The Journal of Arthroplasty Vol. 23 No. 4 2008

Analysis and Results of Dissociation of thePolyethylene Liner in the Harris-Galante I

Acetabular Component

Shu Saito, MD, Junnosuke Ryu, MD, Masayuki Seki, MD, Takao Ishii, MD,Kaichiro Saigo, MD, and Sei Mori, MD

Abstract: The Harris-Galante I modular acetabular components (Zimmer, Warsaw,Ind) were most widely used in the 1980s in primary and revision total hiparthroplasty. We assessed 76 primary total hip arthroplasties performed using Harris-Galante I modular acetabular components. Dissociation of the polyethylene lineroccurred in 5 patients and required revision surgery. None of the 76 hips hadfractures of the metal locking tines on the metal shell. Liner dissociation wasattributed to fatigue fracture of the polyethylene, a result of a gap between the linerand the metal shell. The thin polyethylene liner rim aggravated failure. The revisiontechnique was to cement the polyethylene liner into the well-fixed Harris-Galantemetal shell. The average follow-up observation period after revision surgery was 6.4years; and there was no recurrence of dissociation, which supports the use of ourrevision technique. Key words: Harris-Galante I prosthesis, total hip arthroplasty,dissociation, polyethylene liner, fatigue fracture, polyethylene rim.© 2008 Elsevier Inc. All rights reserved.

Harris-Galante cementless prostheses (Zimmer,Warsaw, Ind) have been used extensively in primaryand revision total hip arthroplasty (THA). Thelongevity of this implant has been well described inthe literature [1,2]. In 1985, the Harris-Galante Icementless prosthesis was introduced in Japan. Inour previous report of good midterm results withHarris-Galante cementless prostheses, 1 dissociationof the polyethylene liner was noted and wasattributed to fatigue fracture caused by an inap-propriate design of the polyethylene liner rim [3].Schmalzried and Harris reported that polyethyleneliner dissociation was found in 1 of 134 hips and was

From the Department of Orthopedic Surgery, Nihon UniversitySchool of Medicine, Tokyo, Japan.

Submitted August 24, 2006; accepted June 8, 2007No benefits of funds were received in support of this study.Reprint requests: Shu Saito, MD, Department of Orthopedic

Surgery, Nihon University School of Medicine 30-1 Oyaguchi,Kamimachi, I tabashi-Ku, Tokyo 173-8610 Japan.

© 2008 Elsevier Inc. All rights reserved.0883-5403/08/2304-0006$34.00/0doi:10.1016/j.arth.2007.06.004

522

caused by a defective design of the liner, which isthin enough to break [4]. Several studies allude tofailure of the tine-locking mechanism of the Harris-Galante I acetabular components as a cause forrevision surgery [5-8]. Werle et al reported 7 hipsthat required revision of the entire acetabular shell,resulting in marked loss of acetabular bone stock [9].Of the 76 Harris-Galante I acetabular componentswe implanted, 5 had liner dissociation; none of the 5had fracture of the metal locking tines on the metalshell. Revision surgery was performed withoutremoval of the well-fixed shell by cementing thenew polyethylene liner into the remaining shell. Weconducted this study to evaluate the failuremechan-ism of theHarris-Galante polyethylene liners and theeffectiveness of cementation of a new polyethyleneliner in revision surgery.

Materials and Methods

From July 1985 to December 1991, we performedTHA on 76 hips (70 patients) using a Harris-Galante

Table 1. Clinical Details of 5 Patients With Dissociation of Polyethylene Liner in Harris-Galante I Acetabular Component

Patient Demographics

Patient Sex

Age (y) atPrimary Total

Hip Arthroplasty Diagnosis ComponentTimeIn Situ

Duration of Follow-UpMonitoring AfterRevision Surgery

1 F 73 Developmental dysplasia of the hip Harris-Galante I 7 y 12 y2 F 54 Developmental dysplasia of the hip Harris-Galante I 11 y 2 mo 7 y 3 mo3 F 65 Developmental dysplasia of the hip Harris-Galante I 12 y 6 mo 5 y 6 mo4 F 65 Developmental dysplasia of the hip Harris-Galante I 8 y 5 mo 4 y 8 mo5 F 56 Developmental dysplasia of the hip Harris-Galante I 11 y 6 mo 4 y

