Secondary Logo

The Role of Constrained Liners in Total Hip Arthroplasty

Su, Edwin P, MD; Pellicci, Paul M, MD

Clinical Orthopaedics and Related Research®: March 2004 - Volume 420 - Issue - p 122-129

Recurrent instability after total hip replacement is a complex problem with extensive literature detailing multiple etiologies and solutions. It has been shown that the success of surgical treatment depends on the identification of the cause. Unfortunately, in certain situations, there may not be an optimal solution for dealing with the cause, or the cause may remain unidentified. In these cases, the success rate of surgical treatment of the unstable total hip replacement is only 40% to 50%. Constrained acetabular liners were developed to address the problem of recurrent instability by holding the femoral head captive within the socket. Before the use of constrained liners, there were no reliable solutions to dislocation arising from inadequate soft tissues, a deficient abductor mechanism, or neuromuscular disorders. We have used a constrained liner for these situations, with poor patient compliance and instability without a clear cause as relative indications for its use. Our experience with attaining joint stability using one type of constrained liner has resulted in a 97.6% success rate (83 of 85 hips) at 4.8 years, surpassing the outcomes achieved by other means. The intermediate followup after implantation of a constrained liner has not shown significant rates of component wear or loosening.

From the Department of Joint Replacement Service, The Hospital for Special Surgery, New York, NY.

Received for publication June 6, 2003.

Reprints: Paul M. Pellicci, MD, The Hospital for Special Surgery, 535 East 70th Street, New York, NY 10021 (e-mail:

Recurrent instability after total hip replacement is one of the most difficult and challenging issues facing the arthroplasty surgeon today. Dislocation after primary THA has been described in the literature with variable rates, depending on the series. Authors have reported dislocation rates ranging from 3% (n = 10,500)30 to 7% (n = 4164)14 after primary THA. Contributing factors to instability include component malposition, impingement, insufficient soft tissue tension, neuromuscular disorders, and patient noncompliance.1,4,23

In the revision situation, inadequate soft tissue and bone stock also may contribute to these issues, increasing the dislocation rate to as much as 15% (n = 12,956)26 in a large review, to 20%29 to 30%18 in smaller series of patients (n = 85 and 45, respectively). Morrey23 reported that dislocation rates after revision surgery were three times higher than after primary THA. The quality of the soft tissues continues to deteriorate with additional revision operations, making it difficult to achieve a stable joint.

Numerous surgical techniques have been devised to combat instability after THA, including component reorientation, capsulorraphy, trochanteric advancement, and conversion to a bipolar component. However, these surgical methods only have shown modest success in controlling instability, achieving a stable joint in 70% to 80% of patients.10,16,23,25 Furthermore, because these procedures mainly are designed to focus on component malposition and soft tissue laxity as a cause of instability, they do not address causes such as inadequate soft tissue structures, a deficient abductor mechanism, or neuromuscular disorders.

Captive, or constrained acetabular components were developed to deal with the problem of recurrent dislocations. Constrained components provide a powerful means to combat instability arising from various causes; to date, they have achieved success rates that surpass other solutions.11,12,28 We discuss our indications for the use of a constrained liner and the technical considerations involved in the implantation of these devices. Finally, we review the early and intermediate (5- to 10-year) experience with their use.

Back to Top | Article Outline


The success of revision operations for instability has been shown to be dependent on the identification of its cause. Revision was 70% successful in treating instability when a cause clearly was identified (29 of 43 hips); however, this rate decreased to 50% when there was no clear underlying cause (nine of 18 hips).6 The problem of recurrent instability after THA should be addressed in a systematic fashion to determine its cause.

A complete physical examination will help determine whether leg length discrepancy (leading to inadequate soft tissue tension), or abductor weakness is contributing to instability. Additionally, an interview with the patient may provide information about whether noncompliance, cognitive disorders, or neuromuscular deficiency have played a role. Good quality radiographs will allow the assessment of component position, impingement, trochanteric malunion or nonunion, or excessive wear as etiologic factors. Finally, the surgeon doing the revision also should be cognizant of infection as a potential cause of recurrent instability, because the inflammation and edema occurring with infection may lead to disruption of the soft tissue envelope about the hip.

