acetabular liner alone revised. One was revised for polyethylene wear and femoral lysis, one was revised for polyethylene wear and hip instability, and one was revised for hip instability alone; two liners were revised at the time of a femoral revision (one loose femoral component, one periprosthetic femoral fracture) for polyethylene wear. None was revised for infection.
Seven patients (seven hips) had femoral component revisions. Five of these revisions were done in combination with the acetabular revisions discussed previously and two (both done for femoral loosening) were not done in combination with an acetabular revision. The reasons for femoral revision in the five patients (five hips) who had a concomitant acetabular procedure were femoral loosening and osteolysis in two patients, femoral osteolysis in one patient, femoral loosening in one patient, and periprosthetic femoral fracture in one patient.
Clinical followup at a minimum of 10 years (mean, 11.93 years; range, 10–16 years) therefore was available in 14 patients (15 hips) who had not had revision surgery (11 hips), died (four hips), or been lost to followup (four hips). Of these patients, eight had no pain, six patients had mild pain, no patients had moderate pain, and no patients had severe pain at final followup. Thirteen patients (14 hips) walked with no support, one patient walked with a cane, and no patients required two-arm support. Eleven patients could walk an unlimited distance, one patient could walk six blocks, and two patients (four hips) could walk less than three blocks.
Of the 14 living patients who did not have revision surgery (15 hips) and who were not lost to followup, 11 patients (11 hips) had radiographs at a minimum of 10 years (mean, 11.6 years, range, 10–15 years). The acetabular component was well-fixed in all patients. There was no discernible acetabular osteolysis in nine hips and mild acetabular osteolysis in two hips. One femoral component was loose and there was marked osteolysis of the greater trochanter around two other femoral components. None of the four patients with radiographs taken at less than 10 years of followup had radiographic loosening or osteolysis at their last radiographic followup examination.
Perioperative and postoperative complications included three minor intraoperative femoral fractures associated with uncemented implant placement, one hip dislocation, and a possible but unconfirmed superficial infection for which the patient was treated with a superficial hip wound debridement.
The current authors showed that total hip arthroplasty with an uncemented porous-coated Ti socket for a diagnosis of posttraumatic arthritis after acetabular fracture provided a very low rate of acetabular component shell loosening. Only two of 26 acetabular components observed until revision or for at least 10 years were loose. The rate of loosening reported in this series is dramatically less than that reported for cemented acetabular components 12 done after previous acetabular fracture and is consistent with some other reports, with shorter followup intervals, of uncemented acetabular components for this diagnosis. 1,14 However, the overall rate of operation in the current series remained high, primarily because of the problem of acetabular polyethylene liner wear in this young patient population. The same general trends, good fixation but high rates of osteolysis, have been reported for first and second generation uncemented cups in young patients. 2
The strengths of the current study include the long followup and the completeness of followup in this group of patients with previous trauma in whom obtaining followup can be difficult. A weakness of the current study is the several uncemented socket designs used during the study; however, all sockets were porous-coated Ti implants of first or second generation design with a modular polyethylene insert.
Patients with previous acetabular fractures frequently have acetabular bone deformity, acetabular bone loss, poor acetabular bone quality, and retained hardware from previous procedures. 3 Despite these challenges, uncemented sockets had excellent fixation in this setting just as they have in multiple series of revision total hip arthroplasty in which some of the same acetabular anatomic and biologic problems are present. In most patients, the procedure could be done without additional bone grafting, but grafts (autologous in all but one patient) were used in nine of the 34 hips. Uncemented sockets were compatible with the particulate and moderate sized bulk grafts 13 that were needed in this series.
The greatest long-term problem with uncemented sockets in this patient population was polyethylene wear and periprosthetic osteolysis. Wear and osteolysis accounted at least in part for six of the nine revision operations in this patient cohort. The most likely explanation for the high rate of osteolysis in these patients is the young patient age combined with monoarticular arthrosis, a combination that leads to a high activity level. For patients with cemented implants, this high activity level probably mostly leads to failures caused by loosening, whereas for patients with uncemented implants, these failures mostly are manifested as osteolysis.
The bearing surface in all patients in the current series was a metal femoral head on conventional polyethylene that had been gamma irradiated in air. In the future, improvements in conventional polyethylene and alternative bearing surfaces that have become available since the time of the study may reduce the rate of failure because of bearing surface wear. The current series shows that uncemented porous-coated sockets markedly reduce long-term socket loosening rates in this challenging patient population. The combination of uncemented acetabular components and improved bearing surfaces may provide more durable arthroplasties for this challenging patient group in the future.
1. Bellabarba C, Berger RA, Bentley CD, et al: Cementless acetabular reconstruction after acetabular fracture. J Bone Joint Surg 83A:868–876, 2001.
2. Berger RA, Jacobs JJ, Quigley LR, Rosenberg AG, Galante JO: Primary cementless acetabular reconstruction in patients younger than 50 years old: 7- to 11-year results. Clin Orthop 344:216–226, 1997.
3. Berry DJ: Primary hip arthroplasty: Operative management problems: Total hip arthroplasty following acetabular fracture. Orthopedics 22:837–839, 1999.
4. Boardman KP, Charnley J: Low-friction arthroplasty after fracture-dislocations of the hip. J Bone Joint Surg 60B:495–497, 1978.
5. Coventry MB: The treatment of fracture-dislocation of the hip by total hip arthroplasty. J Bone Joint Surg 56A:1128–1134, 1974.
6. Jimenez ML, Tile M, Schenk RS: Total hip replacement after acetabular fracture. Orthop Clin North Am 28:435–446, 1997.
Judet R, Judet J, Letournel E: Fractures of the acetabulum: Classification and surgical approaches for open reduction. J Bone Joint Surg 46A:1615–1646, 1964.
8. Letournel E: Fractures of the Acetabulum. Ed 2. New York, Springer-Verlag 1993.
9. Malkin C, Tauber C: Total hip arthroplasty and acetabular bone grafting for unreduced fracture-dislocation of the hip. Clin Orthop 201:57–59, 1985.
10. Pritchett JW, Bortel DT: Total hip replacement after central fracture dislocation of the acetabulum. Orthop Rev 20:607–610, 1991.
11. Rogan IM, Weber FA, Solomon L: Total hip replacement following fracture dislocation of the acetabulum. J Bone Joint Surg 61B:252, 1979.
12. Romness DW, Lewallen DG: Total hip arthroplasty after fracture of the acetabulum. J Bone Joint Surg 72B:761–764, 1990.
13. Spangehl MJ, Berry DJ, Trousdale RT, Cabanela ME: Uncemented acetabular components with structural femoral head autografts for acetabular reconstruction in developmental dysplasia of the hip: Results at 5 to 12 years. J Bone Joint Surg 83A:1484–1489, 2001.
14. Weber M, Berry DJ, Harmsen WS: Total hip arthroplasty after operative treatment of an acetabular fracture. J Bone Joint Surg 80A:1295–1305, 1998.
Richard A. Brand, MD; and Joseph C. McCarthy, MD—Guest Editors