A fifty-seven-year-old woman had a history of a right primary total hip replacement with an uncemented component in 1988 to treat osteoarthritis. This was followed by acetabular revision for isolated aseptic loosening in 2003. The patient presented to the senior author in 2004, eleven months postrevision, with symptoms of increasing inability to bear weight and progressive right hip and buttock pain. Radiographs demonstrated a stable uncemented femoral component as well as gross acetabular loosening and medioinferiorly placed screws (Fig. 1). The patient had a large radiolucency in the area of zone three and superolateral migration of the acetabular component. Preoperative radiographs, including an anteroposterior pelvic view, anteroposterior and lateral views of the hip, and Judet views, demonstrated substantial osseous loss to the posterior wall, posterior column, and acetabular dome. Figure 2 is a representation of the acetabular bone loss seen at the time of revision. The preoperative evaluation included normal serology (C-reactive protein level <3 mg/dL and an erythrocyte sedimentation rate of 10 mm/h), and hip aspiration yielded a negative culture.
Operative Technique
A Kocher approach to the right hip was used, incorporating much of the two previous posterolateral incisions. After posterior hip dislocation and modular femoral head removal, a circumferential capsulectomy was performed. The femoral stem component (Integral; Biomet, Warsaw, Indiana) was deemed to be stable without trunnion damage, and the femur was retracted over the anterior column. The acetabular component was grossly loose and was easily removed.
The acetabular anterior column, anterior wall, and medial wall were all intact. The superior weight-bearing area was estimated to be 50% deficient. There was extensive posterosuperior deficiency, such that the medial wall of the acetabulum communicated directly with the sciatic notch, with the nerve being visible posteriorly in the wound. Although the acetabular component had not migrated in excess of 3 cm (screw fixation had not yet completely failed), the encountered acetabular defect was considered a Paprosky type-IIIA defect because of the pattern and magnitude of the posterosuperior bone loss and the impending “breakout” mode of failure.
The acetabular center was prepared with serial hemispherical reamings up to 58 mm until the reamer matched the profile of the remaining anterior weight-bearing roof and anterior column. This resulted in size selection for the anticipated hemispherical implant (58 mm) and the template for the posterosuperior osseous reconstruction. The posterosuperior acetabular defect, once sized, was reconstructed with an ipsilateral fresh-frozen (AlloSource) allograft from the proximal part of a femur, chosen for its conformity with the defect. The ipsilateral side was chosen because the intertrochanteric line and greater trochanter are easily contoured into a concave surface that can accept the acetabular component. In addition, this allowed the calcar to be oriented superolaterally to recreate the weight-bearing roof. The graft was shaped with an oscillating saw and a burr to mirror the acetabular defect. The subtrochanteric region of the graft was placed superiorly along the ilium and fashioned into a unicortical plate to allow fixation of the graft to the ilium via 7.3-mm screws (Fig. 3). The greater trochanter and intertrochanteric line were oriented inferiorly and posteriorly to recreate the posterior buttress. Once the graft was fixed to the illium, the reconstructed acetabulum was re-reamed to a 58-mm hemisphere. Small gaps between the graft and host bone were filled via reverse reaming of particulate graft. Line-to-line fit with a 58-mm hemispherical trial was excellent and permitted sufficient stability.
A 58-mm Trilogy acetabular component (Zimmer, Warsaw, Indiana) was inserted in approximately 20° of anteversion and 45° of abduction, and it was fixed with ancillary neutralizing screws. Overall host-bone contact with the implant was estimated at approximately 50%. Because of the pattern of bone loss and the direction of the weight-bearing force vector, more than 50% of the load in weight-bearing was being buttressed by the graft. A 40-mm cobalt-chromium bearing (Longevity; Zimmer), made from highly cross-linked polyethylene, was chosen for the purposes of minimizing prosthetic impingement and achieving maximal stability.
Postoperative Course
The patient was flat-foot weight-bearing for six weeks. A custom-fit hip orthosis was placed on postoperative day one and was worn for three months. One month postoperatively, radiographs confirmed stability of the reconstruction (Fig. 4). From six weeks to six months postoperatively, the patient was kept at 50% weight-bearing, and subsequently, she was advanced to full weight-bearing.