Back to Top | Article Outline



It is estimated that recurrent instability will need to be treated surgically in at least 1% of all patients with THA.4,23 Surgical treatment should address any identifiable cause of dislocation for the greatest chance of success. Options such as component reorientation, capsulorraphy, trochanteric advancement, and conversion to a bipolar prosthesis have been described as solutions to instability.6,10,16,25,29 However, because these techniques primarily are designed to address soft tissue laxity and component malposition, their modest success may decrease further when applied to other causes. Furthermore, complications resulting from these methods must be considered, such as trochanter nonunion,10,16 and recalcitrant groin pain.25

The use of a constrained acetabular liner to treat recurrent instability is increasing because of its high success rate and reproducibility of results.11,12,27,28 Early reports cite its ease of use and reliable recreation of a stable joint as reasons for its consideration.11,27

Back to Top | Article Outline


Before the constrained liner, there were no good solutions to directly address inadequate soft tissue structures, a deficient abductor mechanism, or neuromuscular disorders as causes of instability. There may be inadequate soft tissues in patients who have had multiple revision surgeries, leading to a lack of a pseudocapsule providing a checkrein against dislocation. A deficient abductor mechanism may be a result of trauma, postoperative trochanteric malunion, or nonunion. Neuromuscular disorders such as Parkinson’s disease, post-polio syndrome, cerebral palsy, and residual weakness after stroke may affect the dynamic stabilizers of the hip. In these situations, the constrained liner is a powerful solution that confers joint stability. Its use also may be warranted in situations where a cause of instability cannot be identified or patient noncompliance is an issue.

In situations where the surgery must be done expeditiously or with minimal blood loss, revision with a constrained liner may be done quickly and with a limited exposure. Additionally, it does not preclude the use of other means to treat instability, such as trochanteric advancement.

Our indications for use of a constrained liner include inadequate soft tissues, a deficient abductor mechanism, and neuromuscular disorders. Relative indications for its use include instability without a clear cause, and cognitive disorders leading to poor patient compliance.

Back to Top | Article Outline


There are two current designs for constrained acetabular liners: the constrained tripolar, and the locking ring. Each liner is intended for use with its corresponding acetabular shell.

Back to Top | Article Outline

Constrained Tripolar

The Osteonics Omnifit constrained acetabular insert (Osteonics Corporation, Allendale, NJ) is designed with an inner bipolar bearing articulating within an outer, true liner (Fig. 1). The femoral head snaps into the bipolar bearing with an inner diameter of 22, 26, or 28 mm, which is free to rotate within the outer liner. The inner bipolar bearing is restricted from dislocating because the size of its circumferential arc is larger than the introitus of the extended walls of the true PE liner.27 The tested ROM ranges from 72°–84° depending on the inner to outer diameter ratio; pull-out strength before dislocation of the femoral head is 514 lb, and lever-out strength is 450 inch-lb (Osteonics Corporation).



The constrained Omnifit liner is available with a range of outer diameters: for the 22 mm inner diameter, 50 to 74 mm; for the 26 mm inner diameter, 56 to 74 mm; and for the 28 mm inner diameter, 58 to 74 mm. It is designed to be snapped into the hemispheric Osteonics SecurFit acetabular shell, but should be referenced with the compatibility chart provided by Osteonics to determine the size limitations. The smallest outer diameter acetabular shell that can accommodate the constrained liner is 50 mm.

Back to Top | Article Outline

Locking Ring

The Depuy (Warsaw, IN) Corporation manufactures the Duraloc constrained acetabular liner, which uses a locking ring mechanism. This component is designed similarly to the SROM constrained acetabular component (Depuy/Johnson & Johnson, Warsaw, IN). The liner achieves its axial capture of the femoral head with a Ti reinforcing ring that locks into a circumferential groove in the liner face. The advantage of the locking ring mechanism is that the thickness of the liner is not compromised by the constraining mechanism.