Harris-Galante I Acetabular Component � Saito et al 523

I prosthesis. In all 76 hips, the abduction angle of theacetabular component averaged 34.2° (range, 25°-45°). The anteversion angle of the acetabularcomponent, measured on the axial radiographicview of the joints, averaged 14.8° (range, 12°-20°).Five hips underwent revision surgery because ofpolyethylene liner dissociation, followed by cement-ing the liner into the remaining shell.All patients were female. At primary THA, their

mean age was 62.6 years (range, 54-73 years);average height, 156 cm (range, 148-162 cm); andaverage weight, 65 kg (range, 56-73 kg). The initialdiagnosis was developmental dysplasia of the hip inall patients. The acetabular component used for allwas the Harris-Galante modular component; thefemoral component was the Harris-Galante porous-coated stem with a 28-mm head. The externaldiameter of themetal cup averaged 49.2 mm (range,46-52 mm). Five screws were used in all patients.The mean longevity of the implants was 9.6 years(range, 7 years to 12 years 6 months; Table 1).The hips were clinically rated before and after

dissociation and at the most recent follow-upexamination using the Harris Hip Score (HHS)[10]: a score N90 points was excellent; 80 to 89,good; 70 to 79, fair; and b70, poor.Acetabular component fixation was evaluated

using the method described by Tompkins et al [11].Femoral component fixation was evaluated usingthe method described by Engh et al [12]. Moreover,we searched for signs of osteolysis around bothcomponents. All retrieved acetabular componentswere examined grossly for signs of wear and todetermine the cause of liner dissociation.

Results

Liner dissociation occurred in 2 patients whenthey sat down on a toilet. There were no apparentexterior signs of injury in the other 3 patients. Theaverage HHS was 95 points (range, 91-100 points),or excellent, before dissociation in all cases. Afterdissociation, the average score was 75 points (range,

71-79 points), or fair, in all cases. For all 5 patients,the average HHS score once again improved toexcellent, 93 points (range, 91-96 points), afterrevision surgery.

Radiographic Findings

Radiographic findings were similar for all 5patients, including eccentric femoral head placementwithin the metal shell, particularly displaced super-olaterally (Fig. 1). Radiographic evaluation showedthat acetabular component fixation was stable andthat the femoral component was bone ingrown in allcases. There were no clear signs of loosening,osteolysis, or subsidence for either the femoral oracetabular component. The abduction angle of theacetabular component averaged 35.6° (range, 26°-42°), and the anteversion angle of the acetabularcomponent averaged 14.5° (range, 13°-17°).

Intraoperative Findings

The periprosthetic soft tissues exhibited extensivemetallosis, and the liners were stained with titaniumdebris. The liners dissociated inferiorly, with almostall circumferential liner rims being damaged. Thedamage to the superior aspects of the rims inparticular was significant. The femoral neckimpinged against the inferior aspect of the disso-ciated liner, creating pressure scars (Fig. 2). Theseimpingements are considered to have occurred afterliner dissociation, arising secondary to fracture ofthe liners' upper rims. No patient had periprostheticosteolysis of the acetabulum or of the proximalfemur. All of the acetabular shells and femoralcomponents were stable. There were no fractures ofthe metal locking tines on the metal shell.

The revision technique consisted of cementing apolyethylene liner into the existing well-fixedHarris-Galante acetabular shell. The femoral headswere replaced because of burnishing. To ensuregood cement adhesion to the inside of the shell,exhaustive debridement was performed; the screwholes were debrided to aid in anchoring the cement.Five anchor holes (3.2 mm in diameter and 2 mm

524 The Journal of Arthroplasty Vol. 23 No. 4 June 2008

deep) were reamed at the back side of the liner. Acement mantle a minimum of 1 mm thick wasdesired between the inside of the shell and the liner.To achieve this, we used a Harris-Galante liner 2mm smaller than the inside diameter of the shell.During a mean follow-up period after revisionsurgery of 6.4 years (range, 4-12 years), weobserved no recurrence of dissociation.

