Squeaking ceramics bearing surfaces have been recently recognised as a problem in total hip arthroplasty. The position of the acetabular cup has been alluded to as a potential cause of the squeaking, along with particular combinations of primary stems and acetabular cups. This study has used the finite element method to investigate the propensity of a new large diameter preassembled ceramic acetabular cup to squeaking due to malpositioning.

A verified three-dimensional FE model of a cadaveric human pelvis was developed which had been CT scanned, and the geometry reconstructed; this was to be used to determine the behaviour of large diameter acetabular cup system with a thin delta ceramic liner in the acetabulum. The model was generated using ABAQUS CAE pre-processing software. The bone model incorporated both the geometry and the materials properties of the bone throughout based on the CT scan. Finite element analysis and bone material assignment was performed using ABAQUS software and a FORTRAN user subroutine. The loading applied simulated edge loading for rising from a chair, heel-strike, toe off and stumbling.

All results of the analysis were used to determine if the liner separated from the shell and if the liner was toggling out of the shell. The results were also examined to see if there was a propensity for the liner to demobilise and vibrate causing a squeaking sound under the prescribed loading regime.

This study indicates that there is a reduction in contact area between the ceramic liner and titanium shell if a patient happens to trip or stumble. However, since the contact between the liner and the shell is not completely lost the propensity for it to squeak is highly unlikely.

Squeaking in ceramic on ceramic bearing total hip arthroplasty is well documented but its aetiology is poorly understood. In this study we have undertaken an acoustic analysis of the squeaking sound recorded from 31 ceramic on ceramic bearing hips. The frequencies of these sounds were compared with in vitro acoustic analysis of the component parts of the total hip implant. Analysis of the sounds produced by squeaking hip replacements and comparison of the frequencies of these sounds with the natural frequency of the component parts of the hip replacements indicates that the squeaking sound is due to a friction driven forced vibration resulting in resonance of one or both of the metal components of the implant. Finite element analysis of edge loading of the prostheses shows that there is a stiffness incompatibility between the acetabular shell and the liner.

The shell tends to deform, uncoupling the shell-liner taper system. As a result the liner tends to tilt out of the acetabular shell and slide against the acetabular shell adjacent to the applied load. The amount of sliding varied from 4–40μm. In vitro acoustic and finite element analysis of the component parts of a total hip replacement compared with in vivo acoustic analysis of squeaking hips indicate that either the acetabular shell or the femoral stem can act as an “oscillator’ in a forced vibration system and thus emit a squeak.

Introduction: Squeaking has long been recognized as a complication in hip arthroplasty. It was first reported in the Judet acrylic hemiarthroplasty.¹ It was the squeak of a Judet prosthesis that led John Charnley to investigate friction and lubrication of normal and artificial joints which ultimately led to the concept of low friction arthroplasty. Ceramic on ceramic bearings were pioneered by Boutin in France during the 1970’s, but experienced unacceptably high fracture rates. Charnley demonstrated in vitro squeaking when he tested one of Boutin’s ceramic-on-ceramic bearings in his pendulum friction comparator.² Squeaking has also been reported in other hard on hard bearings, and can also occur after polyethylene bearing surface failure resulting in articulation between metal on metal or ceramic on metal surfaces.^3–6 Recently, squeaking has been increasingly reported in modern ceramic-on-ceramic bearings in hip arthroplasty. However, although well-documented, the aetiology of squeaking in ceramic on ceramic bearings is still poorly understood. The incidence ranges from under 1% to 10%.^7–10 It has been reported in mismatched ceramic couples,11and after ceramic liner fracture.^12,13 An increased risk of squeaking has been demonstrated with acetabular component malposition, as well as in younger, heavier and taller patients.⁹ However, it may also occur in properly matched ceramic bearings with ideal acetabular component position and in the absence of neck to rim impingement.^7–9 In rare cases, the squeak is not tolerated by the patient and has prompted a revision.

Under ideal conditions hard-on-hard bearings are assumed to be operating under conditions of fluid film lubrication with very low friction.^14,15 However, if fluid film lubrication breaks down leading to dry sliding contact there will be a dramatic increase in friction. If this increased friction provides more energy to the system than it can dissipate, instabilities may develop in the form of friction induced vibrations and sound radiation¹⁶. Friction induced vibrations are a special case of forced vibration, where the frequency of the resulting vibration is determined by the natural frequency of the component parts. Running a moistened finger around the rim of a wine glass is an example of this. [Appendix].

The hypothesis of this study is that the squeaking sound that occurs in ceramic on ceramic hip replacement is the result of a forced vibration. This forced vibration can be broken down into a driving force and a resultant dynamic response¹⁷. The driving force is a frictional driving force and occurs when there is a loss of fluid film lubrication resulting in a high friction force^14,15,18. The dynamic response is a vibration of a part of the device (the oscillator) at a frequency that is influenced by the natural frequency of the part¹⁶. By analyzing the frequencies of the sound produced by squeaking hip replacements and comparing them to the natural frequency of the component parts of a hip replacement this study aims to determine which part produces the sound.

