Prevention of dislocation in total hip revision surgery using a dual mobility design

ORIGINAL ARTICLE

Prevention of dislocation in total hip revision surgery

using a dual mobility design

Prévention de la luxation par la double mobilité lors de reprise

d’arthroplastie totale de hanche

R. Philippot

, P. Adam

, M. Reckhaus

, F. Delangle

, F.-X. Verdot

,

G. Curvale

, F. Farizon

a_{EA 4338, laboratoire de physiologie de l’exercice, Orthopedic and Trauma Surgery Department, centre hospitalier et}

universitaire de Saint-Étienne, 42055 Saint-Étienne cedex 2, France

b_{Orthopedic and Trauma Surgery Department, hôpital de La-Conception, 13385 Marseille cedex, France}

Accepted: 21 April 2009

KEYWORDS

Dual mobility cup; Revision total hip arthroplasty; Dislocation

Summary

Background: Postoperative dislocation is the commonest complication following revision total hip arthroplasty (THA).

Hypothesis and type of study: Dual mobility cups are supposed to reduce the risk of THA insta-bility. The present retrospective study tested this hypothesis on revision THAs and also, assessed this design contribution to acetabular fixation longevity.

Materials and methods:The series was homogeneous and continuous, comprising a total of 163 revision THAs: 110 of them were bipolar revisions and 53 were restricted to the acetabular component exchange. Mean patient age was 68.7 years (range: 34—92 years). NovaeTM _(SERF, Décines) dual mobility cups were used in all cases: 110 cementless cups were used and 53 cups were cemented in a Kerboull reinforcement ring due to severe acetabular bone loss.

Results:Mean patients’ follow-up (FU) was 60.4±17.6 months. There were six early disloca-tions (which were reduced without additional surgery and remained recurrence-free) and two cases of acetabular loosening. The total postoperative dislocation rate at the end of follow-up was 3.7% and the 7-year cfollow-up survivorship rate was 96.1% (95% CI: 92.8—99.2%). In revision for aseptic loosening, the instability rate was 2.9%; in the higher instability risk groups (i.e., revision for infection and or recurrent instability) the dislocation rate was respectively 9% and 0%.

Discussion: Dual mobility cups provided a dislocation rate of only 3.7% in revision THA, compa-rable to the one reported with standard implants for primary THA. This kind of cup design

∗_{Corresponding author.}

E-mail address:r [email protected](R. Philippot).

1877-0568/$ – see front matter © 2009 Elsevier Masson SAS. All rights reserved. doi:10.1016/j.otsr.2009.04.016

408 R. Philippot et al. is especially suited to deal with high instability risk revision cases, where constrained compo-nents are generally recommended. It can also be indicated in cases of aseptic loosening, where it resulted in a 2.9% dislocation rate and only two impending failures of fixation. In terms of mechanical failure rate, these numbers compare well to the ones pertaining to tripolar and con-strained implants. These later alternatives remain possible options but are not fully efficient in terms of long-term stability and fixation longevity.

Level of study. — Level IV, retrospective or records-based. © 2009 Elsevier Masson SAS. All rights reserved.

Introduction

A 2 to 5% rate of postoperative dislocation following pri-mary total hip arthroplasty (THA) has been reported in large series[1,2]. In revision THA, factors such as muscular insuffi-ciency, enlarged synovectomy, implant positioning problems or the need for bone reconstruction have been implicated in postoperative instability. Depending on the reasons for revision, mean dislocation rates in the literature range from 5 to 30% [3—6]. Constrained or tripolar cup designs have reduced postoperative dislocation rates, but only to a lim-ited extent and to the detriment of long-term acetabular fixation[7—11].

The dual mobility concept was introduced by Gilles Bous-quet to reduce the rate of postoperative dislocation. Some studies of dual mobility cups deployed in primary surgery reported very low rates of postoperative implant instability

[12—14]. The dual mobility design has also proved its worth in revision surgery for chronic implant instability [15,16]. The present study assessed the interest of dual mobility cups in revision THA in a continuous retrospective series of uni- or bipolar revision, focusing on postoperative dislocation rates and long-term implant fixation.

