Demographics
This series consisted of all thirty-eight patients who had acetabular revision with insertion of a bilobed acetabular component without cement between December 1991 and December 1995. The thirty-eight patients had a total of forty-one revisions, done at one institution by three experienced hip surgeons, and they represent a small subgroup of the 414 hips treated with acetabular revision during the same period of time. The decision to use the bilobed component was made at the time of the operation. There were twenty-four women (63 percent) and fourteen men (37 percent), whose mean age at the time of the operation was 61.7 years (range, thirty-four to eighty-four years). The average preoperative weight was 73.0 kilograms (range, 51.7 to 100.7 kilograms). The patients had had an average of 2.5 previous hip operations (range, one to eight operations) on the ipsilateral side. Nineteen (50 percent) of the thirty-eight patients had a total hip prosthesis in the contralateral hip at the time of the index procedure. At the time of the initial arthroplasty, the diagnosis had been degenerative osteoarthritis in nineteen hips (46 percent), avascular necrosis in thirteen (32 percent), congenital dysplasia in four (10 percent), posttraumatic osteoarthritis in three (7 percent), rheumatoid arthritis in one (2 percent), and a displaced fracture of the femoral neck in one (2 percent). The main indication for a revision operation was aseptic loosening in thirty-three hips (80 percent). Other indications included reimplantation after resection arthroplasty in three hips, failed bipolar hemiarthroplasty in three, and failed resurfacing arthroplasty in two. Of the forty-one hips, twenty-one had only the acetabular component revised and twenty had both the acetabular and the femoral component revised. The average blood loss was 1420 milliliters (range, 400 to 4300 milliliters).
Operative Technique
The patient was placed in lateral decubitus, and a standard posterolateral approach was used. A trochanteric osteotomy was performed in fifteen hips, including six in which the procedure was done through the site of a nonunion that developed after a previous trochanteric osteotomy. Once the failed acetabular component, the cement, and the membrane were removed, the bone defect was evaluated. The decision to use the bilobed cup (Joint Medical Products, Stamford, Connecticut) was made intraoperatively on the basis of an absence of superior acetabular bone and an inability to use a porous hemispherical component. More specifically, each acetabulum had an oblong bone defect with a much greater superior-inferior dimension than anteroposterior dimension. A bilobed component was chosen because less than 50 percent of a hemispherical component inserted without cement would have been in contact with host bone.
The bilobed cup system had a special oblong reamer that was difficult to control. Instead of that reamer, conventional reamers with increasing diameters were used inferiorly until healthy bone was encountered. The diameter of the final reamer determined the superior-inferior dimension of the component. Reaming was then performed superiorly to the height of the bone defect; in some cases, overhanging lateral bone was removed. A trial component then was placed in the defect. Usually, the defect was larger than the trial component. In this situation, the inferior portion of the defect was enlarged until the anteroposterior dimension of the pelvis was maximized without disrupting the integrity of the anterior or the posterior column. Reaming to a larger diameter allowed the use of a larger component and improved the ability of the component to bridge the entire acetabular defect. If the component was still not large enough to bridge the oblong bone defect, it was positioned within the defect to allow maximum contact with host bone. In ten hips (24 percent), particulate allograft was used to fill the remaining small cavitary defects in the hemispheres of the bilobed component. No structural or load-bearing graft was used. A so-called press-fit was sought by oversizing the component by one to four millimeters in diameter on the basis of the surgeons' judgment of the quality of the bone. Despite this goal, none of the components were judged to have adequate press-fit stability without augmentation with screws. An average of 4.3 screws (range, two to seven screws) were used for fixation. Screws were placed only when a finger-tight purchase could be obtained during insertion. Usually, the best screw fixation was obtained superiorly, as bone often was insufficient inferiorly for screw fixation.
Postoperatively, every patient was managed with protected weight-bearing with bilateral use of axillary crutches or a walker for a minimum of three months. The duration of protected weight-bearing was determined on the basis of the intraoperative stability of the cup, the hip-joint stability, and the status of the femoral component or the greater trochanter.
Radiographic Assessment
Standardized radiographs were made before the operation, immediately postoperatively, and at follow-up examinations at three months, at six months, and annually thereafter. These radiographs included a standard anteroposterior radiograph of the pelvis, Judet oblique radiographs of the acetabulum, and true anteroposterior and Lowenstein lateral radiographs of the femur. All measurements were made without correction for magnification.