Retrieved Components

The retrieved components were studied so thatwe could try to determine the cause of dissociation.The femoral heads were scratched because ofarticulation of the head with the metal shell. Wearwas observed around almost all of the circumfer-ential rim of the damaged liners, which were stainedblack with titanium debris. Deformation wasobserved in the dissociated bottom aspect of the

Fig. 2. Intraoperative findings in the same patient as inFig. 1: (1) the periprosthetic soft tissues show extensivemetallosis; (2) dissociation of the polyethylene liner fromthe acetabular shell; (3) dissociation of the inferior aspectof the liner and stem neck, causing pressure scars; and(4) no periprosthetic osteolysis of either the acetabulum orthe proximal femur.

Fig 1. Anteroposterior radiograph of one patient's left hip12 years and 6 months after surgery. Dissociation of thepolyethylene liner from the acetabular shell is apparent.The head is positioned eccentrically in the acetabular shell.The acetabular component fixation was stable, the femoralcomponent was bone ingrown, and there was noperiprosthetic osteolysis. There were no fractures of themetal locking tines.

liner as a result of impingement against the stemneck. Similar deformation was observed on thesuperior aspect of the liner (Fig. 3).

Discussion

Harris-Galante porous-coated acetabular compo-nents were introduced in 1984. They have anexcellent survivorship in primary and revisionTHAs [1,2], but dissociation of the polyethyleneliner is an increasingly common complication afterTHA [5-9]. It is believed that numerous factors couldpotentially combine with a weak locking mechan-ism to produce dissociation. Han et al mentioned 2potential causes for liner dissociation in their casereport: One theory is that the polyethylene liner wasnot fixed securely within the metal shell at the timeof the primary THA; the other theory is that thelocking mechanism is not strong enough to hold the

Fig. 3. The dissociated polyethylene liner rim (in the samepatient as in Fig. 1) demonstrates wear, fracture, andseparate fragment from the liner. The femoral head isburnished because of articulation of the head with themetallic shell. Pressure scars caused by impingement ofthe stem neck were observed at the inferior aspect of theliner (black arrows). Similar pressure scars on the femoralhead were also seen in the superior aspect of the liner(white arrows).

Harris-Galante I Acetabular Component � Saito et al 525

liner firmly within the shell [6]. However, because 5patients in this study had a mean 9.6 years ofsurvivorship, there were no cases of tine fracture,casting doubt on the second theory.Gonzalez Della Valle et al presented 3 theories

about the mode of failure [8]: (1) fatigue breakage ofthe titanium-alloy tines; (2) increased micromotionin the shell-liner interface because of polyethylenedegradation, deformation, and wear; and (3) impin-gement of the neck against the liner rim, whichincreases as wear progresses. Broken tines inparticular pose the potential risk of dissociation ofthe polyethylene liner; in 1 patient, who presentedwith a dissociation 6 years after implantation, abroken tine was detected 1 year after implantation.Werle et al reported on 7 cases of liner dissociation.They postulate that dissociation occurs because ofpolyethylene rim wear, rim fragment formation,liner dissociation, and subsequent plastic defor-mation and fatigue fracture of the metal lockingtines [9].Our hypothesis originated from the fact that in

our series, there were no fractures of the metallocking tines. We postulate that either the liner issmaller than the shell, leaving a gap, or that therim of the liner is too thin, which causes fatiguefracture of the liner rim. This failure can beconsidered to reflect poor design of the component.There were no significant differences in abductionangle and anteversion angle as compared with thenondissociation group.