Material and methods

Patients

The study concerned a homogeneous continuous series of 163 revision THAs performed in two teaching hospitals between January 1999 and December 2004. All patients undergoing acetabular revision using a dual mobility cup were included; simple polyethylene insert replacement and unipolar femoral revision were excluded. Mean patient age was 68.7 years (34—92), 103 females and 60 males. Seventy patients had single-joint damage (Charnley A), 62 were classified as Charnley B, and 31 as Charnley C.

One hundred and ten revisions were bipolar and 53 were limited to cup revision. One hundred and sixteen cases were primary and 34 secondary revisions, and in 13 cases this was at least the third revision. The main indications were bipo-lar aseptic loosening (n= 72), aseptic cup loosening (n= 24), recurrent loosening (n= 26), reimplantation following deep infection (n= 33), and periprosthetic fracture (n= 8). The approach was posterolateral in 139 cases, transgluteal in seven and anterolateral in 17.

Acetabular bone loss was classified according to the American Academy of Orthopedic Surgeons (AAOS) scale

[17], definitively assessed peroperatively. Forty-eight cases were stage I, 59 stage II, 47 stage III, and 9 stage IV.

Acetabular implant characteristics

Only dual mobility cups were used. The mobile insert pro-viding dual articulation between metalback and head was in UHMWPE polyethylene, which, being retentive, was force-impacted on to the implant head. The heads were all in cobalt—chrome alloy, with a 22.2 mm diameter. This gave the implant a slight mobility between the implant head and the concave side of the polyethylene insert, and a great mobility between the convex side of the insert and the polished concave surface of the stainless steel cup.

All of the implants were of the NovaeTM _{range (SERF,} Décines):

• 38 Novae SunfitTM_{cups (SERF, Décines). This is a} cylindro-hemispheric stainless steel cup with cementless press-fit fixation ensured by equatorial macrostructures and a bilayer coating comprising an Al2O3 alumina plasma undercoat and a hydroxyapatite plasma top coat. This model was used in case of AAOS stage-I bone loss; • 51 Novae StickTM _{cups (SERF, Décines, France). This is a}

cylindrohemispheric stainless steel cup, intended to be cemented. It is identical to the Novae SunfitTM_{cup except} that the macrostructures cover the entire surface, with-out coating. This model was used when cementing was required, with either a Novae ArmTM _acetabular compo-nent (seven cases) or a Kerboull cross (44 cases); • 58 Novae-1TM _{cups (SERF, Décines). This is a}

cylindro-hemispheric stainless steel cup with cementless so-called ‘‘tripod’’ press-fit fixation, having two deep anchoring plots and a superior mooring screw with a recommended 45◦ iliac orientation. The coating is the same as in the Novae SunfitTM_{cup. This model was used in case of AAOS} stage I or II bone loss;

• 16 Novae CoptosTM_{cups (SERF, Décines, France). This is a} hemispheric stainless steel cup with cementless press-fit fixation, identical to the Novae SunfitTM_{cup but with two} superior feet, each receiving two screws, and an inferior obturator hook to enhance stability. This model was used in case of AAOS stage II or III bone loss.

Stainless steel Novae ArmTM _{acetabular components} (SERF, Décines, France) and KerboullTM _crosses (Strycker-Howmedica) were used for AAOS stages III and IV. Such reinforcements were always fitted using the same protocol: positioning and fixing the reinforcement in healthy bone, acetabular floor allograft using an irradiated and cryopre-served bone-bank head, then cementing the Novae Stick cup into the reinforcement.

