Explant analysis of the Biomet Magnum/ReCap metal-on-metal hip joint

Joints investigated

Between July 2006 and September 2011, 96 patients (110 hips) received the Biomet Magnum/ ReCap/ Taperloc large head MoM THA or the Biomet ReCap resurfacing hip joint as part of the “Magnum Total Hip System International Prospective Data Collection” study at North Bristol NHS Trust (Musculoskeletal Clinical Study 1939). Five patients (six hips) died but the cause of death was unrelated to the hip replacement. Ten patients (11 hips) were lost to follow-up, leaving 93 hips for analysis. A total of 19 patients (20 hips) received the resurfacing joint and 62 patients (73 hips) received the Magnum/ ReCap/ Taperloc THA. There were 38 females (46 hips) and 43 males (47 hips) in the cohort. The mean age at surgery was 57.9 years (26.8 to 73.8) and mean follow-up was 7.1 years (3.7 to 9.2). Eight hips (seven patients) were revised. Two early revisions were performed within the first month, one for a fractured neck of the femur in a resurfacing patient and one for early acetabular component rotation. Six revisions were performed between 31 and 92 months after index surgery; primary diagnosis was osteoarthritis. Written informed consent for retrieval of implants at time of revision was requested from the patient prior to any surgery. This complied with the European Community Regulations. The implant revised at 31 months was not available for investigation in this study.

Five cementless Biomet Magnum/ ReCap/ Taperloc large head MoM THAs and one Biomet ReCap resurfacing hip joint were investigated in this study (examples shown in Fig. 1). The resurfacing joint and four of the five THAs (THA Magnum 1 to 4: Table I) formed part of the prospective Bristol ReCap study. THA Magnum 1 and THA Magnum 3 were sequential bilateral revisions from the same patient. For the resurfacing joint and THA Magnum 1 to 4, index and revision surgeries were performed at the Avon Orthopaedic Centre, North Bristol NHS Trust with revision undertaken between September 2012 and February 2014. For the joint labelled THA Magnum 5 in Table I, both the index and revision surgeries were performed at the Royal United Hospital Bath but by different surgeons. This implant was part of the retrieval study reported here, but is not part of the prospective Bristol ReCap study. Mean time in vivo for all of the implants was 6.7 years (5.3 to 7.6) (Table I). The mean age of the patients was 60.3 years (56.2 to 68.1) at primary surgery. The nominal implant diameters ranged from 44 mm to 52 mm. Five of the implants were retrieved from female patients, one was from a male (THA Magnum 4).

Fig.

a) Biomet Magnum/ ReCap/ Taperloc large head MoM THA and (b) Biomet ReCap resurfacing hip joint.

All patients were revised for hip pain and/or investigations indicative of an ARMD and/or elevated blood metal ion levels. Prior to retrieval, blood was sampled from four patients (five hips) using a stainless steel 21-gauge needle (Becton Dickinson UK Ltd, Oxford, United Kingdom) and collected in a trace element tube that contained sodium ethylenediaminetetraacetic acid (EDTA). The samples were measured by inductively coupled plasma mass spectrometry (ICPMS) for whole blood. Cobalt (Co) and chromium (Cr) levels are represented in µg/l (Table I) and ranged from 1.96 µg/l to 18.1 µg/l for Co, and 2.48 µg/l to 16.5 µg/l for Cr. In accordance with medicines and healthcare products regulatory agency guidelines, the blood metal ion levels in two patients were higher than 7 µg/l and identified as being at risk for metal-on-metal bearings.¹³ Further investigations were performed on three patients (four hips) using metal artefact reduction sequence (MARS) for magnetic resonance imaging (MRI). Evidence of a cystic collection or a soft-tissue mass further supported the presence of an ARMD in all four hips. The MARS MRI was not available in the other two patients. Finally, a positive histological analysis (examination of retrieved cellular tissue) of ARMD was confirmed as the diagnosis in four hips and the reason for failure. Histology was not available in two patients (the resurfacing joint and THA Magnum 5).

