Original Article
Influence of stem type on material loss at the metal-on-metal pinnacle taper junction
Proc IMechE Part H: J Engineering in Medicine 2015, Vol. 229(1) 91–97 Ó IMechE 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411914567931 pih.sagepub.com
Harry S Hothi, Robert K Whittaker, Jay M Meswania, Gordon W Blunn, John A Skinner and Alister J Hart
Abstract The clinical importance of material loss at the head–stem junction is unknown. Comparison of retrievals with different stem types can provide the opportunity to understand the importance of the taper junction. This was a case–control study involving 20 retrieved 36 mm metal-on-metal Pinnacle (DePuy) hips that were paired with either a Corail (n = 10) or S-ROM (n = 10) stem. The median head taper material loss rate for the Corail group was 0.238 (0.0002–2.178) mm3/ year and was significantly greater than the S-ROM group (p = 0.042), which had a median material loss rate of 0.132 (0.015–0.518) mm3/year. The only significant difference between the groups was the stem taper roughness and length: this was rougher and shorter for the Corails. Long and smooth stem taper designs are preferred when used in conjunction with metal heads.
Keywords Wear analysis/testing, orthopaedic tribology, hip prostheses, medical biomaterials, biomedical devices
Date received: 16 October 2014; accepted: 18 December 2014
Introduction Of all the primary metal-on-metal total hip replacement (MOM-THR) procedures performed in the United Kingdom between 2003 and 2011, approximately 52% of these used modular cups consisting of a metal outer shell and metal liner; approximately, 80% of these were paired with a 36-mm femoral head.1 The latest annual report by The National Joint Registry for England, Wales and Northern Ireland2 shows that the most commonly used cup brand in the United Kingdom is the DePuy Pinnacle (DePuy Orthopaedics, Warsaw, IN, USA), which consists of a titanium shell and a polyethylene, ceramic or metal (cobalt–chromium) liner, which may be coupled with a metal or ceramic femoral head. However, the sale of the Food and Drug Administration (FDA)-approved MOM articulation with this cup system (DePuy Ultamet) was discontinued in 2013, reflecting the rapid global decline in the use of MOM hip implants. The Pinnacle Ultamet hip had two regions of modularity: (1) at the cup shell and liner and (2) at the head and stem. While the modular cup element consisted of a single shell/liner design combination, the femoral head could be paired with either a Corail (DePuy) or
S-ROM (DePuy) stem. Munir et al.3 found that the surface topography of the tapers of these two stems differs considerably, such that the S-ROM taper is relatively smooth whereas the Corail is distinctly threaded. Second, the geometries of the two tapers differ between an ‘11/13’ (S-ROM) and ‘12/14’ (Corail) taper. Nassif et al.4 reported that taper design can affect failure of MOM-THRs and Panagiotidou et al.5 presented evidence of enhanced corrosion and fretting of head taper surfaces when paired with roughened stem tapers. Langton et al.6 reported median wear rates of Pinnacle head tapers to be 2.7 times greater with paired with a Corail stem than with an S-ROM. However, Matthies et al.7 demonstrated that there are multiple factors that may contribute to an increase in taper wear that must be taken into consideration. We aimed to investigate
Institute of Orthopaedics and Musculoskeletal Science, University College London and Royal National Orthopaedic Hospital, Stanmore, UK Corresponding author: Harry S Hothi, Institute of Orthopaedics and Musculoskeletal Science, University College London and Royal National Orthopaedic Hospital, Brockley Hill, Stanmore, Middlesex HA7 4LP, UK. Email: [email protected]
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
92
Proc IMechE Part H: J Engineering in Medicine 229(1)
Table 1. Implant and patient data showing median (range) values with p-values indicating the significance of differences between the parameters.