Table 1 Dual mobility cup survivorship (failure defined as surgical revision for aseptic loosening of the component). Interval (months) Number of events (n) Number at risk (n) Cumulative survival Cumulative failure Standard deviation (survival) 0—10 0 163 1 0 0 10—20 0 163 1 0 0 20—30 0 159 1 0 0 30—40 1 144 1 0 0 40—50 0 116.5 0.993 0.007 0.007 50—60 0 85.5 0.993 0.007 0.007 60—70 0 55.5 0.993 0.007 0.007 70—80 1 31 0.993 0.007 0.007 80—90 0 20 0.961 0.039 0.032 Assessment methods

All patients had regular clinical and X-ray follow-up (FU). Clinical assessment used Postel Merle d’Aubigné scoring

[18], and X-ray assessment was based on AP pelvic and AP and lateral hip images. For the cups, radiolucency, osteol-yses and cavities were identified and located by Delee and Charnley zones[19]. On the femoral side, signs of loosening were identified and located in terms of the zones described by Gruen et al.[20]. Heterotopic ossification was classified following Brooker et al.[21].

Cup survivorship at end of FU was analyzed actu-arially [22] (with CI > 95%), with failure defined as any surgical revision for aseptic loosening. The incidence of implant dislocation at end of FU was studied by recording all dislocation episodes; the main disloca-tion risk factors were predefined (age, sex, approach route, and number of previous interventions) and sig-nificant differences were sought between patients with and without postoperative dislocation episodes. Statis-tical analysis on StatviewTM _{software used univariate} parametric tests, with the significance threshold set at p < 0.05.

Results

At end of FU, six patients had died and one was lost to FU. Mean Postel Merle d’Aubigné score had risen from 8.2 (2.4 for pain, 3 for mobility, and 2.8 for walking ability) preoperatively to 14.8 (5.4 for pain, 5 for mobility, and 4.4 for walking ability). Mean FU was 60.4±17.6 months (24—112). End of FU X-ray analysis found three evolutive radiolucent areas in Delee and Charnley zone III and two in zone II.

Failure and complications

Perioperative complications comprised: two greater trochanter fractures and one supracondylar fracture man-aged by plate fixation. There was also one immediate postoperative common fibular nerve paralysis.

Postoperative complications comprised:

• six early dislocations (3.7%), within 6 months, none requir-ing surgical revision (reduction was obtained after close manipulation);

• two fractures under the femoral implant (1.2%), at 36 and 42 months respectively;

• five infections (3%), including two early infections man-aged by simple surgical lavage, and 3 deep infections requiring implant removal at 15, 18 and 20 months, respectively;

• two cases of greater trochanter nonunion (1.2%); • one femoral implant fracture (0.6%) at 36 months,

requir-ing surgical revision;

• two cases of cup loosening (1.2%): one traumatic, at 2 months, and the other aseptic and nontraumatic, at 15 months;

• one polyethylene component replacement (0.6%) at 48 months, for severe early wear.

The overall postoperative dislocation rate at end of FU was 3.7%. There was no recurrence of early dislocation. All cases could be managed by external maneuver under gen-eral anesthetic and curarization. None were complicated by subsequent intraimplant dislocation. The overall end of FU rate of surgical revision for whatever reasons was 6.7%, and 1.2% (two cases) for failed cup fixation.

Two subgroups at particular risk of postoperative instability—–revision for recurrent dislocation (26 cases), and reimplantation following deep infection (33 cases)—–showed postoperative dislocation rates at end of FU of respectively 0% and 9% (equal to three cases). In the subgroup of revision for aseptic loosening (104 cases), the end of FU dislocation rate was 2.9% (equal to three cases).

Statistical analysis

Seven-year dual mobility cup survivorship, with failure defined as revision surgery for aseptic loosening, was 96.1% (95% CI: 92.8—99.2%) (Table 1). None of the prede-fined risk factors (age, sex, approach route, and number of previous interventions) significantly affected disloca-tion rates in the present series, and none could thus be assigned any positive predictive value for postoperative dislocation.

410 R. Philippot et al. Table 2 Comparison of dislocation and acetabular failure rates in series of revision THA secondary to deep infection.

