The PDF of the article you requested follows this cover page.  

This is an enhanced PDF from The Journal of Bone and Joint Surgery

2010;92:550-557.  doi:10.2106/JBJS.H.01326 J Bone Joint Surg Am.Camilo Restrepo, Zachary D. Post, Brandon Kai and William J. Hozack    

Ceramic-on-Ceramic Bearing Total Hip ArthroplastyThe Effect of Stem Design on the Prevalence of Squeaking Following

This information is current as of December 30, 2010

Reprints and Permissions

Permissions] link. and click on the [Reprints andjbjs.orgarticle, or locate the article citation on

to use material from thisorder reprints or request permissionClick here to

Publisher Information

www.jbjs.org20 Pickering Street, Needham, MA 02492-3157The Journal of Bone and Joint Surgery

The Effect of Stem Design on thePrevalence of Squeaking Following

Ceramic-on-Ceramic BearingTotal Hip Arthroplasty

By Camilo Restrepo, MD, Zachary D. Post, MD, Brandon Kai, BS, and William J. Hozack, MD

Investigation performed at the Thomas Jefferson University Hospital, Rothman Institute, Philadelphia, Pennsylvania

Background: The ceramic-on-ceramic bearing for total hip arthroplasty has an extremely low wear rate and demon-strates minimal inflammatory response in comparison with other bearing choices. However, acoustic emissions such assqueaking and clicking are being reported as annoying complications related to its use. The cause or causes of thisphenomenon have not been determined. The purpose of the present study was to evaluate the possibility that designaspects of the femoral component may be a contributing factor to the etiology of squeaking associated with the ceramic-on-ceramic bearing total hip arthroplasty.

Methods: We retrospectively reviewed 266 consecutive patients (304 hips) who had undergone total hip arthroplastywith use of ceramic-on-ceramic bearings. The first 131 consecutive patients (152 hips) (Group 1) received a hydroxyapatite-coated stem composed of titanium-aluminum-vanadium alloy with a C-taper neck geometry and robust midsection withan anteroposterior diameter of 13 mm. The second 135 consecutive patients (152 hips) (Group 2) also received ahydroxyapatite-coated stem, but in that group the stem was composed of titanium-molybdenum-zirconium-iron alloy, witha V-40 neck geometry and a midsection with an anteroposterior thickness of only 10 mm. All 304 hips received the samecup, composed of titanium-aluminum-vanadium alloy. Demographic characteristics, such as age, sex, height, weight, andbody mass index, were similar in both groups. Data regarding the presence of squeaking were obtained prospectively.Patients who were seen for clinical follow-up either expressed the squeaking phenomenon themselves or were askedabout it by the physician. Patients who were not seen at a recent clinical follow-up visit were contacted by telephone andwere asked specifically about squeaking that might be associated with the hip replacement. Only patients with confirmedsqueaking noise were included in the present study. Postoperative radiographs, the Short Form-36 health survey, theHarris hip score, and office or telephone interviews of the patient were used to determine the overall outcome of theprocedure.

Results: The prevalence of squeaking was seven times higher for patients who received the titanium-molybdenum-zirconium-iron-alloy stem (twenty-seven patients, twenty-eight hips [18.4%]) than in those who received the titanium-aluminum-vanadium-alloy stem (three patients, four hips [2.6%]); this difference was significant (p < 0.0001).

Conclusions: Our study suggests that different stem alloys, stem geometries, or neck geometries can have an impacton the frequency of squeaking following a ceramic-on-ceramic total hip arthroplasty.

Level of Evidence: Therapeutic Level III. See Instructions to Authors for a complete description of levels of evidence.

Ceramic-on-ceramic bearings have many advantages foruse in total hip arthroplasty. The hardness of ceramicmakes it more wettable and provides a smoother sur-

face than other bearings do. This results in wear rates as manyas twenty times lower than those for metal-on-metal articu-lations in vitro1. In addition, particle debris from ceramic-on-

Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or grants inexcess of $10,000 from Stryker Orthopaedics. Neither they nor a member of their immediate families received payments or other benefits or acommitment or agreement to provide such benefits from a commercial entity.