Gender (male:female) Age at primary surgery (years) Time to revision (months) Cup inclination (°) Whole blood cobalt (ppb) Whole blood chromium (ppb) Co/Cr ratio
S-ROM group
Corail group
p-value
3:7 64.5 (53–74) 63.5 (40–84) 36 (31–53) 4.95 (1.06–24.6) 4.51 (1.51–90) 1.03 (0.27–2.38)
5:5 63 (26–73) 56 (50–77) 44 (31–55) 10.9 (2.3–130) 4.81 (1.45–42.4) 1.95 (0.98–8.25)
0.649 0.478 0.765 0.159 0.397 0.916 0.034
Figure 1. Examples of (a) S-ROM stem, (b) Corail stem, (c) metal Ultamet head and (d) Pinnacle cup.
the effect of stem taper design on head taper junction material loss in patients.
Method This was a retrieval study with a case–control design involving 20 MOM 36-mm Pinnacle Ultamet hips received at our retrieval laboratory which were used in conjunction with an S-ROM stem (n = 10) or Corail stem (n = 10) (Figure 1). In the majority of cases, the actual stem was retained during revision surgery; only one S-ROM and two Corail stems were retrieved in this study. The femoral heads used in the Pinnacle Ultamet articulation are designed with differing neck lengths between the S-ROM (11/13 taper) and Corail (12/14 taper). Therefore, to match the two designs as closely as possible, we required that the head neck length was + 6 when used with S-ROM stems and + 5 with Corail stems for inclusion in our study. Second, we required that the head taper was visible so that its surface could be assessed. The implants for the S-ROM group were retrieved from three male and seven female patients with a median age of 64.5 (53–74) years at primary surgery, while the Corail group consisted of five male and five female patients with a median age of 63 (26–73) years. Table 1 summarises all the patient and implant demographic data for the two stem designs considered. Mann– Whitney tests confirmed that the two groups were
Figure 2. Study design used.
statistically matched with respect to gender, age, time to revision, inclination and Cr and Cr blood metal ion levels. The Corail group had a median (range) Co/Cr ratio of 1.95 (0.95–8.25) and was significantly greater (p = 0.0341) than the S-ROM group, which had a median (range) Co/Cr ratio of 1.03 (0.27–2.38). The study design of this work is summarised in Figure 2.
Visual assessment of corrosion Macroscopic and stereomicroscopic examinations of the taper surfaces of all 20 heads were performed by a single examiner experienced in retrieval analysis, to score for visual evidence of corrosion. A scale of 1 (none) to 4 (severe) was used as defined by Goldberg et al.8 and a single score was determined for each head, following assessment of the surface as a whole. This method has previously been demonstrated as being both repeatable and reproducible, and corrosion scores
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
93
have been shown to be correlated with measured material loss at the surface.9 Corroded regions were considered as those with discolouration or dullness or with black debris or signs of pitting or etching. A Leica M50 microscope (Leica Microsystems, Germany) at up to 403 magnification was used in examinations.
Measurement of head taper material loss A Talyrond 365 (Taylor Hobson, Leicester, UK) roundness-measuring machine was used to measure the volume of material loss at each head taper (n = 20), using methods previously published.10 The femoral head was positioned on a rotating air spindle, levelled and centred with respect to the spindle axis, and a series of 180 vertical traces were taken axially along the taper surface with a 5-mm diamond stylus. The vertical traces from each taper surface were combined into a rectangular surface map and the unworn taper surface was manually identified in order to level the data obtained. An Abbott–Firestone curve was then used to calculate the volume of material loss of the regenerated head taper surface, taking into account its conical shape.
Measurement of bearing surface material loss A Zeiss Prismo (Carl Zeiss Ltd, Rugby, UK) coordinate measuring machine (CMM) was used to measure the volume of material loss at the bearing surfaces of each component (n = 40), using methods previously published.11 In all, 400 polar scan lines were defined to record up to 300,000 data points using a 2-mm ruby stylus moving at a speed of 3 mm/s. The measurement data were analysed using an iterative least square fitting method and were such that only the geometry of the unworn bearing was used to map the distribution of material loss. The wear maps generated were used to determine whether edge loading of the hip had occurred by identifying the region of greatest wear on the cup surface.