Preoperative Evaluation and Classification of the Acetabular Defect
Preoperatively, the height of the hip center at the site of the failed arthroplasty and that of the contralateral, uninvolved hip were measured. This measurement was defined as the vertical distance between the center of the femoral head and a horizontal line through the inferior margin of the teardrop or the superior border of the obturator foramen. A preoperative hip center could not be measured in the three failed hips that had had a resection arthroplasty. The average height of the hip center was forty-seven millimeters (range, twenty-five to seventy-seven millimeters) in the revised hips and fifteen millimeters (range, nine to twenty-two millimeters) in the contralateral, uninvolved hips. In addition, we measured the horizontal location of the hip centers, defined as the distance from the center of the femoral head to a line perpendicular to the horizontal line at the inferior point of the teardrop or at the medial aspect of the obturator foramen18,26. The average lateral displacement of the hip center was thirty-eight millimeters (range, thirteen to fifty-three millimeters) in the revised hips and thirty-seven millimeters (range, twenty-five to forty-nine millimeters) in the contralateral hips. Three other radiographic parameters also were assessed: (1) the continuity of Kohler's line (the ilioischial line), (2) the visibility of the teardrop, and (3) the presence or absence of osteolysis of the ischium.
Acetabular deficiencies were classified on preoperative radiographs, according to the classification system described by Paprosky et al.14,15, which is based on the presence or absence of an intact acetabular rim and its ability to support an acetabular component. Type 1 indicates an acetabular defect with a completely supportive rim and no osteolysis or migration of the component. A type-2 acetabulum has a distorted hemisphere with intact and supportive anterior and posterior columns. Radiographs show less than two centimeters of migration of the component with minimum osteolysis of the ischium and teardrop. The type-2 category is subdivided into three subcategories. A type-2A defect involves direct superior migration of the component (cavitation of the dome with an intact superior rim); a type-2B defect, superolateral migration of the component (an absent superior rim); and a type-2C defect, medial migration of the cup with destruction of the medial wall, seen radiographically as disruption of Kohler's line. A type-3 acetabular defect indicates severe bone loss with major destruction of the acetabular rim and supporting structure. Radiographs of this type of defect show superior migration of more than two centimeters with or without ischial and medial osteolysis. It was not possible to measure the superior migration of the acetabular component in ten hips because the serial radiographs of the failed component were not available. As the normal height of the hip center has been shown to be 14 ± 2 millimeters (range, ten to twenty millimeters) proximal to the interteardrop line8, a failed hip in which the height of the hip center was more than forty millimeters was also classified as type 3. According to the classification system of Paprosky et al.14,15, type-3 defects were subdivided into two subcategories: type 3A, which indicates that Kohler's line is still intact, and type 3B, which indicates that Kohler's line is no longer intact. Typically, type-3B defects also demonstrate complete destruction of the teardrop and extensive osteolysis of the ischium. However, several of the type-3B hips in our series had disruption of Kohler's line but an intact or only slightly damaged teardrop and ischium. To address this finding, we subdivided the type-3B subcategory of the classification system of Paprosky et al. into type-3B and type-3C defects. Thus, the classification system that we used included type 3A (an intact Kohler's line), type 3B (a disrupted Kohler's line with an intact or only slightly damaged teardrop and ischium), and type 3C (a disrupted Kohler's line with complete obliteration of the teardrop or severe osteolysis of the ischium). In our series of forty-one revisions, there were four type-2A, four type-2B, four type-2C, thirteen type-3A, nine type-3B, and seven type-3C defects.
Postoperative Evaluation
Postoperatively, the height of the hip center (superior migration), the horizontal location (horizontal migration), and the abduction angle were used to determine the stability of the acetabular implant. The abduction angle was measured from the superior to the inferior aspect of the component and, thus, included both lobes of the component. The accuracy of this method of measuring the migration of the acetabular component has been reported to be ±2.0 millimeters24. Russotti and Harris18 reported that a change from 0 to 10 degrees of flexion of the pelvis caused a two-millimeter increase in the apparent measurement of the height of the hip center. Therefore, we defined definite migration of the cup as more than 4 degrees of change in the abduction angle or more than four millimeters of medial or superior movement of the hip center18,26,27. In addition, we noted periacetabular radiolucent areas, radiolucent areas adjacent to supplementary screws, breakage or backout of screws, separation of beads from the metallic shell, and the integrity of the greater trochanter. We used the distance between the inferior edge of the oblong cup and the interteardrop line (a positive measurement indicated that the cup was distal to the line, and a negative measurement indicated that the cup was proximal to the line) to determine whether the component had adequate contact with host bone at the distal portion of the acetabulum (Fig. 1).