Series Cup type Number of

cases

Mean FU (months)

Dislocationsn(%) Revision for aseptic failuren(%) Hartman and Garvin[26]. Single mobility

standard cup

34 24 5 (15%) NS Berend et al.[32] S-ROMTM _{60 120 13 (21.4%) 15 (25%)}

Charlton et al.[27] Single mobility standard cup

44 24—108 6 (14%) NS Present series Dual mobility cup 33 60 3 (9%) 0

Discussion

Series as a whole

Our analysis of the present revision THA series focused especially on the postoperative dislocation rate. The usual rates in the literature, for standard implants, range from 5 to 30% [4,6,23,24], compared to 3.7% in the present series of dual mobility cups—–i.e., close to the rate found with standard implants used in primary surgery. Most pub-lished series do not identify revision for recurrent dislocation or reimplantation following deep infection. The present results, however, highlighted reimplantation following deep infection (33 cases) as the situation at highest risk of postop-erative dislocation. To clarify the discussion, we therefore distinguish three subgroups in the present series: reim-plantation following deep infection (33 cases), recurrent dislocation (26 cases), and aseptic loosening (104 cases).

Reimplantation following deep infection

Vielpeau and Lortat Jacob[25]describe the various instabil-ity risk factors for reimplantation following deep infection: muscular atrophy secondary to prolonged immobilization, the enlarged synovectomy required during reimplantation, significant muscle retraction (especially when traction or a temporary spacer were not used to maintain the joint space), tricky and uncertain implant positioning in what tends to be poor-quality bone, and unequal residual limb length (Table 2). In the present subgroup of 33

reimplan-tations following deep infection, there were three cases (equal to 9%) of postoperative dislocation, which is quite a good result compared to the few published series of revision THA for infection (seeTable 2). Hartman and Gavin[26]and Charlton et al.[27]respectively reported 15% and 14% post-operative dislocation in 34 and 44 reimplantations following deep infection. Charlton et al. [27]concluded by strongly recommending the use of constrained cups, given their high postoperative dislocation rate. Dual mobility cups have a clear role to play in these high dislocation risk situations, reducing the rate of postoperative instability.

Revision for recurrent dislocation

For the recurrent dislocation subgroup, the present series merely confirms the extensive literature findings (see

Table 3). In the subgroup of 26 patients, there was no recur-rence of dislocation and the dual mobility concept once again seemed to be highly effective in terms of stability. Additional elevated posterior wall certainly reduce postop-erative dislocation [28,29]but only to a limited extent as they entail a risk of dislocation in the opposite direction. Moreover, by increasing implant constraint with reitera-tive conflict between neck and elevated rim, they lead to the release of polyethylene particles into the joint, thus potentially increasing the risk of later loosening[30]. Some authors recommend using constrained cups in this indication

[31], but published results still show high rates of residual dislocation of between 3 and 28%, with up to 17% mechanical failure[32,33].

Table 3 Comparison of dislocation and acetabular failure rates according to cup type (constrained or nonconstrained) in revision THA for recurrence of dislocation.

Authors Cup type Number

of cases

Mean FU (months)

Dislocationn(%) Revision for aseptic failuren(%) Toomey et al.[29] Polyethylene

antidislocation crescent

13 72 3 (23%) 0 Gholve et al.[28]. PLADTM_{antidislocation}

crescent

21 17 2 (9,5%) 0 Berend et al.[32] S-ROMTM _{138 120 40 (28.9%) 1 (0.8%)}

Callaghan et al.[40] TripolarTM _{56 48—120 4 (7%) 4 (7%)}

Khan et al.[33] Trident constrained acetabular cupTM

34 36 1 (3%) 6 (17%) Beaule et al.[10] Jumbo HeadTM _{12 72 1 (8%) 1 (8%)}

of dislocation of revision hip arthroplasty using a dual mobility design 411

Table 4 Comparison of dislocation and acetabular failure rates, according to cup type and to degree of bone loss, in revision THA for aseptic loosening.