550

COPYRIGHT � 2010 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED

J Bone Joint Surg Am. 2010;92:550-7 d doi:10.2106/JBJS.H.01326

ceramic hips does not generate the magnitude of inflammatoryresponse that polyethylene does2. The particles are also morebio-inert than those from metal-on-metal articulations. De-spite these advantages, ceramic-on-ceramic bearings have hada mixed experience since their introduction in the early 1970s.Early designs were plagued by catastrophic failure and werelargely abandoned3. Improvements in design and ceramic qualityled to a renewed interest in the 1990s and early 2000s.

However, enthusiasm for ceramic-on-ceramic bearings hasbeen tempered recently by reports of noise, specifically, squeak-ing. All hard-on-hard bearings have been associated with noise4,but, unlike the noise associated with metal-on-metal articula-tions, ceramic-on-ceramic squeaking does not seem to abatewith time. The prevalence of squeaking has ranged widelyamong investigators. Capello and colleagues5 reported a rateof 1%, whereas Jarrett and colleagues6 reported a prevalence of11.4%. A recent study from Denmark7 demonstrated a rate of20%. Data collected at the Rothman Institute8 demonstrated 3%.

The etiology of ceramic-on-ceramic squeaking has beenelusive. Our hypothesis is that the audible squeak associatedwith ceramic-on-ceramic bearings is multifactorial but that animportant contributing factor to squeaking involves specificdesign characteristics of the femoral component.

Materials and Methods

Institutional review board approval was obtained for thepresent study. We retrospectively reviewed 266 consecutive

patients (304 hips) who were managed with ceramic-on-ceramic total hip arthroplasty between 1998 and 2004. Allpatients had a diagnosis of degenerative joint disease of the hipand underwent total hip arthroplasty through a modifiedHardinge approach. All procedures were performed by thesenior author (W.J.H.) at a single institution.

Fig. 1

The Omnifit stem,

made of titanium-

aluminum-vanadium

alloy. This stem has a

hydroxyapatite coating,

a C-taper neck geome-

try, and a robust mid-

section, with an

anteroposterior diame-

ter of 13 mm.

Fig. 2

The Trident cup is made of titanium-aluminum-

vanadium alloy and accepts a ceramic liner within a

commercially pure titanium sleeve.

551

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

Patients fell into one of two groups related to the date ofsurgery. Group 1 included the first 131 consecutive patients(152 hips), who were managed from 1998 to 2002. All patientsin this group received an Omnifit stem (Stryker Orthopaedics,Mahwah, New Jersey), which is composed of a titanium-aluminum-vanadium alloy (Fig. 1). The Omnifit stem has ahydroxyapatite coating, a C-taper neck geometry, and a robustmidsection with an anteroposterior diameter of 13 mm (Fig.1). All of these patients received a Trident cup (StrykerOrthopaedics). The Trident cup is a titanium-aluminum-vanadium alloy and accepts a ceramic liner within a com-mercially pure titanium sleeve (Fig. 2). This specific design hasan elevated titanium rim, which may increase the possibilityof impingement and potentially could be an additional sourceof squeaking.

Group 2 included the second 135 consecutive patients(152 hips), who were managed from 2002 to 2004. All patientsin Group 2 were managed with an Accolade stem (StrykerOrthopaedics) and the same Trident cup that was used inGroup 1. The Accolade stem is made of a titanium-molybdenum-zirconium-iron alloy, with a V-40 taper neck geometry and amidsection with an anteroposterior thickness of only 10 mm (Figs.3 and 4).

Demographic data, including age, sex, height, weight,and body mass index, were recorded prospectively for all pa-tients (Table I).