The liner is available with an inner diameter of 28 mm (outer diameter range, 48- to 66 mm), and 32 mm (outer diameter range, 52- to 66 mm), and is compatible with the Duraloc and Solution system acetabular shells (Depuy). The pull-out and lever-out strength before dislocation are greater than that of the SROM liner: 416 lb versus 300 lb, and 170 inch-lb versus 150 inch-lb, respectively, for the 28 mm inner diameter liner (Depuy Corporation).20 The ROM for the 28 mm inner diameter liner is 80° and 92° for the 32 mm inner diameter (Depuy Corporation).

Back to Top | Article Outline


Before doing revision surgery, one should know the size, make, and manufacturer of the existing implants. This information is necessary to address the issue of compatibility between the acetabular liner and shell, and between the taper of the femoral neck and head, because it may be necessary to change the head size.

When using either constraining liner, the surgeon must determine the optimal method of implantation. Because the available constrained liners are compatible only with the acetabular shells of their own manufacturers, implantation may be done by one of four methods: (1) Uncemented liner inserted into compatible, well-fixed acetabular shell (Fig. 2); (2) Uncemented liner inserted into new, uncemented acetabular shell; (3) Liner inserted with cement into bony acetabulum (Fig. 3); (4) Liner inserted with cement into well-fixed acetabular shell of another manufacturer (Fig. 4).







Cementing a new metal acetabular shell with a constrained liner snapped in place is another possible method of implantation and has been done by us in the past. We do not recommend this mode of fixation, because Shapiro et al28 reported one of eight hips required revision for aseptic loosening at 19 months, three sockets were possibly loose, and two sockets were probably loose at the time of most recent followup.

Currently, our protocol for using the constrained liner proceeds in the usual fashion with a preoperative workup of instability. We do all THAs via a standard posterolateral approach with the patient under hypotensive epidural anesthesia. At the time of surgery, the fixation and position of the femoral and acetabular components are assessed. If a cause of instability is identified, it is addressed in the appropriate manner (removal of impinging bone or soft tissue). We will elect to use a constrained liner if there exists an abductor mechanism or soft tissue deficiency, poor patient compliance, or an unclear cause of instability. Our experience has been limited to the Osteonics Omnifit constrained liner. There are data to suggest that the results with this component are superior to those with other models.11,12,27

The method in which the liner is implanted depends on the stability of the existing cup and the quality of host bone. Our primary objective is to insert the cup without cement; this goal, however, must be balanced with the principle of conserving acetabular bone stock. If the existing cup is loose or can be removed without excessive bone loss, we will implant a new press-fit acetabular shell and liner, with at least two transacetabular screws for additional fixation. If the existing cup is malpositioned, it must be removed regardless of its fixation. During insertion of the new acetabular shell, it is important to obtain good purchase with the screws; if the host bone is osteopenic, we will insert additional transacetabular screws to ensure stability.

If removal of a well-fixed implant is judged by the surgeon to require the destruction of excess host bone, we will cement the liner into a well-fixed shell. A liner with an outer diameter at least 2 mm smaller than the inner diameter of the metal shell is used to allow for an adequate cement mantle. The inside of the well-fixed metal shell and backside of the constrained PE liner are scored with a high-speed burr to roughen the surface for cement fixation. One bag of cement is mixed and pressurized into the shell using the back of a bulb syringe. The liner is placed in the same amount of version as the existing shell, with the elevation placed in the direction of instability.

Although cementing a liner into an existing shell is not our ideal scenario, it represents a compromise between adequate fixation and preservation of bone. Studies have shown that lever-out strength of polyethylene liners cemented into acetabular shells with a 2-mm cement mantle is greater than that of standard locking mechanisms (more than 600 inch-lbs of torque).3,5 These studies, however, did not specifically test constrained liners.