The patient was followed yearly with radiographs, including Judet views, without any appreciable resorption of the graft. Figure 5 shows the most recent image, taken seven and one-half years postoperatively, demonstrating osseous integration of the acetabular component. The leg lengths were equal clinically, and the patient did not have symptoms of any appreciable leg-length discrepancy. At the most recent follow-up, the patient reported unlimited walking with occasional cane use, and she denied any hip pain.
The Paprosky classification categorizes acetabular defects based on several radiographic criteria7; however, it is a preoperative radiographic classification that does not always account for the magnitude of bone loss encountered after implant removal. This case report describes a Paprosky type-IIIA defect with posterosuperior bone loss8. The challenge was to reestablish the hip center, fill the defect to buttress any reconstruction, and prevent acetabular cup migration.
One option to deal with a type-IIIA defect is to fit a hemispherical acetabular component with multiple screws in a high hip center position9, which results in a limb-length discrepancy. A second option is cancellous grafting in combination with cage reconstruction that gains fixation in the ilium and the ischium10. The use of a cemented cage in the reconstructed acetabular defect has also been described11. When an anatomic hip center is desired, a structural graft or metal augments that support a hemispherical cementless cup can be used3. Augments fill the defect and help buttress the reconstruction so that a hemispherical acetabular implant can develop osseous integration, but the defect is still effectively present. Bulk allograft can be used in a similar manner with the advantage of adding bone stock.
In 1986, Gerber and Harris described the use of a femoral head bulk allograft with good overall results12. When using bulk allograft, better results are reported if near-normal hip mechanics are restored and if greater than 50% of host-bone support is available13. The figure-of-seven distal femoral graft, as described by Sporer et al., had success that equaled the results of a femoral head bulk allograft6. Woodgate et al. reported an 80.4% survival rate at ten years in fifty-one patients with structural allografts14. Certainly, not all case series of hip revision with use of bulk allografts have shown such a high survival rate. Hooten et al. used structural acetabular allografts for reconstruction of acetabular bone loss, and saw radiographic evidence of failure in twelve of twenty-seven (44%) patients at a mean of forty-six months15. A postmortem analysis of two structural allografts for reconstruction of acetabular bone loss, with apparent radiographic healing and stability, showed no evidence of ingrowth between the cup and the structural graft16.
In this case report, an ipsilateral bulk allograft from the proximal part of a femur was suggested as a viable source of graft material in addition to grafts previously described in the literature. The intertrochanteric graft from the proximal part of a femur was used in our patient primarily because of the similar geometry between the defect and the proximal part of the femur. A relatively small amount of machining with the saw and the burr was needed to conform the graft to the defect. We believe that, for this specific pattern of bone loss, the native anatomy of the proximal part of the femur makes it an attractive choice because of its ease of use. The proximal part of the femur was also chosen for other anatomic factors. The graft was positioned with the subtrochanteric region superiorly and the greater trochanter posteriorly, such that the secondary compression trabecular group of the calcar formed the superior weight-bearing dome. The calcar was positioned in this manner so that the trabeculae would be oriented in order to theoretically provide the best buttress for the force of weight-bearing17,18. The forces transmitted to the proximal part of the femur with normal weight-bearing15,16 would, in theory, be similar in direction to the forces on the graft. However, any possible biomechanical advantage of the secondary compression trabeculae is purely theoretical and not proven.
We report a single case of a successful reconstruction of acetabular bone loss with use of a bulk allograft from the proximal part of a femur at eight years of follow-up. Certainly, long-term follow-up with a larger number of patients is needed for additional evaluation of this technique. Recognizing the historical limitations and inconsistent outcomes of bulk allograft when used for reconstruction of acetabular bone loss, we suggest that the proximal part of the femur may be a unique source of bulk allograft for the arthroplasty surgeon. Although we do not propose that the proximal part of the femur is a superior source than other bulk allografts, the geometry of the proximal part of the femur may make this a technically desirable graft choice.
Disclosure: None of the authors received payments or services, either directly or indirectly (i.e., via his or her institution), from a third party in support of any aspect of this work. One or more of the authors, or his or her institution, has had a financial relationship, in the thirty-six months prior to submission of this work, with an entity in the biomedical arena that could be perceived to influence or have the potential to influence what is written in this work. No author has had any other relationships, or has engaged in any other activities, that could be perceived to influence or have the potential to influence what is written in this work. The complete Disclosures of Potential Conflicts of Interest submitted by authors are always provided with the online version of the article.