Intraoperative data, including prosthesis type, stem size,head size, and liner size, were also recorded. All patients weremanaged with either a 28, 32, or 36-mm ceramic head. Headsize was based on the maximum allowable size for the im-planted cup. In Group 1, thirty-four hips (22.4%) received a28-mm head, ninety-two hips (60.5%) received a 32-mm head,and twenty-six hips (17.1%) received a 36-mm head. In Group

Fig. 3

The Accolade

stem is made of

a titanium-

molybdenum-

zirconium-iron al-

loy. This stem

has a hydroxyap-

atite coating, a

V-40 taper neck

geometry, and a

midsection with

an anteroposte-

rior thickness of

only 10 mm.

Fig. 4

Close-up photograph showing the neck geometry of the

Omnifit (A) on the left and the Accolade (B) on the right.

552

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

2, twenty-seven hips (17.8%) received a 28-mm head, seventy-nine hips (52.0%) received a 32-mm head, and forty-six hips(30.3%) received a 36-mm head (Table II).

Postoperative anteroposterior and lateral radiographswere reviewed to assess the position of the acetabular com-ponent9. With use of OrthoView software (Meridian Tech-nique, Southampton, Hampshire, United Kingdom), theanteroposterior radiographs were uploaded, the images werecalibrated, and the position of the acetabular component wasassessed with use of the inclination-anteversion tool from thesystem. This evaluation was done by an orthopaedic surgeon(C.R.) and an adult reconstruction fellow (Z.D.P.). Cup posi-tion was assessed with regard to abduction, version, andheight. The stem was evaluated with regard to position andoffset.

Outcome measurements, specifically, the Physical Healthand Mental Health dimensions of the Short Form-36 (SF-36)Health Survey as well as the Harris hip score, were obtainedpreoperatively and at the time of the latest follow-up for allpatients.

Data regarding the presence of squeaking were obtainedprospectively. The 247 patients (282 hips; 92.8%) who wereseen for clinical follow-up either expressed the squeakingphenomenon themselves or were asked about it by the phy-sician. The 19 patients (twenty-two hips; 7.2%) who were notseen at a recent clinical follow-up visit were contacted bytelephone and were asked specifically about noises that mightbe associated with the hip replacement. If squeaking wasconfirmed, the patient was asked to characterize its intensity(slightly, moderately, or highly audible), frequency (continu-ous, intermittent, or a single event), and the time between thedate of the operation and the date of onset. Patients who hadonly clicking were not included in the study.

Complications, including infection, loosening, and re-operation for any reason, were recorded.

Statistical AnalysisDemographic data as well as prosthetic femoral head size andneck length were analyzed and compared between groups of

patients and between squeaking and nonsqueaking hips. Allstatistical tests were performed with alpha set at 0.05. The t testwas performed to evaluate continuous variables (age, height,weight, body mass index, and radiographic findings), and theFisher exact test was used to evaluate nominal values (sex, headsize, and neck length). A baseline logistic regression modeladjusting for the covariates of age, sex, and body mass indexwas performed. We tested confounding by adding severalfemoral variables to the baseline model individually and thentested whether it resulted in a change of odds ratio by 10%.

Source of FundingThere was no direct external funding for this study, althoughone of the authors is a consultant for Stryker Orthopaedics.

Results

The two groups did not differ significantly with regard toage, sex, height, weight, or body mass index (Table I).

On the postoperative radiographs, cup abduction, cupanteversion, cup medialization, and femoral offset were notsignificantly different between the two groups (Table III). Al-though the values for cup abduction and anteversion werehigh, cup malpositioning alone has been demonstrated not tocorrelate with squeaking8.