The problem of loss of acetabular bone stock is ofcritical importance in revision surgery for thiscomplication because it entails removal of a well-fixed cementless acetabular shell [9]. Gonzalez DellaValle et al mentioned that in their 18 cases, theytreated this complication in various ways, such asrevision of the shell, exchange of the liner, cementa-tion of a new liner into the shell, and cementation ofan all-polyethylene cup [8]. Tradonsky et al reportedthat in the case of exchange of a polyethylene liner,the fixation power of the liner weakened by 25%even if there was no damage to the locking mecha-nism [13]. Cameron reported a case of dissociation ofa liner 4 months after an elective exchange [14]. Wedid not remove a well-fixed Harris-Galante acetab-ular shell in any of the hips on which we did revisionsurgery, but we did cement the liner and replace thefemoral head because damage was routinelyobserved. In such procedures, it is necessary tomaintain an adequate cement mantle between theliner and the shell. To this end, a polyethylene liner2 mm smaller in outside diameter than the shell'sinside diameter should be used to secure the 1-mmthickness of the cement mantle. Exhaustive debride-ment is to be carried out tomake screw holes serve asan anchor. Five anchor holes are reamed at the backside of the polyethylene liner. After using thismethod in revision surgery, we monitored patientsfor a mean of 6.4 years (range, 4-12 years); we haveseen no recurrence of dissociation.

Conclusion

The rate of dissociation of polyethylene liners inHarris-Galante acetabular components will increasewith time. Two mechanisms of dissociation are thepresence of a gap between the liner and the metalshell and thinness of the liner rim, among others,all of which are attributable to improper design.The other mechanism in fact may be impingement;but we found no significant differences in abduc-tion angle and anteversion angle as compared withour nondissociation group, which makes us hesi-tate to consider impingement to be the main causeof dissociation.

The revision procedure we recommend consists ofcementation of a new polyethylene liner withoutremoving the acetabular shell.

References

1. Clohisy JC, Harris WH. The Harris-Galante porous-coated acetabular component with screw fixation: anaverage ten-year follow-up study. J Bone Joint SurgAm 1999;81:66.

526 The Journal of Arthroplasty Vol. 23 No. 4 June 2008

2. Clohisy JC, Harris WH. Primary hybrid total hipreplacement, performed with insertion of the acet-abular component without cement and a precoatfemoral component with cement. An average ten-year follow-up study. J Bone Joint Surg Am 1999;81:247.

3. Saito S, Ryu J, Oikawa H, et al. Clinical results ofHarris-Galante total hip arthroplasty without cement:follow-up study of over five years. Bull Hosp Joint Dis1997;56:191.

4. Schmalzried TP, Harris WH. Hybrid total hip replace-ment. A 6.5-year follow-up study. J Bone Joint SurgBr 1993;75:608.

5. Retpen JB, Solgaard S. Late disassembly of modularacetabular components: a report of two cases. ActaOrthop Scand 1993;64:193.

6. Han CD, Choe WS, Yoo JH. Late dissociation of thepolyethylene liner from a modular acetabular metalshell after primary total hip arthroplasty—a report offive cases. Yonsei Med J 1998;39:277.

7. Louwerse RT, Heyliners IC. Late failure of thepolyethylene liner fixation in an uncemented totalhip arthroplasty. J Arthroplasty 1999;14:391.

8. Gonzalez Della Valle A, Ruzo PS, Li S, et al. Dislodg-ment of polyethylene liners in first- and second-

generation Harris-Galante acetabular components.J Bone Joint Surg Am 2001;83:553.

9. Werle J, Goodman S, Schurman D, et al. Polyethyleneliner dissociation in Harris-Galante acetabular com-ponent: a report of 7 cases. J Arthroplasty 2002;17:78.

10. Harris WH. Traumatic arthritis of the hip afterdislocation and acetabular fracture: treatment bymold arthroplasty. An end-result study using a newmethod of result evaluation. J Bone Joint Surg Am1969;51:737.

11. Tompkins GS, Jacobs JJ, Kull LR, et al. Primary totalhip arthroplasty with a porous-coated acetabularcomponent. Seven-to-ten–year results. J Bone JointSurg Am 1997;79:169.

12. Engh CA, Massin P, Suthers KE. Roentgenographicassessment of the biologic fixation of porous-surfacedfemoral components. Clin Orthop Relat Res 1990;257:107.

13. Tradonsky S, Postak PD, Froimson AI, et al. Acomparison of the dissociation strength of modularacetabular components. Clin Orthop Relat Res 1993;296:154.

14. Cameron HU. Dissociation of a polyethylene linerfrom an acetabular cup. Orthop Rev 1993;22:1160.