Authors Cup type Number

of cases

Bone loss distribution (AAOS classification)

Mean FU (months)

Dislocation (n) (%)

Revision for aseptic loosening (n) (%) Parratte et al.[23] Hilock RevTM _{34 III : 100% 72 3}

9%

2 5.8% Pieringer et al.[35] Burch-SchneiderTM_{cage 67 II : 30%}

III : 70%

50 11

16.4%

4 6% Bostrom et al.[34] Burch-SchneiderTM_{cage 31 II : 7%}

III : 74% IV : 19% 30 5 16% 2 7% Unger et al.[36] Monoblock Acetabular Cup

SystemTM 60 I : 2% II : 86% III : 12% 42 7 12% 2 3.3% Kawanabe et al.[37] KerboullTM_{reinforcement}

device with single mobility cup

46 II : 31% III : 67% IV : 2% 105 4 8% 1 2.5% Present series Dual mobility±reinforcement 104 I : 26%

II : 33% III : 28% IV : 13% 60.4 3 2.8% 2 1.9%

412 R. Philippot et al. Revision for aseptic loosening

The end of FU postoperative dislocation rate in the subgroup of revision THA for uni- or bipolar aseptic loosening was 2.9% (Table 4). Many published series using standard cups report much higher rates of between 5 and 30%[23,34—38], and the dual mobility concept emerges as a clear improvement in terms of preventing dislocation. Constrained and tripolar implants have been recommended by some authors where the risk of postoperative dislocation is high. Constrained implants use implant head retention to reduce instability

[8]. Stability is indeed achieved, but the great increase in interface constraint entails a risk of long-term fixation fail-ure, which is the main drawback of this technique [11]. The increase in interface constraint is due to considerably reduced joint amplitude, as the way the implant functions tends to induce impingement, as Murray[39]pointed out for nonconstrained implants involving an antidislocation long posterior wall. The polyethylene component is, moreover, thin in these designs. The risk of long-term fixation fail-ure is increased in revision surgery, where there is often bone loss and poor bone quality. Secondary implant fixation quality is an essential objective, making increased inter-face constraint counterproductive. Tripolar cups, such as the TripolarTM_{model (Osteonics Corp, Allendale, NJ), also} meet the objective of stability[40], and their enhanced joint amplitude reduces interface constraint by limiting mechani-cal conflict. The drawback of the tripolar concept lies in the greater number of interfaces, which doubtless accounts for the excessive incidence of aseptic loosening reported in the literature[40,41].

We therefore give preference to dual mobility cups, which show 7-year survivorship (with failure defined as sur-gical revision for aseptic cup loosening) of 96% (95 % CI: 92.8—99.2%), although FU in the present series was only 60.4 months, so that many events may be yet to come. Studies of implant fixation in revision THA, however, report survivor-ship values very comparable to those for the present series

[23,34—37]. More specifically regarding AAOS stages III and IV acetabular bone loss (with 56 cases in the present series), we found no acetabular failure at end of FU, whereas Par-ratte et al. [23] and Kerboull et al. [42], in comparable series, reported rates of respectively 5.8% and 5% at 6 and 8 years. Thus the dual mobility design does not seem to have a negative impact on acetabular fixation and can be used in all situations of revision involving bone defect, from stage I to stage V, with or without the use of reinforcements or graft.

Conclusion

Our overall postoperative dislocation rate of 3.7% at end of FU in a series of revision THA using a dual mobility cup confirms that this design provides stability, at least up to medium term. This conclusion holds for each of our three subgroups, corresponding to three precise clinical situa-tions: reimplantation following deep infection, recurrent dislocation, and aseptic loosening. With 7-year survivorship of 96%±3.2, the dual mobility design appears to provide better fixation than the constrained or tripolar cups often recommended in these indications. Our FU was relatively

short, but the quality of the results in terms of both stabil-ity and secondary fixation indicates that dual mobilstabil-ity cups are an attitude of choice in revision THA, with or without bone defect.

References

[1] Berry DJ, Von Knoch M, Schleck CD, Harmsen WS. The cumula-tive long-term risk of dislocation after primary Charnley total hip arthroplasty. J Bone Joint Surg Am 2004;86:9—14. [2] Huten D. Luxations et subluxations des prothèses totales de

hanche. In: Masson, editor. Cahier d’enseignement de la Sof-cot, 55. 1996. p. 19—46.

[3] Lewinnek GE, Lewis JL, Tarrr, Compere CL, Zimmerman Jr. Dis-locations after total hip-replacement arthroplasties. J Bone Joint Surg Am 1978;60:217—20.