The only measured parameter that showed a significantdifference between the two groups was the prevalence of

TABLE I Demographic and Follow-up Data

VariableGroup 1

(152 Hips)Group 2

(152 Hips) P Value

Age*† (yr) 45.1 (18.9 to 62.3) 46.4 (16.5 to 65.5) 0.1893

Male/female ratio‡ (no. of hips) 88 (57.9%)/64 (42.1%) 100 (65.8%)/52 (34.2%) 0.1939

Height*†§ (in) 68.1 (58 to 78) 68.3 (58 to 88) 0.7505

Weight*†# (lb) 181.6 (95 to 280) 183.3 (88 to 317) 0.7509

Body mass index*† (kg/m2) 27.3 (15.8 to 41.2) 27.5 (14.3 to 46.2) 0.7958

Duration of follow-up* (yr) 7.1 (5.7 to 9.8) 4.9 (4.5 to 6.1)

*The values are given as the mean, with the range in parentheses. †Evaluated with the two-tailed paired t test. ‡Evaluated with the two-tailedFisher exact test. §1 in = 2.54 cm. #1 lb = 0.45 kg.

TABLE II Head Size Distribution*

Head SizeGroup 1

(152 Hips)Group 2

(152 Hips) P Value

28 mm 34 (22.4%) 27 (17.8%) 0.3903

32 mm 92 (60.5%) 79 (52.0%) 0.1652

36 mm 26 (17.1%) 46 (30.3%) 0.0101

*Evaluated with the two-tailed Fisher exact test.

553

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

squeaking. Three patients (four hips; 2.6%) reported squeak-ing in Group 1. The average time of onset in this group was16.5 months (range, 1.2 to sixty-nine months) postopera-tively. In contrast, twenty-seven patients (twenty-eight hips;18.4%) reported squeaking in Group 2 (p < 0.0001). Theaverage time of onset in this group was 13.2 months (range,0.7 to seventy-seven months) postoperatively. The frequencyand severity of squeaking did not change through the studyperiod for this specific cohort. Of the thirty-two hips with

squeaking, six (all from Group 2) were revised to a polyethyleneliner, with complete resolution of the noise in all six cases. Afterlogistic regression, none of the variables—age, sex, body massindex, head size, stem offset—showed a change of 10% in theodds ratio. After adjusting for the covariates, we found anassociation between the group and squeaking (p = 0.0001)(Table IV).

Compared with preoperative baseline values, both groupsexhibited significantly increased values on the SF-36 ques-

TABLE III Postoperative Radiographic Findings*

VariablesGroup 1

(152 Hips)Group 2

(152 Hips) P Value

Cup abduction† (deg) 40.9 (29.3 to 53) 37.5 (23 to 52) 0.0569

Cup anteversion† (deg) 43.3 (28 to 67) 46.9 (29 to 62.5) 0.9929

Cup medialization† (mm) 3.9 (–10 to 13) 1.9 (–11 to 11) 0.0552

Femoral offset† (mm) 5.3 (–13 to 13) 5.5 (–8 to 19) 0.8475

*Evaluated with the two-tailed paired t test. †The values are given as the mean, with the range in parentheses.

TABLE IV Logistic Regression Model Among Squeaking and Nonsqueaking Hips*

Variable

Symptom

Statistics P ValueSqueaking(N = 32)

Nonsqueaking(N = 272)

Age† (yr) 47.42 ± 9.51 45.86 ± 9.18 T = –0.88 0.39

Sex (no. of hips) Chi square = 0.09 0.76

Male 19 (59%) 169 (62%)

Female 13 (41%) 103 (38%)

Body mass index†(kg/m2)

26.54 ± 3.69 27.53 ± 4.88 T = 1.38 0.18

Head size (no. of hips) Chi square = 4.91 0.03

28 mm 5 (16%) 56 (21%)

32 mm 13 (41%) 158 (58%)

36 mm 14 (44%) 58 (21%)

Stem offset (no. of hips) Chi square = 1.41 0.23

Standard 25 (78%) 234 (86%)

Lateral 7 (22%) 38 (14%)

Group (no. of hips) Chi square = 20.05 <0.0001

1 4 (13%) 148 (54%)

2 28 (88%) 124 (46%)

*The baseline logistic regression model adjusted for covariates of age, sex, and body mass index. We tested confounding by adding severalfemoral variables to the baseline model individually and tested whether it resulted in the change of odds ratio by 10%. None of the variablesshowed a change of 10% in the odds ratio. After adjusting for the covariates, we found an association between the group and squeaking (standarderror = 0.56, odds ratio = 8.50, 95% confidence interval = 2.85 to 25.42, p = 0.0001). †The values are given as the mean and the standarddeviation.