We categorized the stability of the cups as (1) stable, which indicated no definite migration, no continuous radiolucent line wider than two millimeters at the bone-cup interface, no radiolucency adjacent to screws, and no breakage or backout of screws; (2) probably unstable, which indicated no definite migration but breakage or migration of screws or a continuous periacetabular radiolucent line wider than two millimeters; or (3) unstable, which indicated definite migration (more than 4 degrees of change in the abduction angle or more than four millimeters of medial or superior movement of the cup). For the purposes of this study, we considered radiographically unstable hips to be failures.
Statistical analysis of the results was performed with the Wilcoxon signed-rank test, Fisher's exact test, or the two-tailed unpaired Student t test.
Clinical Evaluation
Pain, walking, limp, use of walking aids, functional level, and limb-length discrepancy were assessed preoperatively and at the latest follow-up evaluation. Pain was graded as absent, mild (no effect on average activity), moderate (affecting some activities), and severe (limiting most activities). The rating for pain before the revision was severe for twenty-five hips (61 percent), moderate for eight (20 percent), mild for three (7 percent), and absent for five (12 percent). Preoperatively, seven patients (18 percent) did not need assistive devices to walk, four patients (11 percent) used a cane on long walks, and twenty-seven patients (71 percent) needed full-time support. Twenty-one patients (55 percent) could walk without support indoors only, eight (21 percent) could walk two or three blocks, five (13 percent) could walk four to six blocks, and four (11 percent) could walk an unlimited distance. On physical examination, seventeen patients (45 percent) had a severe limp; thirteen (34 percent), a moderate limp; six (16 percent), a mild limp; and two (5 percent), no limp. The average preoperative limb-length discrepancy was two centimeters shorter on the involved side (range, zero to five centimeters).
Additionally, every patient filled out a questionnaire at the time of the follow-up examination. The subjective outcome was defined by the response to the following three questions. Has the operation increased function and daily activity? Has the operation decreased hip pain? Are you satisfied with the result of the operation?
Of the thirty-eight patients (forty-one hips), one was lost to follow-up and one died during the study period. Two patients who could not return for radiographic evaluation completed questionnaires. The remaining group of thirty-four patients (thirty-seven hips) had four type-2A, four type-2B, three type-2C, twelve type-3A, seven type-3B, and seven type-3C defects, and they were followed both clinically and radiographically for a minimum of two years. The average duration of follow-up was forty-one months (range, twenty-four to sixty-six months).
Radiographic Findings
At the immediate postoperative evaluation, the vertical position of the hip center had improved in all patients. The average height of the hip center was twenty-three millimeters (range, five to forty-four millimeters). The hip center was lowered an average of twenty-three millimeters (range, six to fifty millimeters) compared with the preoperative position. The average horizontal location or lateral displacement of the hip center was thirty-one millimeters (range, twenty-five to forty millimeters), and the location was an average of seven millimeters medial to the preoperative position. The inferior edge of the oblong cup was located one centimeter or more distal to the interteardrop line in nine hips, less than one centimeter distal to the line in nine hips, on the line in four hips, less than one centimeter proximal to the line in ten hips, and one centimeter or more proximal to the line in five hips. (The average location was 1.6 millimeters distal to the line, and the range was twenty-three millimeters distal to the line to twenty-four millimeters proximal to it.) The angle of abduction of the oblong acetabular component averaged 54 degrees (range, 31 to 74 degrees).
At the latest follow-up evaluation, seven cups (19 percent) showed separation of the beads from the metallic shell, seven (19 percent) had backout or breakage of supplementary screws, and two (5 percent) had radiolucency adjacent to the screws. Incomplete radiolucent lines of one millimeter or less in width appeared at the acetabular bone-prosthesis interface in all but six hips. Six hips (16 percent) had a continuous periacetabular radiolucent line wider than two millimeters.
Twenty-eight acetabular components (76 percent) were stable (Figs. 2-A and 2-B). Three components (8 percent) had changes in screw position but no definite migration, and they were classified as probably unstable. All three components were in hips with a type-3B acetabular bone defect.
