Midterm results of uncemented hemispheric acetabular components for revision total hip arthroplasty have been mostly favorable and show the utility and versatility of this means of acetabular revision.12,16,17,19 However, when large superior segmental acetabular bone deficiencies are present, a standard hemispheric uncemented socket cannot be placed solely on native bone at the normal hip center. In these circumstances, traditional treatment options for reconstruction include structural bone grafting with acetabular reconstruction at an anatomic hip center, acetabular reconstruction at a superiorly positioned high hip center, or when the defect allows, conversion of the oblong defect to a larger hemispheric defect followed by placement of a jumbo socket.
In the last decade, off-the-shelf oblong porous-coated components have become available. The theoretical advantages of these devices include an increased surface contact area between the porous metal and native acetabular bone, avoidance of structural bone grafts, and the potential to normalize the center of hip rotation. The theoretical disadvantages include the lack of bone stock restoration and few published clinical results. DeBoer and Christie2 reported good results in 18 hips (15 revisions) using a porous uncemented oblong bilobed socket and Köster et al10 reported mostly favorable results using a different oblong uncemented socket in 102 hips, 98 of which were revisions. To date, the authors are aware of no published reports in the peer reviewed literature reporting the results of using these devices exclusively for revision hip arthroplasty. The purpose of this study was to evaluate the results of using an oblong porouscoated noncustom acetabular component of a single design exclusively in acetabular revision surgery associated with a large superior segmental bone deficiency. This study evaluated clinical results, including pain relief and leg length restoration; radiographic results, including implant fixation and hip center restoration; and complications associated with the procedure.
MATERIALS AND METHODS
The results of 38 revision total hip arthroplasties performed using a Ti bilobed oblong off-the-shelf porous-coated uncemented acetabular component of a single design (SROM-Standard Range of Motion, DePuy, a Johnson and Johnson Company, Warsaw, IN) were assessed a minimum of 2 years after surgery. The acetabular revisions were performed from 1992 through 1995 at four centers and, for the implant under study, represent a consecutive series by the surgeons included from each of those institutions. The implant was placed in 13 hips, nine hips, nine hips, and seven hips at each of the four institutions. The decision to use this implant, rather than an alternative reconstruction technique, was based on the preference of the operating surgeon according to the pelvic bone deficiency encountered at surgery. In general, the implant was used when a large superior lateral segmental deficiency was present in which a satisfactory fit could be obtained with trial oblong bilobed implants. The study was approved by the Institutional Review Board at the authors' institutions.
There were 14 men and 24 women. The mean patient age was 60.5 years (range, 35-88 years). There were 20 right hips and 18 left hips. The mean patient weight was 80 kg (range, 53-130 kg). The underlying diagnosis before total hip arthroplasty was osteoarthritis in 15 patients, posttraumatic arthritis in nine patients, developmental dysplasia of the hip in seven patients, rheumatoid arthritis in three patients, and other in four patients. The procedure was the first ipsilateral hip revision in 18 patients, second in 12 patients, third in five patients, fourth in three patients, and sixth in one patient. The failed implant removed at the time of the index revision arthroplasty was cemented in 30 patients and uncemented in eight patients. The bony acetabular bed after implant removal consisted partially of healed (but probably incompletely revascularized) allograft in two patients. In one patient, the allograft made up more than 50% of the acetabular surface, and in one, less than 50%.
Patients were seen in followup for interview and examination or answered a letter or telephone questionnaire. The Harris hip score was calculated according to published criteria; when patients provided data by answering a questionnaire, the score was modified by assigning arbitrarily 4 points of 5 for motion.6 Leg lengths were measured clinically using blocks or by measuring apparent leg length discrepancy. Review of this specific patient population was performed retrospectively. Radiographs were reviewed by the senior author at each institution, who also was the operating surgeon and who was not blinded to the outcome of the procedure. Acetabular prostheses to bone interfaces were evaluated using a modification of the method of DeLee and Charnley.4 For the current study, Zone 1 was considered to encompass the entire superior lobe of the socket, Zone 2 the superior half of the lower lobe, and Zone 3 the inferior half of the lower lobe. All patients had superior segmental acetabular bone deficiencies. The prerevision acetabular bone deficiency was categorized according to D'Antonio et al1 as Type I (superior segmental) in five patients, Type III (combined cavitary and segmental) in 32, and Type IV (pelvic discontinuity) with associated superior segmental bone loss in one. The hip center was measured on the anteroposterior radiographs of the pelvis as the shortest distance between the center of the femoral head and an interteardrop line.