When the original shell diameter is less than 50 mm, it is clear that a constrained liner will not fit in this existing shell. Therefore, we remove the original metal shell using bone conservation techniques, and directly cement the constrained liner into the acetabulum using the same method of cement pressurization.

The inner diameter of this specific type of constrained liner is determined by the outer diameter of the acetabular shell; thus the surgeon does not have a choice of femoral head size. To optimize hip range of motion and minimize the risk of impingement, we avoid the use of skirted femoral heads, if possible. Clinically, however, we have not found a significant limitation of hip motion regardless of head or liner size.

After implantation of the liner, we proceed to close the posterior pseudocapsule via drill holes in the greater trochanter, if possible. The usual postoperative hip precautions are maintained to allow the soft tissues to heal.

Back to Top | Article Outline


The success of reoperation for chronic instability is reported variably from 50% to 70% using methods such as component reorientation, removal of impingement, capsulorraphy, conversion to bipolar arthroplasty, and trochanteric advancement.24 As seen from the aforementioned results, operative correction of instability after THA using these traditional means has had only moderate success, failing in as many as 50% of cases.

A review of the literature encompassing the experience to date with constrained liners is summarized in Table 1. The early experience using the SROM constraining liner was encouraging, resulting in success rates comparable with those achieved by other means. Lombardi et al20 used the SROM liner in a group of patients having primary and revision surgery and achieved stability after an average of 30 months in 50 of 55 patients (91%). Anderson et al2 reserved its use for revision circumstances and achieved stability in 15 of 21 patients (71%) during an average 31-month followup. In both studies, mechanical component failure (pullout of the acetabular liner from the shell and disengagement of the femoral head from the liner) played a significant role in the repeat dislocators. Some of these early failures may be attributable to component design.

Table 1

Table 1

Goetz et al11 reported greater success using the Osteonics constrained acetabular insert in patients with recurrent THA dislocations. The constrained liner prevented instability in 54 of 56 hips (96%) after an average followup of 3 years, which is a significant improvement over the traditional methods of achieving stability. Goetz et al12 also used the constrained liner in situations of neuromuscular deficiency and intraoperative instability, implanting 101 liners. In this setting, there was a failure rate of only 4%. Reoperations, other than for dislocation, were done for infection, heterotopic bone excision, periprosthetic fracture, and allograft failure.

Our experience with the Osteonics constrained liner also has been equally successful at controlling instability. Shapiro et al28 have reported on the results using the Osteonics constrained liner for recurrent dislocation. Eighty-five hips were analyzed in patients with an average followup of 4.8 years. Fifty-three constrained liners were inserted without cement into press-fit shells, 16 liners were cemented into well-fixed shells, eight liners were cemented directly into the acetabulum, and eight liners with their metal shells were cemented into place. Only two of the 85 hips had repeat dislocation (2.4%), which only occurred with dissociation of liners that had been cemented into well-fixed shells. Additional reoperations in this group of patients were done for aseptic loosening of the acetabular and femoral components, and excessive PE wear. The high percentage of stable hips achieved using the constrained liner is impressive and exceeds the success achieved by other operative means.

Additionally, this population of patients has had good functional results. Of 82 patients (85 hips) with a minimum of 3 years followup, 41 used no assistive device, 19 used a cane for ambulation, and 10 used a walker. There was no pain reported in 42 of the patients, mild pain in 19 patients, moderate pain in six patients, and severe pain in three patients.28

To date, no studies using the Depuy constrained liner have been published.

Back to Top | Article Outline


Failure to Prevent Dislocation

Most of the complications of constrained liners have involved failure of the capture mechanism or dissociation of the liner from the acetabular shell. Anderson et al2 and Lombardi et al20 reported failure of the SROM liner capture mechanism and disassembly of the liner from the shell. Fisher and Kiley9 also reported two mechanical failures of the SROM liner that occurred by disassembly of the liner from the shell. Additionally, Kaper and Bernini15 reported the failure of four SROM constrained liners. Two occurred by fracture of the constraining ring on the neck of the liner, with only one resulting in clinical instability. The other two resulted from the femoral head pulling out of the liner. Because of the design of the constrained component, closed reduction was not possible and open revision was necessary.