The bearing surfaces of the ReCap resurfacing are identical to those of the THA design. Both the resurfacing joint and THA were made from high carbon (> 0.2%) as-cast cobalt-chromium-molybdenum (CoCrMo). The radial clearance is specified as being between 75 µm and 150 µm with sphericity of less than 5 µm. The femoral head extends approximately 23° beyond a full hemisphere. The outside geometry of the acetabular component is hemispherical. The CoCr THA femoral head is designed to accept a tapered titanium (Ti) alloy sleeve. This Ti sleeve has an internal female taper and is used with a forged Ti Taperloc stem to give a Ti/Ti taper/trunnion junction. These data are taken from the ReCap design rationale booklet.²¹ The acetabular component internal arc of cover ranges from 154.6° for a 38 mm diameter implant to 163.6° for the 60 mm diameter implant.²⁰

All explanted joints were sterilised in 10% formaldehyde solution and cleaned thoroughly before being measured to determine wear volumes.

Equipment

The volumetric wear from the articulating surfaces (femoral head and acetabular component) of the six retrieved implants was measured on a CMM (Legex 322; Mitutoyo UK Ltd, Andover, United Kingdom) with an accuracy of 0.8 µm.^8,18 The volumetric wear of the tapers of the five femoral heads from the Magnum/ ReCap/ Taperloc THAs was also measured on this CMM. No femoral stems were available for analysis. The CMM is a contacting instrument where a stylus is used to map the geometrical profile of the specimen surface.^8,18 A custom-designed Matlab programme was used to provide the volumetric wear of the surfaces. The CMM and Matlab method have been shown to have an accuracy of approximately 0.5 mm^318,22 for the wear measurements of the articulating surfaces, and 0.2 mm³ for the wear measurements of the tapers.⁸

In addition to the wear measurements, surface roughness measurements were taken on the articulating surfaces using a Zygo NewView 5000 non-contacting white light interferometric profilometer (Zygo Corporation, Middlefield, Connecticut). This has previously been used to measure the roughness of explanted MoM hips.^23-25 The ×10 lens was used with a x2 zoom, giving an area of view of 317 x 238 µm. The manufacturers of the NewView 5000 state that the vertical resolution of this profilometer is better than 1 nm with a 0.99 µm lateral (x, y) resolution for the magnification used. Measurements of S_a (mean surface roughness over the 3D area of view), S_q (root mean square surface roughness over the 3D area of view), and S_sk (surface skewness over the 3D area of view) were taken. The skewness of a surface is a measure of symmetry of the profile about the mean line. A positively skewed surface has a predominance of peaks, and a negatively skewed surface has more pronounced valleys. All six heads and all six acetabular components were measured in the unworn region and the worn region. In order to determine the areas of the surfaces that were unworn and worn, the surfaces of these joints were visually inspected – this was then confirmed by images taken using the CMM. Ten measurements were taken within each region and the mean calculated.

Surface roughness measurements were also taken on the five femoral head tapers using a Mitutoyo Surftest SJ-210 (Mitutoyo Corporation, Kanagawa, Japan). Four 5 mm line profiles were taken at 90° angles to each other. These profiles were taken in the region where the taper was engaged with the trunnion (as shown by the CMM measurements). The mean R_a (mean surface roughness of the 2D line profile) and R_q (root mean square surface roughness of the 2D line profile) values of these four roughness profiles were calculated for each of the five tapers.

Theoretical analysis

Theoretical studies were also performed. The predicted minimum film thickness using the theory by Hamrock and Dowson,²⁶ along with the surface roughness measurements of the articulating surfaces, was used to calculate the theoretical lubrication regimes acting within the joints.²⁷ These equations have been used previously in the calculation of the theoretical lubrication regimes of MoM joints,²⁸ and the equations are given below.

The minimum film thickness equation:²⁶

h_min is the minimum film thickness, R_x is the equivalent radius for a ball on plane model (equation shown below), η is the viscosity of the lubricant (taken as 0.010 Pa s, a similar viscosity to pathological synovial fluid under the shear-thinning encountered during the stance phase of the walking cycle²⁹), u is the entraining velocity (0.012 ms⁻¹, again, during the stance phase of walking), E′ is the equivalent elastic modulus of the material pairing (equation shown below; 2.3×10¹¹ Pa for CoCrMo) and L is the applied load (taken as 2000 N).

The equivalent radius equation:

R_x is the equivalent radius, R_h is the radius of the femoral head and R_c is the radius of the acetabular component.