Acetabular components of the design used in this study are Ti and have a single metallic shell with two lobes. The lobes are of equal diameter for a given component. The diameters range from 51 mm to 66 mm in 3-mm increments. For each diameter, the cup is available in two versions, one with the center of the upper lobe offset superiorly 15 mm compared with the center of the lower lobe (E-15), and one with the upper lobe offset 25 mm (E-25) (Fig 1). For the E-15 implants, the lower lobe is adducted (more horizontal) 10° relative to the upper lobe, whereas for the E-25 implants, the lower lobe is adducted (more horizontal) 20° and anteverted (more flexed) 15° relative to the upper lobe. Double reamers are available to prepare the acetabulum to fit the implant. The implants provide fixation with 6.5-mm diameter Ti screws through the dome of the implant and the upper lobe of the implant and 5-mm screws through the periphery of the implant. Modular polyethylene liners are available with 0°, 10°, 15°, and 20° elevated rims, with a lateralized hip center and a constrained liner. The liner is fixed in the shell with locking pegs or 5-mm screws.
The median implant size was 57 mm (range, 51-66 mm). There were 22 E-15 and 16 E-25 implants. All metallic implants were fixed to bone with screws. The number of 6.5-mm screws ranged from one to four (median, three screws), and the number of 5-mm screws ranged from zero to four (median, two screws). A 28-mm head size was used in 26 patients, a 32-mm head size in nine patients, a 22-mm head size in two patients, and a 26-mm head size in one patient. A flat polyethylene insert was used in three patients, a 10° elevated rim in 18, a 15° elevated rim in seven, a 20° elevated rim in four, and a lateralized insert in six. No constrained polyethylene inserts were used.
Revision total hip arthroplasty was performed using an anterolateral approach in seven patients, a posterolateral approach in 21 patients, and a transtrochanteric approach in 10 patients. Concomitant femoral revision was performed in 19 patients. No bulk structural acetabular bone grafts were used. Particulate bone grafting (autograft in seven hips, allograft in 14 hips, and autograft and allograft in one hip) was used in 22 hips to fill remaining cavitary bone deficiencies. The patient with a pelvic discontinuity also had the pelvis stabilized with a posterior plate.
Patients were followed up a mean of 3 years (range, 2-5 years) after surgery. No patients were lost to followup, and none have died. One patient underwent revision surgery for aseptic loosening 5.2 years after the index revision. The patient had a structural acetabular bone graft from a previous revision surgery, and more than 50% of the bilobed socket was placed against the bone graft. The oblong socket failure was associated with precipitous failure of the socket and collapse of the allograft, and the patient's hip required reconstruction with an antiprotrusio device. No other patients have had their implants removed, and no patients have had another ipsilateral hip operation (Fig 2).
The mean preoperative modified Harris hip score was 54 (range, 28-86); the mean hip score at latest followup was 90 (range, 79-100). The improvement was statistically significant (p < 0.0001). At latest followup, 23 patients had no hip pain, 10 had mild hip pain, two had moderate hip pain, and none had severe hip pain. One of the patients with moderate hip pain had a trochanteric nonunion.
The average preoperative leg length discrepancy was 15 mm shorter on the ipsilateral side (range, 50 mm shorter-12 mm longer); this was decreased to an average of 0 mm (range, 40 mm shorter-30 mm longer) after surgery. Thus, the mean increase in leg length was 15 mm (range, 7 mm shortening-30 mm lengthening). In two patients in the series, the surgically treated leg was more than 1 cm longer than the opposite leg after revision surgery (in one 20 mm, in one 30 mm). In the patient with a 30-mm leg length discrepancy, the opposite leg was short because of intrapelvic protrusion of another failed total hip arthroplasty; the lengths later were equalized when the opposite hip arthroplasty was revised.