Materials and methods: In vitro determination of the natural frequencies of implant components Modal analysis has suggested that resonance of the ceramic components would occur only at frequencies above the human audible range and that resonance of the metal parts would occur at frequencies within the human audible range. Furthermore, that resonance of the combined ceramic insert and titanium shell would not be within the human audible range. To test this hypothesis we performed a simple acoustic analysis. The natural frequency of hip replacement components was determined experimentally using an impulse-excitation method (Grindo-sonic). Components were placed on a soft foam mat in a quiet environment and struck with a wooden mallet. The sound emitted from the component was recorded on a personal computer with an external microphone with a frequency response which ranges from 50Hz to 18,000Hz (Beyerdynamic MCE87, Heilbronn, Ger-many). The computer has an integrated sound card with a frequency response from 20Hz to 24kHz (SoundMAX integrated digital audio chip, Analogue Devices Inc, Norwood, M.A.) and we used a codec with a frequency response from 20Hz to 20kHz (Audio Codec ’97, Intel, Santa Clara, CA). Sound files were captured as 16 bit mono files at a sample rate of 48000Hz using acoustic analysis software (Adobe Audition 1.5, Adobe Systems Incorporated, San Jose, California, USA). We performed fast Fourier transform (FFT) of the sound using FFT size 1024 with a Blackmann-Harris window to detect the frequency components of the emitted sound. (Fast Fourier transform is an accepted and efficient algorithm which enables construction of a frequency spectrum of digitized sound).

We tested the following components: modular ceramic/titanium acetabular components, which included testing the titanium shell and the respective ceramic inserts both assembled according to the manufacturer’s instructions and unassembled; titanium femoral stems and ceramic femoral heads both assembled and unassembled. A range of sizes of each component was tested according to availability from our retrieval collection.

In vivo acoustic analysis: Sound recordings were collected from 31 patients. Nineteen recordings were made at our institution: 16 of these were video and audio recordings and 3 were audio only recordings. Video recording was with a digital video camera recorder (Sony DCR-DVD101E Sony Electronics, San Diego, CA, USA) with the same external microphone used in the in vitro analysis. For 3 patients who could not reproduce the sound in the office we lent them a digital sound recorder for them to take home and record the sound when it occurred (Sony ICD-MX20, Sony Electronics, San Diego, CA, USA). This device has a In vivo acoustic frequency range from 60Hz to 13,500Hz. The remainder of the recordings were video and audio recordings made by surgeons at three other institutions on digital video camera recorders.

Sound files were captured and analyzed by the same method used in the in vitro analysis. Each recording was previewed in the spectral view mode which allows easy visual identification of the squeak in the sound recording. In addition all sound recordings were played, listening for the squeak. Once a squeak was identified a fast Fourier transform (FFT) was performed. We used FFT size 1024 with a Blackmann-Harris window which allowed us to easily pick out the major frequency components. All prominent frequency components were recorded at the beginning of the squeak and at several time points during the squeak if there was any change. A range was recorded for the fundamental frequency component. We were able to determine the frequency range of the recording device used by observing the frequency range of the background noise on the recording. We found that if a squeak was audible on the recording we had no difficulty determining its frequency regardless of the quality of the device used to make the recording or the amount of background noise.

The mean age of the patients was 54 years (23 to 79 years), mean height was 171cm (152 to 186cm) and mean weight was 79kg (52 to 111kg). There were 17 female and 14 male patients. There were nineteen ABGII stem and ABGII cup combinations, 10 accolade stem and trident cup, 1 Exeter stem and trident cup and 1 Osteonics Securfit stem with an Osteonics cup. Ethics committee approval was obtained for this project from our institution and from the referring institutions and informed consent was gained from the patients.

Finite element analysis of edge loading: Edge-loading wear which may provide a mechanism for failure of fluid film lubrication and may therefore play a role in squeaking. To evaluate edge loading further we conducted finite-element analysis (FEA).⁹ Computed tomography (CT) scans of an intact pelvis were obtained from visual human data set (VHD, NLM, Bethesda, Maryland). Slices were taken at 1mm thick with no inter-slice distance through the entire pelvis. The CT files were then read into a contour extraction program and saved into an IGES file format which was imported into PATRAN (MSC Software, Los Angeles, CA) to develop the pelvic geometry. The pelvis was meshed with 10 noded modified tetrahedral elements. The model was reconstructed with a 54mm titanium alloy generic acetabular shell and a 28mm alumina ceramic liner. The acetabular shell and ceramic liner were meshed using 8 noded hexahedral elements. The shell-liner modular taper junction incorporated an 18° angle. The implant contact conditions (Lagrangian multiplier) allowed the liner and shell to slide with a friction coefficient of 0.9. Tied contact conditions were applied between the generic acetabular shell and the bone representing bone ongrowth. Bone material properties were extracted from the CT files by taking the Hounsfield value and the coordinates and mapping to the element in the model allowing us to calculate the Young’s modulus for each element ¹⁹. Material properties for the shell and liner were based on published values²⁰ for titanium alloy and alumina ceramic

Introduction: Squeaking is reported ceramic-on-ceramic hip bearings in association with acetabular component malposition – particularly too much or too little anteversion. Acoustic analysis of squeaking hips with modular ceramic-titanium acetabular components suggests that there may be dynamic uncoupling of the ceramic insert from the titanium shell with edge loading of the ceramic. The aim of this study was to investigate edge loading of a modular ceramic-titanium acetabular component during gait at different positions of anteversion using the finite element (FE) method.