Stem taper analysis A Contour GT-K 3D optical profilometer (Bruker, UK) was used to examine the retrieved stems (n = 3) to determine the height and pitch of the machined thread on the surface of the tapers. An objective lens of 53 was used to scan the surface with a backscan of 80 mm and length of 50 mm.
Statistical analysis An independent samples t-test was performed to assess the significance of taper material loss (mm3/year) differences between the two stem groups. Additionally, we examined the significance of differences between the two groups in relation to (1) taper engagement length,
(2) bearing surface wear, (3) cup edge wear and (4) increased vertical femoral offset.
Results The median vertical femoral offset with an S-ROM and Corail stem was 69 (63–82) mm and 70 (59–103) mm, respectively. There was no significant difference between the two stems (p = 0.44). Edge wearing of the cup occurred in five hips with an S-ROM and five with a Corail. The 11/13 S-ROM stem tapers had a median taper engagement length of 14 (10.5–14) mm at their longest point of engagement in the medial–lateral plane (Figure 3(a)) and were significantly greater (p \ 0.001) than the 12/14 Corail stem tapers, which had a median taper engagement length of 10.5 (10–11) mm. The median taper engagement length of the S-ROM tapers in the regions of the scallops was comparable to that of the Corail stems. Figure 3 illustrates the differences in taper engagement length and position between the two stem designs.
Visual assessment of corrosion All head tapers were found to have evidence of corrosion; statistical analysis revealed that the differences were not significant (p = 0.0769).
Measurement of bearing surface material loss The median material loss of the combined cup and head bearing surfaces of hips with an S-ROM stem it was 3.92 (1.20–7.81) mm3/year, while with a Corail stem, it was 3.21 (0.87–62.12) mm3/year (Table 2). There was no significant difference between the two groups (p = 0.3192).
Measurement of head taper material loss The median material loss rate of the head tapers with an S-ROM was 0.132 (0.015–0.518) mm3/year, while with a Corail it was 0.238 (0.0002–2.178) mm3/year (Figure 4). The Corail group was found to have significantly more material loss than the S-ROM group (p = 0.036).
Stem taper analysis Figures 5(a) and 5(b) presents the taper surfaces of the S-Rom and Corail stems, respectively, as viewed under the optical microscope. The median (range) thread heights for the single S-ROM stem and two Corail stems retrieved in this study were 1 (1–1.1) mm and 11.5 (11.3–11.7) mm, respectively (using three different scan regions on each taper). The median (range) thread spacing for the S-ROM and Corail stems was 0.099 (0.955–0.102) mm and 0.2 (0.193–0.228) mm,
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
94
Proc IMechE Part H: J Engineering in Medicine 229(1)
Figure 3. Schematic illustrations presenting the taper engagement of the S-ROM stem in the (a) medial–lateral and (b) anterior– posterior views. The shorter engagement length of the scalloped regions can be seen. The Corail stem is presented in the (c) distal– proximal and (d) anterior–posterior views (examples of points of elevated stress due to toggling are circled).
Table 2. Median (range) material loss values, taper engagement length, presence of edge wearing and vertical femoral offset.
Taper engagement length (mm) Vertical femoral offset (mm) Cup edge worn? Yes:no Bearing surface material loss (mm3/year) Taper surface material loss (mm3/year)
respectively. This revealed that Corail taper had a distinctly rougher surface topography than the S-ROM stem.
