A thirty-five-year-old male computer technician with osteopetrosis presented with progressively worsening symptoms of osteoarthritis in both hips. The left hip was more symptomatic than the right. His orthopaedic history included approximately forty previous fractures of various bones, none of which involved the femora. Over the course of the previous nine years, he had experienced increasing bilateral hip pain that necessitated a reduction of his daily functional activities. He walked with a cane and was being treated with nonsteroidal anti-inflammatory agents. The University of California at Los Angeles (UCLA) hip scores were 4 (on the right) and 2 (on the left) for pain, 5 for walking, 6 for function, and 3 for activity.
On physical examination, the patient had a markedly antalgic gait and had a positive Trendelenburg sign in both hips. The left hip had 90° of flexion with a 15° flexion contracture, 20° of abduction, 15° of adduction in extension, and a 20° arc of rotation in extension. The right hip had the same range of motion measurements except that adduction in extension was 20°. Plain radiographs revealed the classic manifestations of osteopetrosis, including diffuse sclerosis of the femoral heads and the pelvis, markedly reduced or absent medullary canals in the proximal portion of the femora, and severe degenerative osteoarthritis of both hips (Fig. 1-A).
We planned to perform a staged bilateral hybrid metal-on-metal resurfacing total hip arthroplasty procedure, and it was elected to first proceed with resurfacing of the more symptomatic left hip. Before the operation, a computed tomography scan was performed to precisely measure the dimensions of the head, neck, and acetabulum in order to more accurately determine the optimal size of the components and to minimize the amount of femoral and acetabular reaming that would be necessary. Several sets of new femoral and acetabular reamers and tungsten carbide drill-bits were obtained to ensure the availability of sharp cutting tools for the dense, sclerotic, and brittle bone.
The operation on the left hip was performed through a posterolateral approach, with use of a previously described technique17,20. Intraoperatively, there was no cancellous bone in the femoral head or the acetabulum (Fig. 1-A). The femoral preparation began with pin-centering procedures, with use of a tungsten-carbide-steel one-eighth-inch (3.18-mm) drill-bit because an ordinary Kirschner wire would not penetrate the sclerotic osteopetrotic bone. Reaming of the head was done slowly in incremental steps. Multiple small drill-holes (more than thirty) were created in the dome and chamfered area of the femoral head to increase the surface area for cementation. A 50-mm femoral component was used. Simplex-P regular viscosity cement (Stryker Orthopaedics, Mahwah, New Jersey) was used, but only while it was in a state of relatively low viscosity so that the component could be press-fit with hand pressure only, without the need for hammer blows, which might cause a neck fracture. Very sharp acetabular reamers (Model DR; Precimed, Exton, Pennsylvania) were used to expand the acetabular cavity to 56 mm to achieve a press-fit of 1 mm with the 56-mm (anticipated outside diameter of 57 mm) Conserve Plus socket (Wright Medical Technology, Arlington, Tennessee). Despite the use of new, very sharp reamers, the bone was extremely difficult to ream, leading to one of the most time-consuming stages of the procedure. The assessment of the depth of the acetabular cavity involved the use of a template with a see-through gauge, while sphericity and size were checked with use of ring gauges. The component was attached to the inserter and impacted with multiple hammer blows by the assistant while the surgeon held the inserter in proper alignment17. The procedure was uneventful, with an estimated blood loss of 500 mL. No blood transfusion was necessary. The total time for the operation was approximately two hours and thirty minutes.
The immediate postoperative radiographs revealed two fractures in the medial wall of the acetabulum, which undoubtedly occurred during the impaction of the component (Fig. 1-B). In contrast, we had never experienced an acetabular fracture in patients with osteoarthritis when we used our customary technique of achieving a 1-mm press-fit. Apparently, the magnitude of the press-fit was too great for this brittle bone. As a result, the patient had an unusual amount of pain postoperatively, and he was placed on a nonweight-bearing regimen for rehabilitation. The pain subsided gradually over the next several weeks, and radiographs made two months postoperatively suggested to us that the acetabular fractures were healing (although the fracture lines were still visible) and that the component had not undergone a change in position (Fig. 1-C). Three months later, the patient could walk with a cane in his left hand to minimize the osteoarthritic pain in the uninvolved right hip. He tended to have slight discomfort in the left hip after walking.
Five months after the resurfacing procedure on the left hip, the patient underwent resurfacing of the right hip. In order to minimize the risk of acetabular fracture, three "custom" Conserve Plus sockets, which ranged in size in increments of 0.5 mm, were obtained to allow for a wider choice of socket size during the procedure. Their outer diameters were measured precisely with use of a shadow-graph comparator (S-T Industries, St. James, Minnesota) with a lens of five times magnification (Model #20-1600; S-T Industries). (The actual size of the "off-the-shelf" components ranges between 1.25 and 1.75 mm greater than the nominal dimension. The variation in size is due to the variability of the thickness of the sintered porous beads.) To further reduce the risk of fracture, the acetabulum was carefully reamed to 57 mm rather than 56 mm (as in our usual technique), and this was verified with the metallic ring gauges to ensure a more precise press-fit of no more than 0.5 mm for the custom 56-mm Conserve Plus socket (actual size, 57.5 mm). The impaction force necessary for seating the component was considerably less than that used for the left hip. The femoral bone preparation was uneventful with use of the special instruments that had been obtained for the left hip; for the purpose of keeping them sharp, these instruments had been set aside and not used for other operations. The operation time was just under two hours. The estimated blood loss was 300 mL, and no blood transfusion was given. The patient was placed on an ordinary postoperative rehabilitation program17. Walking began on the first postoperative day, with weight-bearing allowed as tolerated. Crutches were used for three weeks.
At the patient's most recent evaluation as of the time of writing (twenty-five months after the operation on the right hip and thirty months after the operation on the left hip) (Fig. 1-D), the UCLA hip scores were 9 (on the left) and 10 (on the right) for pain, 10 for walking, 10 for function, and 7 for activity. The patient reported occasional pain in the left hip (perhaps related to the delayed union of the intraoperative fractures), but the pain was so minimal that it did not restrict walking or any function. The range of motion for both hips improved to 120° of flexion, 45° of abduction, 25° of adduction, and 85° of rotation arc (assessed with the hip in extension). The patient said that he now occasionally and carefully participates in softball batting practice with his children.