554

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

tionnaire in terms of both the Physical Health score (withincreases of 45 points in Group 1 and 47 points in Group 2)and the Mental Health score (with increases of 35 points inGroup 1 and 37 points in Group 2), but there was no differencebetween the two groups (p = 0.956 and p = 0.987, respec-tively). As expected, the Harris hip score was better after sur-gery in both groups in comparison with the baseline values(with improvement of 27 points in Group 1 and of 25 points inGroup 2), but again there was no significant difference be-tween the two groups (p = 0.893).

All patients, including the ones who experiencedsqueaking, were satisfied with the result of total hip arthroplasty(except for the squeaking itself). No other orthopaedic andnon-orthopaedic complications were found in either group.

Discussion

Adefinitive etiology for squeaking in ceramic-on-ceramic-bearing hips has been elusive and controversial. Factors

such as cup position and patient age, height, and weight haveall been implicated as being associated with audible squeak-ing10. However, other reports and our own data have shown noassociation with these same factors8. The current study sug-gests a new explanation. We found a clear relationship betweenthe prevalence of squeaking and the type of femoral compo-nent implanted. Patients in our series who were managed witha thicker femoral component with a C-taper neck and a stemmade of titanium-aluminum-vanadium were seven times lesslikely to report squeaking than those with a thinner stem com-ponent with a V-40 taper neck made of a titanium-molybdenum-zirconium-iron alloy.

The current explanation for ceramic-on-ceramic squeakingis based on the phenomenon of stripe wear. Walter et al.11 first

reported finding stripe wear in 2004. In a retrieval analysis,those investigators found a heavy wear pattern in a location ofthe head that suggested wear during activities at the extremesof motion. Specifically, their analysis showed that this wear didnot usually occur during normal walking but rather occurredduring rising from a seated position or other high-flexionactivities. During these activities at the extremes of motion,edge loading between the ceramic head and the posterior rimof the ceramic cup occurred. The investigators concluded thatit was these edge-loading activities that led to the visible wearstripe.

In a retrieval analysis, Lusty et al. reported that twelveof twelve squeaking ceramic-on-ceramic bearings had evi-dence of edge-loading wear, suggesting that a wear stripe is likelyinvolved in the production of squeaking12. As a ceramic-on-ceramic bearing passes over the stripe, it could generate avibration that produces an audible sound (Fig. 5). This ex-planation is consistent with the fact that ceramic-on-ceramicsqueaking does not occur until an average of fourteen toeighteen months after surgery8,10. During this time, the wearstripe has not fully developed. Once a wear stripe is estab-lished, squeaking occurs. In addition, the theory that edgeloading creates a wear stripe is consistent with studies thathave shown an increased prevalence of squeaking in patientswith malpositioned acetabular cups10. A cup in excessiveretroversion, for example, would lead to increased edgeloading in flexion and, subsequently, a higher prevalence ofsqueaking.

The basic presumption about squeaking is that the noiseis due directly to vibration of the ceramic bearing. However, astudy by Walter et al. demonstrated that ceramic heads andceramic liners generally resonate at a frequency that is above

Fig. 5

A: Photograph showing stripe wear on an explanted alumina ceramic head. B: Diagram of the stripe wear in the contact

area.

