➢ In an acetabulum with an oblong defect, creation of a supportive hemisphere of bone is critical. If during reaming this cannot be achieved without the loss of supportive host bone, then a porous metal augment device should be considered.
➢ In general, acetabular augment devices are used in Paprosky IIIA and IIIB defects.
➢ The combined reports to date on the use of augment devices in acetabular revision surgery have shown a mean survivorship of 98% (range, 91% to 100%) at a mean follow-up of forty-five months (range, eighteen to seventy-four months).
➢ Pelvic discontinuities remain difficult to manage even with the use of augment devices.
As the number of total hip arthroplasties performed in the United States increases, and as more young patients undergo this procedure, it can be expected that the number of revision hip arthroplasties will also increase. Most acetabular revisions can be performed with a hemispherical shell. Results with these components show a rate of survivorship free of aseptic loosening of 97% (range, 94% to 100%) in 729 revision surgeries1-9 at a mean follow-up time of 9.8 years (range, six to 21.3 years).
Use of a hemispherical shell requires the creation of a supportive, concave hemisphere of bone. While in the majority of acetabular revisions this goal is attainable, at times the nature of the acetabular osseous defect makes this goal unobtainable. This can be understood as the inability to ream a large enough hemisphere, with the goal of reaching the apex of the superior aspect of the deformity, without in the process removing supportive anterior and posterior host bone. In this setting, an oblong osseous deficiency is left over (defined as the superior to inferior length of the overall defect being greater than the anterior to posterior width of the remaining acetabular bone).
The development of highly porous metal technology has provided surgeons with an expanded range of implants to address these bone defects. One of the earliest highly porous materials to be utilized in hip reconstructive surgery was porous tantalum. This material has numerous properties that make it suitable for treating bone defects encountered in revision acetabular surgery. Compared with conventional porous metals, porous tantalum has an increased porosity that allows for improved bone ingrowth, and it has a structural stiffness similar to subchondral bone10. It also has a higher coefficient of friction as compared with that of other porous coatings, allowing for a higher initial interface strength that, when combined with improved bone ingrowth, has produced excellent short-term and midterm results11,12. Building on the initial short-term success with porous tantalum hemispherical acetabular components, porous tantalum augment devices were developed that, in theory, allow for structural stability when unitized with a hemispheric porous tantalum cup through cementation of the cup to the augment device13. Thus, an augment device unitized with a hemispherical shell expands the range of defects that porous metals can address; previously, treatment of such defects required the use of a structural allograft and a hemispherical cup, a reconstruction ring, or an antiprotrusio cage.
Porous metal augment devices are used primarily for these indications: replacement of deficient segmental acetabular bone (typically, loss of the superior lateral rim), filling of large cavitary defects where cementation of the augment to the porous metal shell will improve component stability (typically, anterior superior or medial defects), or for combined defects. This review will focus on three areas: (1) understanding the likely revision scenarios in which an augment device may need to be used, (2) describing general and specific operative techniques, and (3) evaluating the published results to date on the use of augment devices.
Classification of Acetabular Defects
The classification scheme of the American Academy of Orthopaedic Surgeons (AAOS) is geographic and divides osseous deficiencies by anatomic location and type14,15. The possible implant needs are not addressed in this system. The classification system reported by Paprosky et al. describes the osseous defect, quantifies superior acetabular component migration, and aids in identifying the anticipated implants needed for the reconstruction (Fig. 1)16. The critical difference among the Type-IIA, Type-IIB, and Type-III defects is the severity of superior migration of the femoral head as measured above a line transversing the superior obturators. To identify the superior obturator line, on an anteroposterior pelvic radiograph the most prominent point at the superior-lateral aspect of each obturator foramen is marked, and a line connecting these two points is drawn (Fig. 2). This point is used rather than the acetabular teardrop, as that osseous landmark may not be identifiable as a result of ischial osteolysis. If the existing femoral head center is <3 cm above this line, then by definition a Type-II defect has been identified; if ≥3 cm of migration is measured, a Type-III defect is present16,17.