The equation for the equivalent elastic modulus:

υ₁ and E₁ are the Poisson’s ratio and elastic modulus of the head (0.3 and 2.1×10¹¹ Pa) and υ₂ and E₂ are the Poisson’s ratio and elastic modulus of the component (0.3 and 2.1×10¹¹ Pa).

Finally, the calculated minimum film thicknesses, together with the recorded mean values of average root mean square S_q for the femoral head and acetabular component (S_qh and S_qc) for each joint, were used to determine the dimensionless parameter λ and therefore the theoretical lubrication regimes acting within the joints.²⁷

Previous work has shown that if the ratio of the minimum film thickness to the combined surface roughness is greater than one but less than three, mixed lubrication is likely.²⁷ If λ > 3 then a full-fluid film-lubricating regime is predicted,²⁷ i.e. the asperities of the bearing surfaces are completely separated by the lubricant film. A λ value of less than one indicates boundary lubrication.²⁷

Wear of the articulating surfaces

Wear data for the articulating surfaces of the femoral head and acetabular component of each joint are shown in Table II, along with the radial clearances of the unworn sections (as measured on the CMM). These wear values are given as means because the data were found to be normally distributed using the Anderson-Darling test for normality (p = 0.918). The total wear volume for the head and the acetabular component of the THAs and resurfacing joint was between 30.0 mm³ and 57.3 mm³, and the volumetric wear rate was between 4.1 mm³/year and 7.6 mm³/year (mean: 6.1 mm³/year). The mean volumetric wear rate measured for the five THAs was also 6.1 mm³/year, with a range of 4.1 to 7.6 mm³/year. The ReCap resurfacing hip joint provided a wear rate of 5.9 mm³/year. In all cases, the femoral head wear volume was larger than the wear volume from the acetabular component. The mean femoral head:acetabular component wear ratio was 62:38 (54:46 to 69:31). Examples of the images obtained showing the wear profiles for the articulating surfaces of the femoral head and acetabular component are shown in Figure 2. The femoral head wear can be seen as being located close to the pole and the acetabular component wear was mainly at the rim. The measured radial clearances were within the range given in the manufacturer’s specifications.²¹

Fig. 2

Examples of worn images (blue/green) of areas on femoral head (left) and acetabular component (right) indicative of all joints measured in this study.

Surface roughness measurements of the articulating surfaces

The surface roughness measurements performed on the unworn and worn regions of the femoral heads and acetabular components are shown in Tables III and IV. It was not possible to distinguish an unworn area on the acetabular component of THA Magnum 3, therefore, no unworn measurements were available for this acetabular component. Figures 3 and 4 show examples of images acquired by the non-contacting profilometer on the unworn (a) and worn (b) regions of the femoral heads and acetabular components, respectively.

Fig.

Surface profilometry images of a femoral head (a) unworn region showing carbides (S_qh: 0.047 µm; S_sk: 0.929), and (b) worn region showing scratching (S_qh: 0.075 µm; S_sk: 0.473).

Fig.

Surface profilometry images of an acetabular component (a) unworn region showing carbides (S_qc: 0.035 µm; S_sk: 1.032) and (b) worn region showing scratching (S_qc: 0.054 µm; S_sk: -0.112).

Theoretical lubrication regimes

Using the equations for predicted minimum film thickness, h_min, and the dimensionless parameter, λ, the theoretical lubrication regime acting within each joint during the stance phase of the walking cycle was calculated. Table V shows the predicted lubrication regimes for each joint using the surface roughness values obtained in the unworn region and the worn region of each component.

Taper wear

The taper angles and wear volumes measured on the female tapers of the five THA femoral heads are shown in Table VI. The volumetric wear for each of the five Biomet Magnum THA tapers ranged from 0.16 to 0.96 mm³, with a mean total volumetric wear of 0.38 mm³ and a mean wear rate of 0.054 mm³/year. The maximum wear depths ranged from 4.6 µm to 8.5 µm, with a mean of 5.7 µm. Indicative images displaying the positioning of the wear on the tapers are shown in Figure 5.

Fig.

Images of worn (blue/green) areas on the femoral head tapers of the THAs.

Table VII provides the surface roughness results for the tapers of each THA Magnum femoral head. These were measured in the regions that were engaged with the trunnion when implanted.