Discussion The release of cobalt and chromium metal ions from the taper junction of large diameter (536 mm) MOMTHRs is speculated to contribute to the high revision rates of these hip systems. A number of retrieval studies have reported on a wide range of material loss at the
S-ROM group
Corail group
14 (10.5–14) 69 (63–82) 5:5 3.92 (1.20–7.81) 0.132 (0.015–0.518)
10.5 (10–11) 70 (59–103) 5:5 3.21 (0.87–62.12) 0.238 (0.0002–2.178)
surfaces of modular junctions10,12,13 together with evidence of substantial corrosion;14–16 this has been associated with periprosthetic soft tissue damage in patients.17–20 It is speculated that the underlying cause of increased mechanical wear and corrosion at the head taper is multifactorial. In order to better understand the clinical impact of an individual factor, there is considerable benefit in employing a case–control design in retrieval analysis whereby hips are selected such that they are statistically matched in relation to other influencing
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
95
factors. In this study, we reported on the effect of stem type (longer, smoother S-ROM taper vs shorter, rougher Corail taper) on head taper material loss. This study was limited by the comparatively low numbers of hips in each group; however, our inclusion criteria meant that the two groups were closely matched with respect to gender, age, time to revision, neck length and inclination. Furthermore, no significant differences in bearing surface wear, cup edge wearing or vertical femoral offset were found, which are thought to contribute to differences in taper material loss. As such we are able to draw clinically significant conclusions from our study regarding the impact of stem design. We found that head tapers that had been paired with Corail stems had significantly greater rates of material loss per year than heads paired with S-ROM stems. Additionally, we found a statistically significant difference in the taper engagement length: a median of 10.5 mm for the Corail group compared with 14 mm for the S-ROM group at its greatest length of engagement. While the S-ROM taper was engaged up to the opening
Figure 4. Distribution of head taper material loss values for the two stem groups.
of the head taper (excluding scalloped regions), the Corail stem taper in all cases was found to be seated fully within the head taper, such that at the base of the stem, taper was in contact with the head taper surface at a distance from the taper edge, Figure 3. This positioning of the stem taper may increase the susceptibility of it toggling within the head taper, resulting in localised regions of increased contact stresses (circled in Figure 3(b)), suggesting a pattern of mechanical wear not observed with the longer S-ROM taper. The significantly greater pre-revision Co/Cr ratios in the Corail group suggest that the taper junction in these hips would have corroded more, with much of the Co ions being retained on the surface of the surface, while much of the Cr ions would be released into the blood. The comparable volumes of material loss measured at the bearing surfaces of each group support differences in Co/Cr ratios as being due to corrosion at the taper junction. Cooper et al.20 reported high Co/Cr ratios in hips with severely corroded modular junctions and nonMOM bearings, while Hart et al.21 found implant corrosion to be associated with higher Co than Cr in periprosthetic tissue. This, however, was not reflected in the visual corrosion scoring of head tapers in this study; however, this may be due to the subjective nature of this scoring method. The 12/14 head tapers were found to have evidence of imprinting of the rougher thread of the Corail stem taper; this was not observed in S-ROM tapers. It has been suggested that this is due to galvanic corrosion in which the head taper, rather than the stem taper, is preferentially corroded.10 This corrosion mechanism coupled with an increase in the risk of toggling and edge loading of the shorter stem taper (when fully seated inside the head taper) may explain the greater material loss observed at these junctions.
Figure 5. Image generated via the optical profilometer of the (a) smooth S-ROM taper and (b) rough Corail taper.
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
96
Proc IMechE Part H: J Engineering in Medicine 229(1)
The findings of our study are in agreement with those reported by Langton et al.,6 and the magnitude of head taper material loss is comparable. While notably larger volumes of material loss were measured at the bearing surface than the head taper in this study, it has been speculated that the metal debris from the taper junction may be more potent in causing damage to soft tissue, suggesting that the comparatively low volumes of material loss may still be clinically relevant.
Future work While the two stems in this study demonstrate clear differences in their taper surface topography and geometry, future work will examine the significance of more subtle variations in surface roughness and size on head taper material loss. Furthermore, there is some debate regarding the clinical impact of the mixing and matching of components from different manufacturers in THRs. It is speculated that stem–head taper incompatibility due to varying manufacturing tolerances may impact on taper junction performance22 and future work will examine this issue. Corrosion of the modular neck junction in THRs has been reported previously23 and we will examine the significance of this in relation to the stem–head junction.