Paprosky Type-I defects feature an intact acetabular rim and no osseous osteolysis or component migration. After removal of the current implant, reaming should leave an intact supportive osseous acetabular bed upon which a hemispherical shell can be placed. In a report on thirteen revisions performed for Type-I defects, the mean increase in reaming to obtain a stable press fit was 3 mm (range, 2 to 6 mm)18.
Paprosky Type-IIA and Type-IIB defects have superior component migration of <3 cm above the transverse superior obturator line. In a Type-IIA defect, the component migrates superiorly and medially (often anteriorly), causing damage to the acetabular dome but leaving an intact superior lateral rim. In a Type-IIB defect, the component migrates superiorly and laterally, damaging the superior rim. A Type-IIC defect features medial component migration, which leaves a protrusio defect. For the majority of these hips, a hemispherical acetabular component can be used.
In Type-III defects, the component migrates superiorly >3 cm above the transverse superior obturator line. This results in an unsupported superior rim and a large oblong defect. Greater ischial and medial osteolysis is seen, which leaves less supportive bone for the reconstruction. In the Type-IIIA defect, the columns are deficient but intact. In a Type-IIIB defect, a break in the Kohler line is seen, there is less supportive bone, and often a pelvic discontinuity is present. For these defects, implant options are a hemispherical acetabular component combined with either a porous metal augment(s) or structural allograft, reconstruction cages with or without structural allograft, a cup-cage construct, a cup-cup construct, or a custom triflange component. Caution must be exercised in the treatment of a Type-IIIB defect, as often the acetabular bone is unsupportive of a hemispherical shell and a pelvic discontinuity may exist.
Informed consent, in addition to discussion of the general risks of revision surgery, should focus on specific items applicable to the use of an augment device in revision acetabular surgery. The specific items to discuss include the risk of injury to the sciatic nerve, preoperative evaluation of limb lengths and postoperative expectations, the risk of dislocation, and both the expected treatment option (use of augment devices combined with a hemispherical shell) and the possible need for use of other techniques (e.g., cup-cage reconstruction, cup-cup reconstruction, or reconstruction cage).
The approach should be predicated on two conditions: the prior approach used, and the surgeon’s comfort. The use of a prior approach (e.g., anterolateral) is most reasonable when the dissection plane is easily identified (such as when the surgeon finds that the abductor repair has not completely healed). It is our opinion that taking the abductors off the trochanter a second time may compromise the subsequent repair. In this situation, a trochanteric osteotomy may be considered.
Important general concepts for preparation of the deficient acetabulum are (1) recognition of when the defect is oblong and (2) the ability to begin reaming in the anatomic position. The true inferior osseous acetabulum should be identified, which may be accomplished by removal of inferior scar tissue and placement of the curved end of a cobra retractor inferior to the osseous edge. During reaming, the surgeon should evaluate whether the acetabular defect is becoming a hemisphere (i.e., the superior rim is being contacted) while concomitantly not removing supportive anterior and posterior host bone. Obtaining osseous contact in two key anatomic areas of the acetabulum (anterior-superior and posterior-inferior) is important for implant stability. If the surgeon is concerned that the superior rim is not reachable without loss of supportive anterior-superior and/or posterior-inferior bone, then a high hip center may be chosen or, alternatively, an augment may be used. Adjunctive screw fixation (dome and/or peripheral screws) is recommended through the acetabular shell and, if possible, through the shell-augment construct. Generally at least two screws are placed superiorly in the so-called safe zone19.
Technique by Paprosky Classification
In a Type-IIB defect, secondary to superior and lateral migration of the failed acetabular component, a superior rim deficiency exists. The key decision to make is whether augmentation of the superior rim deficiency will be needed. To use a hemispherical shell alone for a Type-IIB defect, three conditions should be satisfied: (1) reaming to obtain both anterior-superior and posterior-superior host bone contact with creation of concave walls should not remove supportive lateral bone; (2) reaming should not elevate the joint center >1.5 cm over the anatomic center, as this can create a risk for osseous impingement and instability20; and (3) some concavity of the superior rim must be created with the reamer so that the trial implant will not tilt superiorly and laterally when pressure in an upward direction is applied to the insertion handle (or else the real implant may not be stable). If these conditions are met, then a hemispherical shell can be used.