Six hips (16 percent) had radiographic evidence of migration (Figs. 3-A, 3-B, and 3-C). In one of these hips, the component tilted to a more vertical position. In the other five, it tilted to a more horizontal position. In addition, all six unstable sockets had breakage or backout of the supplementary screws and a continuous periacetabular radiolucent line wider than two millimeters. Five of the six demonstrated separation of beads from the metallic shell. The average time to radiographic evidence of failure was thirty-four months (range, ten to sixty months). All six unstable components were in hips with a type-3 defect: four of the hips had a type-3C defect, one had a type-3B defect, and one had a type-3A defect.
Two of the six hips with an unstable component needed a revision with an antiprotrusio cage. These were complex repeat revisions with extensive bone loss. The original acetabular bone defect and the loosening of the bilobed component contributed to the complexity of these cases. The bone removed at the time of the initial reaming for the bilobed component was not a contributing factor. The remaining four patients who had a loose bilobed component were not symptomatic enough to warrant a reoperation.
Clinical Findings
A clinical evaluation was performed at the time of the latest follow-up for all of the patients except the two who had a repeat revision, who were evaluated just before the reoperation.
Pain
Twelve (31 percent) of the thirty-nine hips were pain-free; fifteen hips (38 percent) were mildly painful; seven (18 percent) were moderately painful, causing limitation of some activities; and five (13 percent) were severely painful, limiting most activities. Overall, pain was less severe than it had been preoperatively in thirty-two (82 percent) of the thirty-nine hips (p < 0.001, Wilcoxon signed-rank test). Two patients (6 percent) who had a loose acetabular component had worse pain than before the operation.
Walking
Fifteen patients (42 percent) did not need an assistive device, eleven (31 percent) needed a cane for long walks, and ten (28 percent) still needed full-time support. Five of the ten patients who needed full-time support had a loose acetabular component. Overall, twenty patients (56 percent) had less need for walking aids postoperatively than they had had preoperatively (p = 0.004, Wilcoxon signed-rank test).
Fifteen patients (42 percent) could walk an unlimited distance, ten (28 percent) could walk four to six blocks, four (11 percent) could walk two or three blocks, and seven (19 percent) could walk without support indoors only. Four of the seven who only walked indoors had a loose acetabular component. Overall, twenty-five patients (69 percent) could walk farther postoperatively than they had been able to walk preoperatively (p < 0.001, Wilcoxon signed-rank test). Three patients (8 percent) had a decreased ability to walk. Two of these three patients had a loose acetabular component.
Gait
We evaluated the gait of each of the thirty-four patients (thirty-seven hips) who returned to the clinic for an examination. Eleven patients (32 percent) had no limp, fifteen (44 percent) had a mild limp, six (18 percent) had a moderate limp, and two (6 percent) had a severe limp. Of the six patients who had a loose acetabular component, four had a moderate or severe limp. Twenty-four patients (71 percent) had a decrease in the limp compared with the preoperative status (p < 0.001, Wilcoxon signed-rank test). Two patients (6 percent) had postoperative worsening of the limp; both had a loose cup.
Limb-Length Discrepancy
The limb lengths were evaluated in each of the thirty-four patients (thirty-seven hips) who returned to the clinic, and they were equal in seventeen patients (50 percent). Eleven patients (32 percent) had a discrepancy of one centimeter or less, five patients (15 percent) had a difference of more than one to two centimeters, and one (3 percent) had a discrepancy of more than two centimeters (3.5 centimeters). The average postoperative discrepancy was 0.5 centimeter shorter on the side of the revision, with an average improvement of 1.5 centimeters (range, zero to five centimeters) compared with the preoperative status (p < 0.001, Wilcoxon signed-rank test).
Overall Satisfaction
According to the questionnaires that were completed by thirty-six patients, twenty-nine (81 percent) considered the function of the hip and the ability to perform daily activities to be improved and twenty-eight (78 percent) were satisfied with the result of the operation. Five of the six patients who had a loose acetabular component were not satisfied.
Statistical Findings
All six hips that failed were in patients who had a type-3 acetabular defect; the defects included one type-3A, one type-3B, and four type-3C defects. The rate of failure for the hips that had a type-3C acetabular defect preoperatively (four of seven hips) was significantly increased compared with that for all of the other hips (two of thirty hips) (p = 0.007, Fisher's exact test).