The prerevision position of the hip center averaged 37 mm (range, 22-75 mm) superior to a line drawn between the radiographic teardrops before revision and averaged 25 mm (range, 12-50 mm) superior to the same line after revision. In 31 of the 38 hips, the hip center of rotation after reconstruction was 30 mm or less superior to a line drawn between the radiographic teardrops. The mean abduction angle of the lower lobe of the metal socket was 42° (range, 28°-57°).
All patients had radiographs a minimum of 2 years after surgery. Serial radiographs showed acetabular component migration of 3 mm in one case and 7 mm in another case within the first year. The implants subsequently did not show progressive migration in either case at 2 and 3 years, respectively. The number of screws used to fix the two sockets with early migration (five screws and four screws respectively) did not differ significantly from the number of screws used in the group as a whole. Excluding the patient who required another revision surgery, none of the other patients had implants with measurable acetabular migration or change in position. On the latest radiographs, two implants showed bead shedding from the porous surface but did not have measurable migration or change in position at 2.1 and 5 years after surgery. Neither of the patients with bead shedding had pain. One of the patients with bead shedding had a complete radiolucent line at the prosthesis-bone interface. None of the other 36 sockets had a complete radiolucent line at the prosthesis-bone interface, but a partial nonprogressive line was seen in 24 hips (involving all or part of one zone in 16 hips and all or part of two zones in eight hips). Twelve hips had no radiolucent lines at the prosthesis-bone interface. No implants had broken fixation screws.
Complications included three postoperative partial palsies of the peroneal branch of the sciatic nerve and two postoperative dislocations. Neither dislocation was recurrent, and both were treated with closed reduction. Two of the three nerve palsies occurred in patients who had undergone a posterolateral approach to the hip. The mean amount of leg lengthening in the three patients with nerve palsies was 20 mm (range, 12-30 mm), compared with 15 mm for the group as a whole. In two of these patients, the leg lengths were equal after the operation, and in one, the surgically treated leg was 30 mm longer than the contralateral side (attributable to a short opposite leg from another failed total hip arthroplasty). The etiology of the nerve injury was uncertain in the three hips. In one patient, the nerve injury was associated only with dysesthesia and numbness but no weakness. One of the other nerve palsies resolved completely; one persisted, and the patient required an ankle-foot orthosis. There were no infections and no cases of symptomatic deep venous thrombosis or pulmonary embolism.
This study assessed the early results of revision total hip arthroplasty performed for large superior segmental acetabular bone deficiency using an off-the-shelf bilobed uncemented porous-coated acetabular component of one design. There has been only one revision surgery to date, and clinical results were satisfactory in most cases.
This series shows that an oblong uncemented acetabular component can provide several advantages when used in the reconstruction of a large superior segmental acetabular bone deficiency. Large superior segmental bone deficiencies were reconstructed in all patients without the need for bulk structural acetabular bone grafts, and in all patients, the hip center was brought to a more normal position. These implants gained satisfactory acetabular fixation in most patients; only one implant has been rerevised for loosening. The single rerevision occurred when the acetabular bed consisted of greater than 50% structural allograft placed at a previous revision.
The most important limitation of this study is the relatively short-term followup. The findings show the short-term efficacy of the procedure with respect to clinical results and implant fixation, but longer followup is needed before the durability of this reconstructive technique can be assessed. In addition to the one patient who had revision surgery, two patients had early implant migration (after which the implants have been stable) and two had bead shedding. Although none of the procedures in these four patients currently are clinical failures (none of the patients have pain or gross implant instability), the longer-term fixation and clinical function of the four sockets must be viewed with concern. Many components in this series had partial radiolucent lines at the prosthesis-bone interface. Similar lines have been recognized as common after uncemented acetabular revision, but these sockets require ongoing observation.
Several other reports on the use of uncemented oblong implants for acetabular bone deficiencies have been published.2,3,10 The series are not strictly comparable to the current report because they report on primary and revision hip arthroplasties, although in each case, most of the cases were revision arthroplasties. Using the same implant evaluated in the current study, DeBoer and Christie2 reported on 18 patients (15 revision surgeries) followed up 4 to 6.9 years (mean, 4.5 years) after surgery. They reported excellent clinical results, no cases of implant migration, and no revision operations. Köster et al10 reported on 102 hips (98 revisions) treated with the LOR uncemented oblong socket. This implant differs from the Standard Range of Motion oblong implant: the socket is oblong but not bilobed and has a rough but not porous Ti surface. The LOR socket is fixed with screws, and the polyethylene liner is oblong and fits into the thin metal shell. Of the 102 hips, only two required reoperation for aseptic loosening. Eight other hips had migration, but in all of these hips the migration was 5 mm or less, and none had failed clinically.