If, due to the superior-inferior length of the oblong defect, enough host bone does not exist to allow reaming that will create a hemispherical concavity, then an augment device can be used. Once reaming in the anatomic position is finished, a trial shell is placed to assess host bone contact. This allows evaluation of the size in width, length, and depth of the defect to be augmented21. Prior to preparation for an augment device, a trial augment (often smaller than the final augment used) can be inserted above the trial shell for the purpose of determining augment size and positioning relative to the osseous defect. As a general rule, the outer diameter of the chosen augment should match the osseous defect without necessitating removal of a large amount of supportive bone. Matching of the existing osseous defect to the outer diameter of the augment may create a mismatch between the inner diameter of the chosen augment and the outer diameter of the hemispherical shell. We favor placing cement on the outer face of the augment prior to impacting the acetabular shell, both to fill voids due to the potential use of mismatched augment-shell diameters and to unitize the construct22.
If the augment system allows for preparation of the augment device with reciprocating rasps, the surgeon will prepare for the augment device with the acetabular trial component in place. Otherwise, the surgeon will use high-speed burrs to contour the bone for the augment21. Once the augment osseous bed is prepared, the porous metal augment (with the trial shell in place) is placed in the defect and secured with screws that are not yet fully tightened. The trial shell is then removed, any osseous defects are bone grafted, cement is applied to the face of the augment, and the new acetabular component is driven into place (cement unitizes the construct, and the asymmetric geometry of the cup-augment construct theoretically may be more resistant to migration). An assistant tightens the screws on the augment device, and then dome screws are placed through the shell. When possible, placement of a dome screw through both the acetabular shell and the augment is recommended. Where the augment interferes with drilling for dome screws, a hole may be created in the augment or the shell with a metal cutting burr21. An important caveat should be noted: intraoperatively, two highly porous roughened surfaces (acetabular shell and augment) placed against one another in an osseous defect may, because of the frictional interlock of the two surfaces, falsely give the appearance of being stable and capable of resisting motion12. For this reason, in addition to cementation of the augment to the shell, adjunctive screw fixation is recommended.
Type-IIIA defects involve compromise of the columns both in bone width and quality, with deficient but intact columns and with component migration >3 cm above the teardrop line. For a Type-IIIA defect, if anterior and posterior host bone is available and there is sufficient superior bone to allow reaming to be performed and a concave surface made of the remaining walls, either a hemispherical shell in an anatomic position combined with an augment, or a hemispherical shell in a high position, may be utilized. Given the large amount of deficient segmental bone, it may be necessary to place both a superior augment device and a second anterior or posterior augment device23. Medial cavitary defects may be filled with bone, or an augment may be placed. The techniques discussed above with regard to Type-IIB deficiency apply (Fig. 3, Fig. 4, and Fig. 5).
In these defects, if a concavity of bone capable of supporting a hemispherical shell with or without an augment cannot be created, traditionally two implant options have existed: an ilio-ischial reconstruction cage, or a structural bone graft with a reconstruction cage. If enough host bone exists to allow reaming and placement of a hemispherical acetabular component, yet this component when placed does not have osseous stability, the porous metal component (with or without augment devices) can be transfixed with multiple screws, after which a reconstruction cage can be placed over the acetabular component. This so-called cup-cage construct theoretically protects the cup, allowing osseointegration24. Recently, one of the authors (T.J.B.) and an associate reported on a modification of the cup-cage construct, which is used when a superiorly placed hemispherical porous metal acetabular component achieves stability in a healthy bone bed but the shell placement results in the joint center being either too superior, still protrusio, or both. In this setting, a second porous metal shell, placed in appropriate inclination and anteversion, is cemented into the first shell. Early results show no failures in eight hips, but the average follow-up time was less than two years25.