Similarly, there have been sporadic cases of mechanical failure with the Osteonics constrained liner. Table 2 shows the incidence of failure by method of implantation. Two failures reported by Shapiro et al28 and one failure reported by Goetz et al11 resulted from pullout of a liner cemented into a well-fixed acetabular shell. An additional failure occurred by dislodgement from the acetabulum of a newly inserted press-fit shell with a constrained liner snapped in place11; however, this was thought to be caused in part by its placement into a bulk allograft. In two other cases in a larger series12 there was failure of the metal retaining ring (Fig. 5) and dissociation of the liner from a new uncemented shell, leading to repeat dislocation. There was one failure attributable to excessive PE wear.28

Table 2

Table 2



Back to Top | Article Outline

Early Loosening and Osteolysis

Another concern with the use of a constrained liner is the potential for early loosening of the prosthesis, because there may be increased forces distributed to the host-implant interface. Investigators have reviewed followup radiographs using the criteria of Loudon and Charnley,21 Engh et al,7 Harris et al,13 and Massin et al22 to evaluate implant loosening. Table 2 shows the published incidence of aseptic loosening requiring revision for the Osteonics constrained liner, stratified by method of implantation; and Table 3 reviews the radiographic results of constrained liners to date. In reviewing 70 hips in patients with a 2-year minimum followup, Goetz et al12 reported five acetabular components that seemed to be loose (7.1%), one of which was revised. On the femoral side, 5.7% seemed to be loose, one of which was revised. Acetabular osteolysis was present in 2.9% of patients, and femoral osteolysis was present in 5.7%.

Table 3

Table 3

Shapiro et al28 have evaluated the largest series of constrained liners with a minimum of 3 years of radiographic followup. Of 85 patients, three acetabular components (3.5%) showed aseptic loosening and required revision. Two of these were implanted with a new press-fit shell without cement; one, as described previously, was inserted by cementing the metal acetabular shell with the liner snapped in place. There were two cases (2.4%) of focal, nonprogressive acetabular osteolysis in hips with a newly cemented metal shell and liner. On the femoral side, all newly inserted femoral stems were stable, without osteolysis. In the 49 existing femoral components, there were two cases of loosening (2.4%), one of which was revised. One femoral component (1.1%) had evidence of progressive osteolysis and was revised.

These numbers are comparable with previously reported rates of loosening and osteolysis in complex revision THA,17,19 albeit the followup is of intermediate length. Therefore we are cognizant of the possibility of increased rates of loosening of constrained implants and continue to monitor for it accordingly.

Back to Top | Article Outline

Excessive Polyethylene Wear

The additional constraint also may lead to increased PE wear via increased forces and an additional bearing surface in the case of the Osteonics Omnifit liner. Additionally, because of its design, the thickness of PE is limited in the Omnifit liner. Goetz et al11 did not find any cases of increased PE wear, although they acknowledge it is difficult to measure radiographically. Shapiro et al28 reported that there was one case of excessive PE wear necessitating revision.

Future considerations may include the use of highly cross-linked PE within the constrained liner.

Back to Top | Article Outline


Recurrent instability after THA is a frustrating problem leading to significant physical, emotional, and financial burden to the patient and arthroplasty surgeon. There is an extensive literature detailing the potential causes of recurrent instability, including component malposition, impingement, inadequate soft tissue tension, deficient abductor musculature, neuromuscular disorders, and patient noncompliance.

In the past, it has been reported that the results of treatment depend on identification of the cause of instability. Daly and Morrey6 reported that operative treatment of instability was successful in 70% of cases (29 of 43 hips) when a cause was clearly identified. The most common cause of instability was component malposition.1,8,23 However, even in this situation, reorientation of the components successfully achieved stability in only 80% of patients. The success of operative treatment decreased to 50% when there was no clear underlying cause of the instability.