The average postoperative distance (and standard deviation) between the inferior edge of the cup and the horizontal line was -8.8 ± 11.3 millimeters in the hips that had a failed cup and 3.3 ± 10.3 millimeters in the hips that did not have a failed cup (p = 0.02, two-tailed unpaired Student t test). The location of the inferior edge of the cup was proximal to the horizontal line in five of the six failed hips, whereas it was proximal to the horizontal line in only fourteen (45 percent) of the thirty-one hips that did not fail. Although the finding was not found to be significant (p = 0.2, Fisher's exact test), the relative risk for failure was found to be 4.7 times greater (95 percent confidence interval, 0.6 to 36.7) when the inferior edge of the cup was proximal to the interteardrop line than when it was distal to the line.
The average postoperative height of the hip center was 30.4 millimeters (range, fourteen to forty-four millimeters) in the hips that failed and 21.0 millimeters (range, five to forty-three millimeters) in the hips that did not fail; the difference was significant (p = 0.04, two-tailed unpaired Student t test). However, the average postoperative lateral displacement was not found to differ significantly between the hips that failed (33.0 millimeters) and those that did not fail (32.6 millimeters) (p = 0.8, two-tailed unpaired Student t test). The average angle of abduction was 51 degrees for the hips that failed and 55 degrees for the hips that did not fail. Again, the difference was not found to be significant (p = 0.4, two-tailed unpaired Student t test). With the numbers available, we could not detect a significant difference between the hips that failed and those that did not fail with respect to the age or body weight of the patient at the time of the index revision operation or the number of screws used for stabilization (p > 0.09, two-tailed unpaired Student t test).
Complications
Complications in the forty-one hips included four dislocations, three trochanteric nonunions, two intraoperative femoral fractures, and one early superficial infection. Including the two revised cups, six hips (15 percent) in six patients needed a reoperation.
Of the four patients who had a dislocation of the hip, two had no additional dislocations after an initial closed reduction and two had recurrent dislocations. One patient subsequently had an exchange of the polyethylene liner, and one had a reattachment of the trochanter secondary to trochanteric migration.
Of the three trochanteric nonunions, two were associated with minimum displacement (less than one centimeter), which did not need treatment. The third trochanter had two centimeters of displacement and was associated with recurrent dislocation, resulting in the trochanteric reattachment that was already mentioned.
The patient who had the superficial wound infection was successfully managed with multiple débridements and six weeks of parenteral antibiotic therapy. The two intraoperative femoral fractures around the tip of the stem were treated with open reduction and internal fixation at the time of the index revision. One of these fractures, however, did not unite, necessitating treatment with additional autogenous bone graft and electrical stimulation.
The results of acetabular reconstruction with use of a bilobed socket have not been widely reported. We know of four recent reports on the use of an oblong-shaped cup inserted without cement for acetabular revision3,12,19,23, but those reports included small numbers of hips and involved both custom and off-the-shelf components. We are aware of only two reports involving the same acetabular component as the one used in our series. Sutherland23 reported disappointing results in six hips that had a severe acetabular bone defect and were revised with a bilobed custom component (four hips) or an off-the-shelf bilobed component (two hips). Three of the six sockets in his series had failed at the time of follow-up, which was a minimum of four years. The failed implants were custom devices, whereas the two off-the-shelf bilobed components were stable. In the other series3, an off-the-shelf bilobed component was used for fifteen revisions; there was no loosening of the component at an average of 4.5 years. Those authors reported excellent results, and they were able to relocate the hip center to a normal level with the bilobed acetabular component. They also noted the technical difficulty associated with implanting a bilobed component. In our study, the 24 percent rate of failure (nine of thirty-seven hips), which was due to a probably or definitely loose component, and the 5 percent rate of revision (two of thirty-seven hips) were inferior to the results achieved in the other studies. Without analysis, it was difficult to surmise whether our higher rate of failure was related to the design of the component, the operative technique, or the parameters that had been used for the selection of patients.
We analyzed the technical factors that could have contributed to our high rate of failure. Difficulty in reaming the acetabular defects is an aspect of the technique that is specific to the instrumentation. The bilobed reamer was awkward and difficult to use on the sclerotic acetabular bone remaining after removal of the failed implant. Other difficulties, such as placement of the screws, are common to revisions of all acetabula with large defects. In these hips, the sclerotic superior acetabular bone held the screws well, but the bone of the inferior aspect of the acetabulum often was absent or too weak to hold the screws. The complexity of these procedures was evidenced by the fact that 37 percent (fifteen) of the forty-one hips needed a trochanteric osteotomy. Although these are important technical aspects, it is difficult to ascertain how they affected the rate of loosening.