Oblong bilobed sockets were developed to manage large superior acetabular bone deficiencies.3,18 Numerous other techniques are available to manage these defects. A superior bulk bone graft may be used in conjunction with a hemispheric cemented or uncemented socket implanted in anatomic or near anatomic position. Although such bone grafts provide the potential for bone stock restoration, reliance on bulk acetabular bone grafts for structural support of a large percentage of the socket may lead to problems,7,9 whether a cemented or uncemented implant is used. Jasty and Harris9 described a high rate of acetabular bone graft collapse and socket loosening when much of a cemented socket is supported by bone grafts. Pollock and Whiteside,14 Patch and Lewallen,13 and Paprosky et al12 identified a high rate of acetabular component loosening when uncemented sockets are supported largely by bone grafts, although it is recognized that structural bone grafts can be combined successfully with uncemented sockets when sufficient contact of the socket with native bone is present.
Large superior segmental deficiencies can be managed by placement of a cemented or uncemented socket in a superior position, a high hip center. An advantage of using a high hip center is that structural bone grafting usually is not necessary, and the socket can be placed mostly on native bone. Schutzer and Harris15, using an uncemented porous socket fixed with screws, reported excellent results with respect to acetabular fixation. However, placement of the socket in a superior position also has some practical disadvantages: leg length may not be restored; hip instability may be a problem because of femoral impingement against the pelvis; and most importantly, placement of a high hip center necessarily places constraints on the type of femoral reconstruction required.
Finally, the acetabulum may be reamed from an oblong shape to a hemispheric shape with subsequent insertion of an uncemented porous coated hemispheric component, usually of an extra large size.5,8,11 However, when the dimensions of the deficiency are much greater in an inferior to superior, rather than an anterior to posterior direction, insertion of a hemispheric component large enough to fill the deficiency requires reaming of the anterior and posterior columns with potential compromise of bone stock. In specific circumstances, oblong sockets have several advantages compared with the techniques described. By filling superior bone deficiencies with metal, rather than bone graft, the potential problems associated with structural graft collapse and socket failure on that basis are avoided. Oblong cups place a large porous metal surface in contact with native bone and thereby theoretically optimize the likelihood of stable biologic fixation. By providing acetabular reconstruction with the hip center in a more normal position, the practical problems associated with a high hip center are avoided.
Oblong components have some theoretical and practical drawbacks. Because superior bone deficiencies are filled with metal, rather than bone graft, bone stock restoration is minimal, and in some circumstances bone can be sacrificed to make the bone fit the implant. The bilobed components used in the current study fit only a few of acetabular defects, and it can be difficult to predict from preoperative templating whether a given bone deficiency will be ideal for a bilobed cup in advance of surgery. The implant is most useful for large superolateral segmental defects and is not suited as well for superior medial bone deficiencies because, in such circumstances, the superior lobe of the implant does not obtain good native bone contact without placement of the device in an excessively vertical orientation.
Despite the fact that oblong components obviated the need for structural bone grafts in these revision arthroplasties, they are more difficult to insert than are porous hemispheric implants. Double reaming of the pelvis to accommodate the two lobes must be performed, and the complex geometry typically does not provide the secure initial press fit customarily associated with uncemented hemispheric components. In addition, the thick metal of the shell limits the surgeon's ability to orient screws, thereby making good screw purchase more difficult to obtain. To position the socket properly the surgeon must be aware of the implant geometry and use the face of the lower lobe of the implant to determine implant orientation. Finally, careful preoperative planning is needed before an oblong cup is inserted to optimize leg length and implant sizing.