555

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

the upper limit of human hearing (20,000 Hz) and well abovethe in vivo range of squeaking hips (400 to 7500 Hz)13. Thisdiscrepancy poses an obvious question: If ceramic heads andliners do not resonate in a range that can be heard, what iscausing the audible squeaking? Further evaluation of the datafrom the study by Walter et al. showed that a titanium femoralstem vibrates between 2000 and 20,000 Hz, easily within theaudible human range. These data, combined with our findings,suggest that the difference in the rate of squeaking in our twogroups is directly related to the differences in the compositionand design of the femoral components. It is our hypothesisthat there is a breakdown of fluid film lubrication in the hip,with an increase in friction. This generates increased frictionalenergy in the form of a vibration, which is then transmitted toa flexible stem that amplifies the vibration by resonating, re-sulting in an audible sound.

We found no difference between Groups 1 and 2 interms of patient-related parameters. In addition, we found nodifference between the patients with and without squeaking interms of cup or stem position. These findings are in contrast tothose of some published studies10 but are consistent withothers7. All procedures in the present study were performed bythe same surgeon, who has performed thousands of primarytotal hip arthroplasties during his career. The only differencenoted between groups was the prevalence of squeaking.

The stem used in Group 1 was made of a traditionaltitanium-aluminum-vanadium alloy. The neck on that stemhad a C-taper, and the stem itself was more robust proximally.The stem used in Group 2 was made of a titanium-molybdenum-zirconium-iron alloy. That stem had a V-40 neck geometry andwas much more slender proximally. One of the purportedbenefits of the titanium-molybdenum-zirconium-iron mate-rial is that it has a modulus of elasticity that is 30% to 40% lessthan that of titanium-aluminum-vanadium. Therefore, it isa more flexible material even with the same geometry. Fur-thermore, the titanium-molybdenum-zirconium-iron stem inour study was thinner from front to back (10 mm comparedwith 13 mm), thus reducing the bending stiffness in the sagittalplane by an additional 50%. These properties create the po-tential for less thigh pain and better stress distribution to bone,with less proximal bone atrophy. In addition, they make for amuch more flexible stem. The V-40 neck is smaller than theC-taper on the titanium-aluminum-vanadium stem. The de-creased cross section of the V-40 neck is another purportedadvantage of using titanium-molybdenum-zirconium-iron.

The smaller neck diameter should lead to less impingement.However, the smaller diameter of the V-40 neck leads to alower bending stiffness and lower resonant frequency andis more capable of amplifying vibrations generated by theceramic-on-ceramic articulation to cause an audible squeak.The relatively high prevalence of squeaking in Group 2 is con-sistent with this mechanism. In contrast, the thicker C-taperneck in Group 1 has a higher bending stiffness and a highernatural frequency and is therefore less able, in theory, to am-plify vibrational energy into audible resonance.

Three published studies support our proposed theory.Jarrett and colleagues reported that ten (7%) of 143 patientsreported squeaking6. Of note, the stem that was used also had athin profile with a V-40 neck and was composed of titanium-molybdenum-zirconium-iron alloy. Another study, whichdemonstrated a 20% prevalence of squeaking, evaluated adifferent femoral component, but one that was also composedof titanium-molybdenum-zirconium-iron with a V-40 neck14.In contrast, Capello and colleagues reported a 1% prevalenceof squeaking in a study that evaluated a component with aC-taper that was composed of titanium-aluminum-vanadiumalloy5.

In conclusion, the etiology of squeaking of a ceramic-on-ceramic-bearing hip appears to be complicated and mul-tifactorial. Our data show a direct relationship between theprevalence of squeaking and the type of femoral componentused. It is our belief that ceramic-on-ceramic continues to be agood alternative for young patients. Avoiding the use of a V-40neck and titanium-molybdenum-zirconium-iron stem mayreduce substantially the prevalence of squeaking. n

NOTE: The authors thank Dr. William Walter for his invaluable input on the revision of thismanuscript.