Methods: An intact and reconstructed 3D FE model of a human pelvis was generated using PATRAN. Bone properties extracted from the CT data were applied using FORTRAN subroutines. A generic acetabular titanium shell and ceramic liner were modelled and placed in the pelvis in two different positions: ideal anteversion and 18 degree excess anteversion. The contact conditions simulated a fully osseointegrated acetabular shell and a matched taper junction with a friction coefficient of 0.2. We ran FE analysis with ABAQUS software to determine the stress distribution and surface separation of shell and liner at toe-off.

Results: The separation distance between the ceramic liner and the acetabular shell for the anteverted component (40mm) was an order of magnitude greater than that for the ideally positioned component (4mm). There was “tilting” of the ceramic liner out of the acetabular shell in both cases.

Discussion: Based on clinical observations, the toe-of phase of gait is a common position for squeaking to occur. Clinical retrievals also show evidence of edge loading wear and contralateral taper interface separation with the “tilting” of the liner out of the acetabular shell. It is envisaged that the “tilting” of the liner in the acetabular shell may allow forced vibrations associated with the squeaking phenomena, possibly in combination with edge loading.

Introduction: Bone resorption at the bone-implant interface is still a problem, leading to pain, poor function and the possibility of bone fracture. This loss of supporting bone tissue is due to resorption and impaired bone formation. Loosening of an implant is often not clinically or radiographically apparent for 8–10 years. It would be beneficial if these potential failures could be identified early so that revision surgery can be avoided. The aim of this study was to investigate the influence of implant material property changes and its influence on the trabecular loading patterns of the underlying supporting bone structure.

Methods: An intact and reconstructed 3D finite element (FE) model of a human femur was developed. The model was generated using PATRAN and CT scans. This was used to determine the stress, strain and interface sliding of a knee implant at heel-strike and stair climbing phases of gait. FE analysis of the model was performed using ABAQUS software. The materials properties of the bone were extracted from the CT data and applied using FORTRAN subroutines. Implant-bone interfaces were simulated using cementless fixation concepts. Sliding contact conditions were applied to simulate the immediate post-operative period.

Results: Three material property cases were analysed, with respect to the intact bone, at 100%, 25% and 2.5% of cobalt chrome’s (CoCr) Youngs modulus. At heel-strike, for the 100% case, higher stress was found at anterior flange while lower stress dominated around the pegs and intercondylar notch. For the 25% case, lower stresses were found in the intercondylar notch and higher stresses above the pegs. For the 2.5% case, stresses resembled that of intact bone, higher stresses were found above the pegs and lower stress in the intercondylar notch. In stair-climbing, for the 100% case, lower stresses were found around the pegs and in the intercondylar notch. For the 25% case, lower stresses were found in the intercondylar notch and higher stresses in areas above the pegs. For the 2.5% case, higher stresses were found at the distal condyles and lower stresses were observed in the intercondylar notch.

Discussion: The analysis presented changes in the trabecular loading and subsequently resulted in stress shielding. The general trend showed that the majority of stress shielding is occurring at the posterior flange and medial condyle while increased trabecular loading occurred at the anterior flange and lateral condyle regions. As the stiffness of the implant decreases from 100% to 25%, the differences in trabecular loading are extremely small. Both these implant material properties are very stiff in comparison to the underlying trabecular bone. However, as CoCr stiffness is decreased to 2.5% this yields a more homogenous stress distribution at the contact interfaces.

Introduction: The process of impacting cemented hip resurfacing components may, in part, be associated with femoral neck fracture. The impaction process may introduce fractures due to the impact shock wave passing through the bone during the setting of the implant and achieving a completely seated position. The aim of this study was to measure the impaction loads during hip resurfacing surgery and correlate the measured loads to theoretical calculations.

Methods: Following ethical approval 3 patients have been enrolled out of 24 patients in a pilot study. A surgical mallet was manufactured and instrumented with a calibrated impact load cell. During the impaction procedure the impact loads are recorded to a laptop using Labview software. An Excel spreadsheet has been written using the finite difference method to calculate the impact loads based on a mass (hammer, impactor and implant) and spring system (compression only) defining each part of the surgical instrumentation used to impact the resurfacing component onto the femoral head.

Results: Clinically, upto 19 impacts are used to seat the resurfacing implant onto the femoral head. Loads upto 24kN were recorded. The finite difference model was calibrated to the clinical measurements. The Pearson’s R correlation coefficient for the net force on the mallet was 0.91 and for the impulse was 0.98

Discussion: This study has investigated the clinical impaction loads imparted onto an implant during resurfacing surgery and developed a finite difference model of the process. The finite difference approach can be used to better understand the loads applied to not only the implant, but the underlying bone. This may, in part, give the surgeon a better understanding as to whether the bone has been predisposed to fracture following the high impact loads and thereby affecting the long-term integrity of the joint replacement.