[4] Gioe TJ. Dislocation following revision total hip arthroplasty. Am J Orthop 2002;31:225—7.

[5] Alberton GM, High WA, Morrey BF. Dislocation after revision total hip arthroplasty: an analysis of risk factors and treatment options. J Bone Joint Surg Am 2002;84:1788—92.

[6] Ballard WT, Callaghan JJ, Johnston RC. Revision of total hip arthroplasty in octogenarians. J Bone Joint Surg Am 1995;77:585—9.

[7] Callaghan JJ, Heithoff BE, Goetz DD, Sullivan PM, Peder-sen DR, Johnston RC. Prevention of dislocation after hip arthroplasty: lessons from long-term follow-up. Clin Orthop 2001;393:157—62.

[8] Anderson MJ, Murray WR, Skinner HB. Constrained acetabular components. J Arthroplasty 1994;9:17—23.

[9] Williams JT, Ragland PS, Clarke S. Constrained components for the unstable hip following total hip arthroplasty: a literature review. Int Orthop 2007;31:273—7.

[10] Beaule PE, Roussignol X, Schmalzried TP, Udomkiat P, Amstutz HC, Dujardin FH. Tripolar arthroplasty for recurrent total hip prosthesis dislocation. Rev Chir Orthop 2003;89:242—9. [11] Yun AG, Padgett D, Pellicci P, Dorr LD. Constrained

acetab-ular liners: mechanisms of failure. J Arthroplasty 2005;20: 536—41.

[12] Aubriot JH, Lesimple P, Leclercq S. Study of Bousquet’s non-cemented acetabular implant in 100 hybrid total hip prostheses (Charnley type cemented femoral component). Average 5-year follow-up. Acta Orthop Belg 1993;59(Suppl. 1):267—71. [13] Farizon F, De Lavison R, Azoulai JJ, Bousquet G. Results with a

cementless alumina-coated cup with dual mobility. A twelve-year follow-up study. Int Orthop 1998;22:219—24.

[14] Philippot R, Adam P, Farizon F, Fessy MH, Bousquet G. Survival of cementless dual mobility sockets: ten-year follow-up. Rev Chir Orthop 2006;92:326—31.

[15] Beguin L, Adam P, Farizon F, Fessy MH. Total hip arthroplasty: treatment of chronic instability using a double-mobility cup. J Bone Joint Surg Br 2002;84(Suppl. 1):52.

[16] Leclercq S, El Blidi S, Aubriot JH. Bousquet’s device in the treatment of recurrent dislocation of a total hip prosthesis. A propos of 13 cases. Rev Chir Orthop 1995;81:389—94. [17] D’antonio JA, Capello WN, Borden LS, et al. Classification and

management of acetabular abnormalities in total hip arthro-plasty. Clin Orthop 1989;243:126—37.

[18] Merle D’aubigné R. Cotation chiffrée de la fonction de hanche. Rev Chir Orthop 1970;56:481—6.

[19] Delee JG, Charnley J. Radiological demarcation of cemented sockets in total hip replacement. Clin Orthop 1976;121:20—32. [20] Gruen TA, Mcneice GM, Amstutz HC. ‘‘Modes of failure’’ of cemented stem-type femoral components: a radiographic anal-ysis of loosening. Clin Orthop 1979;141:17—27.

[21] Brooker AF, Bowerman JW, Robinson RA, Riley LH. Ectopic ossification following total hip replacement. Incidence

and a method of classification. J Bone Joint Surg Am 1973;55:1629—32.

[22] Lettin AW, Ware HS, Morris RW. Survivorship analysis and confidence intervals. An assessment with reference to the Stan-more total knee replacement. J Bone Joint Surg Br 1991;73: 729—31.

[23] Parratte S, Argenson JN, Flecher X, Aubaniac JM. Acetabular revision for aseptic loosening in total hip arthroplasty using cementless cup and impacted morselized allograft. Rev Chir Orthop 2007;93:255—63.