Camilo Restrepo, MDZachary D. Post, MDBrandon Kai, BSWilliam J. Hozack, MDDepartment of Orthopaedic Surgery,Thomas Jefferson University Hospital,Rothman Institute, 925 Chestnut Street,5th Floor, Philadelphia,PA 19107.E-mail address for W.J. Hozack:research@rothmaninstitute.com

References

1. Clarke IC, Good V, Williams P, Schroeder D, Anissian L, Stark A, Oonishi H,Schuldies J, Gustafson G. Ultra-low wear rates for rigid-on-rigid bearings in total hipreplacements. Proc Inst Mech Eng H. 2000;214:331-47.

2. Catelas I, Petit A, Marchand R, Zukor DJ, Yahia L, Huk OL. Cytotoxicity andmacrophage cytokine release induced by ceramic and polyethylene particles in vitro.J Bone Joint Surg Br. 1999;81:516-21.

3. Willmann G. Ceramic femoral head retrieval data. Clin Orthop Relat Res.2000;379:22-8.

4. Back DL, Dalziel R, Young D, Shimmin A. Early results of primary Birmingham hipresurfacings. An independent prospective study of the first 230 hips. J Bone JointSurg Br. 2005;87:324-9.

5. Capello WN, D’Antonio JA, Feinberg JR, Manley MT, Naughton M. Ceramic-on-ceramic total hip arthroplasty: update. J Arthroplasty. 2008;23(7 Suppl):39-43.

6. Jarrett CA, Ranawat AS, Bruzzone M, Blum YC, Rodriguez JA, Ranawat CS. Thesqueaking hip: a phenomenon of ceramic-on-ceramic total hip arthroplasty. J BoneJoint Surg Am. 2009;91:1344-9.

556

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y

7. Murphy S, Ecker T, Tannast M. Incidence of squeaking after alumina ceramic-ceramic total hip arthroplasty. J Arthroplasty. 2008;23:327.

8. Restrepo C, Parvizi J, Kurtz SM, Sharkey PF, Hozack WJ, Rothman RH. The noisyceramic hip: is component malpositioning the cause? J Arthroplasty. 2008;23:643-9.

9. Ackland MK, Bourne WB, Uhthoff HK. Anteversion of the acetabular cup. Mea-surement of angle after total hip replacement. J Bone Joint Surg Br. 1986;68:409-13.

10. Walter WL, O’Toole GC, Walter WK, Ellis A, Zicat BA. Squeaking in ceramic-on-ceramic hips: the importance of acetabular component orientation. J Arthroplasty.2007;22:496-503.

11. Walter WL, Insley GM, Walter WK, Tuke MA. Edge loading in third generationalumina ceramic-on-ceramic bearings: stripe wear. J Arthroplasty. 2004;19:402-13.

12. Lusty PJ, Watson A, Tuke MA, Walter WL, Walter WK, Zicat B. Wear and ace-tabular component orientation in third generation alumina-on-alumina ceramicbearings: an analysis of 33 retrievals [corrected]. J Bone Joint Surg Br. 2007;89:1158-64. Erratum in: J Bone Joint Surg Br. 2008;90:543.

13. Walter WL, Waters TS, Gillies M, Donohoo S, Kurtz SM, Ranawat AS, HozackWJ, Tuke MA. Squeaking hips. J Bone Joint Surg Am. 2008;90 Suppl 4:102-11.

14. Keurentjes JC, Kuipers RM, Wever DJ, Schreurs BW. High incidence ofsqueaking in THAs with alumina ceramic-on-ceramic bearings. Clin Orthop RelatRes. 2008;466:1438-43.

557

TH E J O U R N A L O F B O N E & JO I N T SU R G E RY d J B J S . O R G

VO LU M E 92-A d NU M B E R 3 d M A R C H 2010EF F E C T O F ST E M DE S I G N O N PR E VA L E N C E O F SQ U E A K I N G

FO L LO W I N G CE R A M I C - O N -CE R A M I C HI P A RT H R O P L A S T Y