Complications in the current series occurred infrequently. However, a complication of particular concern was partial palsy of the peroneal branch of the sciatic nerve, which occurred in three patients. In none of the three patients was the etiology of the palsy certain, but lowering of the hip center with oblong cups and the associated potential stretching of the sciatic nerve is a possible source. Efforts to avoid excessive tension on the sciatic nerve when the hip center is normalized may be warranted.
For selected patients with large superolateral acetabular bone deficiencies, a bilobed, oblong porous-coated uncemented acetabular component fixed with screws facilitated a complex reconstruction and allowed the hip center to be placed in a more anatomic position. This device has provided satisfactory stability and good clinical results at short-term followup of 2 to 5 years in most patients, but longer followup is needed to determine whether fixation and results remain satisfactory with time.
1. D'Antonio JA, Capello WN, Borden LS, et al: Classification and management of acetabular abnormalities in total hip arthroplasty. Clin Orthop 243:126-137, 1989.
2. DeBoer DK, Christie MJ: Reconstruction of the deficient acetabulum with an oblong prosthesis. Three- to seven-year results. J Arthroplasty 13:674-680, 1998.
3. DeBoer DK, Christie MJ: Revision of the Acetabular Component: Oblong Cup. In Callaghan JJ, Rosenberg AG, Rubash HE (eds). The Adult Hip. Philadelphia, Lippincott-Raven Publishers 1461-1468, 1998.
4. DeLee JG, Charnley J: Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop 121:20-32, 1976.
5. Emerson Jr RH, Head WC: Dealing with the deficient acetabulum in revision hip arthroplasty: The importance of implant migration and use of the jumbo cup. Sermin Arthroplasty 4:2-8, 1993.
6. Harris WH: Traumatic arthritis of the hip after dislocation and acetabular fractures: Treatment by mold arthroplasty. J Bone Joint Surg 51A:737-755, 1969.
7. Harris WH: Bulk versus morselized bone graft in acetabular revision total hip replacement. Semin Arthroplasty 4:68-71, 1993.
8. Hozack WJ: Techniques of acetabular reconstruction. Semin Arthroplasty 4:72-79, 1993.
9. Jasty M, Harris WH: Salvage total hip reconstruction in patients with major acetabular bone deficiency using structural femoral head allografts. J Bone Joint Surg 72B:63-67, 1990.
10. Köster G, Willert H-G, Köhler H-P, Döpkens K: An oblong revision cup for large acetabular defects. Design rationale and two- to seven-year follow-up. J Arthroplasty 13:559-569, 1998.
11. Padgett DE, Kull L, Rosenberg A, Sumner DR, Galante JO: Revision of the acetabular component without cement after total hip arthroplasty. J Bone Joint Surg 75A:663-673, 1993.
12. Paprosky WG, Perona PG, Lawrence JM: Acetabular defect classification and surgical reconstruction in revision arthroplasty: A 6 year follow-up evaluation. J Arthroplasty 9:33-44, 1994.
13. Patch DA, Lewallen DG: Reconstruction of a deficient acetabula using bone graft and a fixed porous ingrowth cup: A 5 year roentgenographic study. Orthop Trans 17:151, 1993.
14. Pollock FH, Whiteside LA: The fate of massive allografts in total hip acetabular revision surgery. J Arthroplasty 7:271-276, 1992.
15. Schutzer SF, Harris WH: High placement of porous-coated acetabular components in complex total hip arthroplasty. J Arthroplasty 9:359-367, 1994.
16. Silverton CD, Rosenberg AG, Sheinkop MB, Kull LR, Galante JO: Revision total hip arthroplasty using a cementless acetabular component. Clin Orthop 219:201-208, 1995.
17. Silverton CD, Rosenberg AG, Sheinkop MB, Kull LR, Galante JO: Revision of the acetabular component without cement after total hip arthroplasty. A follow-up note regarding results at seven to eleven years. J Bone Joint Surg 78A:1366-1370, 1996.
18. Sutherland CJ: Treatment of type III acetabular deficiencies in revision total hip arthroplasty without structural bone graft. J Arthroplasty 11:91-98, 1996.
19. Tanzer M, Drucker D, Jasty M, McDonald M, Harris WH: Revision of the acetabular component with an uncemented Harris-Galante porous-coated prosthesis. J Bone Joint Surg 74A:987-994, 1992.