From a practical standpoint, it often is difficult to determine the cause of dislocation after THA; the existing literature also reveals difficulty in establishing the underlying reason.6,23,24 Although component position may be assessed from preoperative radiographic studies, there are no reliable methods to assess soft tissue laxity or deficiency. Frequently the cause cannot be determined, or there may be multiple reasons for instability. As such, it is difficult to standardize treatment groups to compare the success rates for different surgical procedures. This additionally complicates the surgeon’s decision as to the best method of addressing recurrent instability. As stated previously, traditional surgical means of treating instability have failed in as many as ½ of all cases.

Constrained acetabular liners are a powerful revision option for achieving joint stability particularly when the cause of dislocation cannot be determined. They are designed to hold the femoral head captive via a secure locking mechanism. Although earlier studies using the SROM constraining liner offered success rates comparable with those achieved by other means (71% stable),2 more recent studies using the Osteonics liner have shown a dramatic improvement in stability. Goetz et al11 reported no additional dislocations in 54 of 56 hips (96.4%) treated for recurrent instability; Shapiro et al28 reported no dislocations in 83 of 85 hips (97.6%).

Another advantage of a constrained liner is its ease of use, allowing revision surgery to be done expeditiously or through a limited exposure. Additionally, using a constrained liner does not preclude the use of secondary procedures to achieve stability, such as trochanteric advancement.

The potential drawbacks of a constrained liner must be considered carefully. Because of the constraining mechanism, there is decreased ROM, although it does not seem to be clinically noticeable. There is a theoretical possibility of the increased transmission of stress to the implant-bone or implant-cement interface, leading to early loosening. To date, after an average of 4.8 years (with as much as 8 years followup), Shapiro et al28 have not found this to be the case. Additionally, there is the possibility of increased polyethylene wear and osteolysis; again this has not been seen in the followup period.

Finally, there is the possibility of component failure, either by the dissociation of the liner from the shell or of the disruption of the locking mechanism. Sporadic cases have been described, although much less frequently with the Osteonics device. When it occurs, however, open reduction must be done.

Despite these potential disadvantages, we think that a constrained acetabular liner is an excellent option for treating recurrent instability. Our algorithm to treat recurrent instability proceeds by seeking out identifiable etiologies. The constrained liner should not be used to compensate for other mechanical causes of instability such as component malposition. We reserve the use of a constrained liner for situations in which there are inadequate soft tissues, a deficient abductor mechanism, or neuromuscular disorders. Instability without a clear cause and poor patient compliance are relative indications for its use. We have been impressed by the success of constrained liners in achieving joint stability without evidence of significant drawbacks.