In Type-IIIB defects, greater columnar deficiencies are found, often with an associated pelvic discontinuity. The techniques described above with regard to the Type-IIIA defects may be used; in the Type-IIIB defect, the often-associated pelvic discontinuity must be addressed. For these extreme cases, use of a hemispherical shell and augments may not yield a stable reconstruction.
Isolated Acetabular Revisions—Points to Consider
Two points are critical to consider in the setting of an isolated acetabular revision. The first point is that, with retention of a well-fixed femoral component, the surgeon must plan the anticipated position of the new acetabular component so that its placement will allow the hip to be reduced. While the stems encountered in revision surgery are usually modular, thus allowing intraoperative flexibility in optimizing limb-length gain and soft-tissue stability, there are still revisions performed with retention of a nonmodular femoral component. In this setting, particularly when dealing with superior acetabular component migration, it is important to mobilize the femur to allow reduction into the new acetabular center. For this purpose, a Schnidt clamp may be placed superior to the lesser trochanter and used to pass the femur in an anteromedial to anterolateral direction, which will protect the iliopsoas tendon. The surgeon can incise with Bovie electrocautery any capsular or pericapsular tissue. A Cobb elevator is then used to strip down the anterior part of the femur if any remaining soft tissue is encountered. After removal of the failed acetabular component, the retained femur can be reduced in the osseous acetabulum and an assessment made as to whether or not the hip is reducible into the planned new component position. In a setting in which the retained prosthetic neck cannot be reduced into the proposed new acetabular position without the creation of excessive soft-tissue tightness, two options exist: acceptance of a higher hip center, or use of a trochanteric osteotomy of sufficient length such that the distal aspect of the osseous osteotomy may be removed, allowing distalization of the trochanter and effective femoral shortening.
The second point is that the anteversion of a well-fixed femoral component is not alterable; however, the acetabular component may be placed to optimize combined version. If a trial shell and liner can be placed, the hip can be gently reduced and tested for stability, and then the final acetabular component position can be determined. If the trial shell moves during range-of-motion testing, then, at a minimum, a gentle reduction of the hip can be performed to evaluate the proposed positioning of the acetabular component. The Ranawat sign is helpful in this regard26. With this technique, evaluation of the combined (femoral and acetabular) anteversion is performed by placing the operatively treated limb in extension and internally rotating the leg 45°. If the femoral head is parallel to the face of the acetabular component, this implies a combined anteversion of 45°, assuming that the pelvis is positioned on the operating table in neutral rotation.
In the peer-reviewed literature from 2004 to 2013 on the use of augment devices, it was reported in six papers that an augment device was used 15% of the time or less. We therefore restricted our review to papers reporting primarily on the treatment of either Type-II defects (one paper) or Type-III defects (eight papers). In these series, at least fifty percent of the reconstructions were performed with use of augment devices. Results from seven of the eight papers are summarized in Table I.
Fernández-Fairen et al. reported on 263 reconstructions, with 155 of these cases regarding the treatment of Paprosky Type-IIA or Type-IIB defects. A total of three augments were used, all in Type-IIB defects27.
Eight papers were with regard to the use of augments in Paprosky Type-IIIA or Type-IIIB defects. To our knowledge, Nehme et al. were the first to report on the use of trabecular metal augment devices and revision shells. Eleven of sixteen operations were performed for the treatment of Type-III deficiencies, and no cup loosening was seen at a mean follow-up of thirty-two months24. In seventeen Type-IIIA and six Type-IIIB defects treated with porous metal hemispherical shells and one or more augment devices, all reconstructions were stable at a mean follow-up of thirty-five-months (range, twenty-four to fifty months)28. In that series, eight preoperatively unsuspected pelvic discontinuities were identified. In a larger series of forty-three patients (thirty-three Type-IIIA defects and ten Type-IIIB defects) at a mean follow-up time of thirty-four months, only one failure was reported, and that failure was due to recurrent sepsis22. Sporer and Paprosky described a novel treatment for Type-IIIB chronic pelvic discontinuity; this treatment included placement of augments to provide a supportive rim and also filling of cavitary defects with particulate bone graft, followed by distraction of the discontinuity achieved during impaction of an oversized acetabular shell23. At mean follow-up of 2.6 years, no implants were loose. The use of fourteen buttress augments in fourteen of nineteen Type-IIIA or IIIB defects was described by Ballester Alfaro and Sueiro Fernández29. After acetabular preparation, a buttress augment device was fixed with screws to the ilium, after which cement was applied to the face of the augment device and the porous metal shell was impacted and fixed with screws. At a mean follow-up of just over two years, all implants were stable.