The size and placement of a bilobed component, however, is a technical aspect that can be analyzed. We found that the relationship between the postoperative position of the bilobed component and the radiographic teardrop influenced the outcome. In our series, five of the six unstable components initially had not extended to the level of the radiographic teardrop or distal to it. Furthermore, the average distance between the inferior edge of the cup and the interteardrop line was significantly greater in the hips that failed than in the hips that did not fail (p = 0.02, two-tailed unpaired Student t test). It appears that, if a bilobed component does not extend to the level of the interteardrop line or distal to it, the component is undersized and is in danger of failing because of inadequate fixation to the inferior aspect of the acetabulum. The size of the component is especially important in hips with a defect of the medial wall. In these hips, a small component has neither medial nor distal acetabular support and it migrates into the pelvis.
DeBoer and Christie3 suggested that this implant should not be used in patients who have a major anterior or posterior column defect. To evaluate the possibility that the failures in our study resulted from improper selection of patients, we grouped the patients retrospectively with use of an acetabular bone-stock classification system. Although the reproducibility of this classification system was not verified, it is based on easily identifiable radiographic landmarks.
With use of this classification system, we determined that the most important predictor of outcome was the amount of superior migration of the component and disruption of Kohler's line. No patient who had had less than two centimeters of superior migration of the component (a type-2 acetabular defect) before the revision had loosening of the bilobed component. Only one of the twelve hips that had had more than two centimeters of migration of the component and an intact medial acetabular wall (a type-3A defect) before the revision had instability of the bilobed component. Conversely, eight of the fourteen hips with preoperative migration of the component of more than two centimeters and disruption of Kohler's line (a type-3B or type-3C defect) had a loose or probably unstable bilobed component. On closer examination of these hips, we noted an especially high rate of failure in the type-3 group with a defect of the medial acetabular wall and ischial osteolysis or obliteration of the radiographic teardrop (a type-3C defect). Four of the seven implants inserted into a hip with this pattern of bone loss migrated. Therefore, it seems that preoperative superior migration of the component of more than two centimeters combined with a defect of the medial wall is a contraindication to the use of a bilobed component. On the basis of this analysis, only hips with a type-2 or type-3A defect could be treated with a bilobed component.
Currently, the best results of acetabular revision that we are aware of have been achieved with components inserted without cement and fixed with supplemental screws. After seven to eleven years of follow-up, Silverton et al.22 reported that none of the 115 hips that were available for review in their series had had aseptic loosening. They concluded that use of a hemispherical porous-coated component combined with an operative technique that maximizes host-bone contact, attains peripheral rim-fit fixation, and uses either autogenous graft or allograft in contained defects was successful. In view of this finding, it would seem prudent to consider these components as the first option for reconstruction of type-2 and type-3A defects.
However, in our experience, the inferior-superior dimension of some type-2 and type-3A defects has exceeded the anteroposterior dimension of the pelvis. In these instances, converting the oblong-shaped defect to a hemispherical cavity compromises the anterior column of the acetabulum or the posterior column of the acetabulum, or both3. If a hemispherical component is used and the acetabular columns are maintained in these hips, it is difficult or impossible to restore the anatomical hip center. Because the bilobed acetabular component can lower and restore the anatomical hip center, it presents an opportunity to restore the normal mechanics of some hips with type-2 and type-3A defects. As alternatives to the bilobed cup, structural allografts and antiprotrusio cages can be used to achieve the same objectives.
In conclusion, insertion of a bilobed component is a technically demanding procedure that should not be used in patients who have extensive bone loss in the superior aspect of the acetabulum combined with a defect of the medial wall (a type-3B or type-3C defect).
A bilobed component may be considered for use in hips that have mild-to-moderate bone loss in the superior aspect of the acetabulum (a type-2 defect) or an intact medial acetabular wall with severe bone loss in the superior aspect of the acetabulum (a type-3A defect), especially if the surgeon wants to correct an elevated hip center.
If the bilobed component is used, it should be large enough so that the inferior aspect of the cup extends to the level of the radiographic teardrop, allowing secure fixation to healthy bone in the inferior portion of the acetabulum.