[24] Thomasson E, Guingand O, Terracher R, Mazel C. Perioperative complications in revision hip surgery and their predictive fac-tors. A series of 181 consecutive procedures. Rev Chir Orthop 2001;87:477—88.

[25] Vielpeau C, Lortat Jacob A. Les reprises de prothèses totales de hanche infectées. Rev Chir Orthop 2002;88(Suppl.):162—216. [26] Hartman CW, Garvin KL. Dislocation of the hip after

reimplan-tation for infection: an analysis of risk factors. Clin Orthop 2006;447:24—7.

[27] Charlton WP, Hozack WJ, Teloken MA, Rao R, Bissett GA. Complications associated with reimplantation after girdlestone arthroplasty. Clin Orthop 2003;407:119—26.

[28] Gholve PA, Lovell ME, Naqui SZ. Minimal surgical approach for recurrent hip dislocation using the posterior lip augmen-tation device for the Charnley hip arthroplasty. J Arthroplasty 2006;21:865—8.

[29] Toomey SD, Hopper RH, Mcauley JP, Engh CA. Modular com-ponent exchange for treatment of recurrent dislocation of a total hip replacement in selected patients. J Bone Joint Surg Am 2001;83:1529—33.

[30] Nicholas RM, Orr JF, Mollan RA, Calderwood JW, Nixon JR, Watson P. Dislocation of total hip replacements. A compar-ative study of standard, long posterior wall and augmented acetabular components. J Bone Joint Surg Br 1990;72: 418—22.

[31] Bottner F, Steinbeck J, Winkelmann W, Gotze C. Acetabular augmentation ring for recurrent dislocations in revision arthro-plasty. Clin Orthop 2005;436:151—7.

[32] Berend KR, Lombardi AV, Mallory TH, Adams JB, Russell JH, Groseth KL. The long-term outcome of 755 consecutive constrained acetabular components in total hip arthro-plasty: examining the successes and failures. J Arthroplasty 2005;20(Suppl. 3):93—102.

[33] Khan RJ, Fick D, Alakeson R, Lim G, Nivbrant B, Wood D. The constrained acetabular component for hip instability. J Arthro-plasty 2007;22:377—82.

[34] Bostrom MP, Lehman AP, Buly RL, Lyman S, Nestor BJ. Acetab-ular revision with the Contour antiprotrusio cage: 2- to 5-year follow-up. Clin Orthop 2006;453:188—94.

[35] Pieringer H, Auerspergv, Bohler N. Reconstruction of severe acetabular bone-deficiency: the Burch—Schneider antiprotru-sio cage in primary and reviantiprotru-sion total hip arthroplasty. J Arthroplasty 2006;21:489—96.

[36] Unger AS, Lewis RJ, Gruen T. Evaluation of a porous tantalum uncemented acetabular cup in revision total hip arthroplasty: clinical and radiological results of 60 hips. J Arthroplasty 2005;20:1002—9.

[37] Kawanabe K, Akiyama H, Onishi E, Nakamura T. Revision total hip replacement using the Kerboull acetabular reinforcement device with morsellised or bulk graft: results at a mean follow-up of 8, 7 years. J Bone Joint Surg Br 2007;89:26—31. [38] Courpied JP, Migaud H. Reprise fémorale dans les

arthro-plasties itératives aseptiques de la hanche. Rev Chir Orthop 2000;86(Suppl.):33—90.

[39] Murray DW. Impingement and loosening of the long posterior wall acetabular implant. J Bone Joint Surg Br 1992;74:377—9. [40] Callaghan JJ, Parvizi J, Novak CC, et al. A constrained liner cemented into a secure cementless acetabular shell. J Bone Joint Surg Am 2004;86:2206—11.

[41] Bremner BR, Goetz DD, Callaghan JJ, Capello WN, Johnston RC. Use of constrained acetabular components for hip insta-bility: an average 10-year follow-up study. J Arthroplasty 2003;18(Suppl. 1):131—7.

[42] Kerboull M, Hamadouche M, Kerboull L. The Kerboull acetabu-lar reinforcement device in major acetabuacetabu-lar reconstructions. Clin Orthop 2000;378:155—68.