Three articles described the mean five-year follow-up results associated with porous metal reconstructions. In thirty-seven cases involving Paprosky Type-IIIA defects and in which augment devices were used in all hips, at a mean of sixty months of follow-up only one component had failed21. In the paper by Abolghasemian et al., in which the Gross classification system30 and at least one augment was used in each of thirty-four hips (eighteen minor column defects [Gross Type III], fourteen major column defects [Gross Type IV], and two pelvic discontinuities [Gross Type V]), there were three failures, two of which occurred in hips with pelvic discontinuity31. In a series of twenty-four hip reconstructions (involving fifteen Type-IIIA defects and nine Type-IIIB defects) in which a wedge-shaped augment was combined with impaction grafting and cementation of an all-polyethylene acetabular component, it was reported at a median follow-up time of five-years that one augment had fractured and five cups had migrated >5 mm32.
The literature would suggest that only in rare cases is an augment device needed for a Paprosky Type-IIB defect. Most Type-IIB defects are managed by reaming to convert an oblong defect to a hemisphere; therefore, augments are indicated primarily in the treatment of Type-IIIA and Type-IIIB defects.
In the series reporting on the treatment of Type-IIIA and IIIB defects, representing a total of 149 revision surgeries, the longest mean follow-up was just over five years (twenty-seven months to 107 months). With failure defined as aseptic loosening, only one series (thirty-four surgeries) showed a lower than 97% success rate31. The majority of the failures occurred in the setting of a pelvic discontinuity, suggesting that augmentation without direct fixation of the discontinuity may fail in the long-term. The concept of placement of augment devices to recreate an acetabular rim, followed by distraction of the discontinuity by impacting an oversized hemispherical shell, is of interest, but the follow-up is too short to know the final results with regard to longevity.
The limitations of the literature are inherent both to the short-term use of these implants and to the combining in these reports of cases in which an augment device was or was not used. The available literature includes only Level-IV evidence (case series). In only two series were augment devices used in all hips.
The primary complications are dislocation and infection. In the selected literature in which augment devices were used in at least half of the reconstructions, of a combined 149 revision cases in which a porous metal shell and augment were used, there were seven dislocations (5%) and nine infections (6%) (Table II). Other series have reported higher dislocation rates. Van Kleunen et al., in a large series of ninety-seven cases, reported a 10% infection rate and a 7% dislocation rate33. In eight of the ten infected hips for which a reoperation was performed, seven of the eight acetabular components were found to have bone ingrowth. A dislocation rate of 12% (seven of sixty) was reported by Unger et al.34.
In general, the use of acetabular augment devices in revision acetabular surgery is confined to Paprosky Type-IIIA and IIIB defects, with occasional use in the Type-IIB defect. Unitization of the augment device to the shell with cement seems reasonable, as the combined construct’s geometry may provide resistance to migration. The early results of these reconstructions are promising, yet it must be remembered that the mean follow-up is limited. Successful treatment of pelvic discontinuity with the use of porous metals still remains difficult. The next decade will determine whether the intraoperative flexibility afforded by the use of augment devices has resulted in a durable construct.
Source of Funding: No funding was received by either author for this work.
Investigation performed at Sutter Medical Center, Sacramento, California
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.
- Copyright © 2014 by The Journal of Bone and Joint Surgery, Incorporated