Back to Top | Article Outline


1. Ali Khan MA, Brakenbury PH, Reynolds IS. Dislocation following total hip replacement. J Bone Joint Surg. 1981;63B:214–218.
2. Anderson MJ, Murray WR, Skinner HB. Constrained acetabular components. J Arthroplasty. 1994;9:17–23.
3. Bensen CV, Del Schutte H Jr, Weaver KD. Mechanical stability of polyethylene liners cemented into acetabular shells. Crit Rev Biomed Eng. 2000;28:7–10.
4. Berry DJ. Unstable total hip arthroplasty: Detailed overview. Instr Course Lect. 2001;50:265–274.
5. Bonner KF, Delanois RE, Harbach G, et al. Cementation of a polyethylene liner into a metal shell: Factors related to mechanical stability. J Bone Joint Surg. 2002;84A:1587–1593.
6. Daly PJ, Morrey BF. Operative correction of an unstable total hip arthroplasty. J Bone Joint Surg. 1992;74A:1334–1343.
7. Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop. 1990;257:107–128.
8. Fackler CD, Poss R. Dislocation in total hip arthroplasties. Clin Orthop. 1980;151:169–178.
9. Fisher DA, Kiley K. Constrained acetabular cup disassembly. J Arthroplasty. 1994;9:325–329.
10. Fraser GA, Wroblewski BM. Revision of the Charnley low-friction arthroplasty for recurrent or irreducible dislocation. J Bone Joint Surg. 1981;63B:552–555.
11. Goetz DD, Capello WN, Callaghan JJ, et al. Salvage of a recurrently dislocating total hip prosthesis with use of a constrained acetabular component: A retrospective analysis of fifty-six cases. J Bone Joint Surg. 1998;80A:502–509.
12. Goetz DD, Capello WN, Callaghan JJ, et al. Salvage of total hip instability with a constrained acetabular component. Clin Orthop. 1998;355:171–181.
13. Harris WH, McCarthy JC Jr, O’Neill DA. Femoral component loosening using contemporary techniques of femoral cement fixation. J Bone Joint Surg. 1982;64A:1063–1067.
14. Heithoff BE, Callaghan JJ, Goetz DD, et al. Dislocation after total hip arthroplasty: A single surgeon’s experience. Orthop Clin North Am. 2001;32:587–591.
15. Kaper BP, Bernini PM. Failure of a constrained acetabular prosthesis of a total hip arthroplasty: A report of four cases. J Bone Joint Surg. 1998;80A:561–565.
16. Kaplan SJ, Thomas WH, Poss R. Trochanteric advancement for recurrent dislocation after total hip arthroplasty. J Arthroplasty. 1987;2:119–124.
17. Katz RP, Callaghan JJ, Sullivan PM, et al. Long-term results of revision total hip arthroplasty with improved cementing technique. J Bone Joint Surg. 1997;79B:322–326.
18. Kavanagh BF, Fitzgerald RH Jr. Multiple revisions for failed total hip arthroplasty not associated with infection. J Bone Joint Surg. 1987;69A:1144–1149.
19. Lawrence JM, Engh CA, Macalino GE, et al. Outcome of revision hip arthroplasty done without cement. J Bone Joint Surg. 1994;76A:965–973.
20. Lombardi AV Jr, Mallory TH, Kraus TJ, et al. Preliminary report on the S-ROM constraining acetabular insert: A retrospective clinical experience. Orthopedics. 1991;14:297–303.
21. Loudon JR, Charnley J. Subsidence of the femoral prosthesis in total hip replacement in relation to the design of the stem. J Bone Joint Surg. 1980;62B:450–453.
22. Massin P, Schmidt L, Engh CA. Evaluation of cementless acetabular component migration: An experimental study. J Arthroplasty. 1989;4:245–251.
23. Morrey BF. Instability after total hip arthroplasty. Orthop Clin North Am. 1992;23:237–248.
24. Morrey BF. Difficult complications after hip joint replacement. Dislocation. Clin Orthop. 1997;344:179–187.
25. Parvizi J, Morrey BF. Bipolar hip arthroplasty as a salvage treatment for instability of the hip. J Bone Joint Surg. 2000;82A:1132–1139.
26. Phillips CB, Barrett JA, Losina E, et al. Incidence rates of dislocation, pulmonary embolism, and deep infection during the first six months after elective total hip replacement. J Bone Joint Surg. 2003;85A:20–26.
27. Shapiro GS, Weiland D, Sculco TP, et al. The use of a constrained acetabular component for recurrent dislocation. Instr Course Lect. 2001;50:281–287.
28. Shapiro GS, Weiland DE, Markel DC, et al. The use of a constrained acetabular component for recurrent dislocation. J Arthroplasty. 2003;18:250–258.
29. Williams JF, Gottesman MJ, Mallory TH. Dislocation after total hip arthroplasty: Treatment with an above-knee hip spica cast. Clin Orthop. 1982;171:53–58.
30. Woo RY, Morrey BF. Dislocations after total hip arthroplasty. J Bone Joint Surg. 1982;64A:1295–1306.
© 2004 Lippincott Williams & Wilkins, Inc.