Conclusion This was a closely matched case–control study involving a single femoral head design paired with two distinctly different stem tapers from the same manufacturer. We showed that shorter and rougher stems can increase the volume of taper material loss; the consequential release of metal ions from this junction has been implicated in implant failure. When using a metal head, we recommend the use of stems with smooth and long tapers to minimise the clinical impact of damage at the taper junction. Acknowledgements We are grateful for the support of Gwynneth Lloyd, Charlotte Page and Elizabeth Ellis for their coordination of the retrieval centre; Bob Skinner and Siva Mahindan for their support in metrology; Erica Cook for statistical analysis and Christian Klemt for assistance with the analysis of plain radiographs. Declaration of conflicting interests The authors declare that there is no conflict of interest. Funding Two authors received funding from the British Orthopaedic Association through an industry consortium of nine manufacturers: DePuy International Ltd, Zimmer GmbH, Smith & Nephew UK Ltd, Biomet UK Ltd, JRI Ltd, Finsbury Orthopaedics Ltd, Corin
Group PLC, Mathys Orthopaedics Ltd and Stryker UK Ltd. References 1. Smith AJ, Dieppe P, Vernon K, et al. Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry of England and Wales. Lancet 2012; 379(9822): 1199–1204. 2. National Joint Registry for England and Wales (NJR) 10th annual report, 2013, www.njrcentre.org.uk 3. Munir S, Imbuldeniya A and Walter WL. Variations in the taper surface topography between 11 different commercially available hip replacement stems. Presented at the annual meeting of the American academy of orthopaedic surgeons, New Orleans, LA, 11–15 March 2014. 4. Nassif NA, Nawabi DH, Stoner K, et al. Taper design affects failure of large-head metal-on-metal total hip replacements. Clin Orthop Relat Res 2013; 472: 564–571. 5. Panagiotidou A, Meswania J, Hua J, et al. Enhanced wear and corrosion in modular tapers in total hip replacement is associated with the contact area and surface topography. J Orthop Res 2013; 31(12): 2031–2039. 6. Langton D, Nargol A, Sayginer O, et al. Taper failure: wear is the corrosion? Presented at the annual meeting of the British hip society, Exeter, 5–7 March 2014. 7. Matthies AK, Cro S, Bills P, et al. Which factors determine the volume of material lost from the taper junction of metal-on-metal hip replacements? Presented at the annual meeting of the American academy of orthopaedic surgeons, New Orleans, LA, 11–15 March 2014. 8. Goldberg JR, Gilbert JL, Jacobs JJ, et al. A multicentre retrieval study of the taper interfaces of modular hip prostheses. Clin Orthop Relat Res 2002; 401: 149–161. 9. Hothi HS, Matthies AK, Berber R, et al. The reliability of a scoring system for corrosion and fretting, and its relationship to material loss of tapered, modular junctions of retrieved hip implant. J Arthroplasty 2014; 29(6): 1313–1317. 10. Matthies AK, Racasan R, Bills P, et al. Material loss at the taper junction of retrieved large head metal-on-metal total hip replacements. J Orthop Res 2013; 31(11): 1677– 1685. 11. Bills PJ, Racasan R, Underwood RJ, et al. Volumetric wear assessment of retrieved metal-on-metal hip prostheses and the impact of measurement uncertainty. Wear 2012; 274: 212–219. 12. Langton DJ, Sidaginamale R, Lord JK, et al. Taper junction failure in large-diameter metal-on-metal bearings. Bone Joint Res 2012; 1: 56–63. 13. Bolland BJ, Culliford DJ, Langton DJ, et al. High failure rate with a large diameter hybrid metal-on-metal total hip replacement: clinical, radiological and retrieval analysis. J Bone Joint Surg Br 2011; 93–B: 608–615. 14. Cross MB, Esposito C, Sokolova A, et al. Fretting and corrosion changes in modular total hip arthroplasty. Bone Joint J 2013; 95-B(Suppl. 15): 127. 15. Higgs G, Kurtz S, Hanzlik J, et al. Retrieval analysis of metal-on-metal hip prostheses: characterising fretting and corrosion at modular interfaces. Bone Joint J 2013; 95-B (Suppl. 15): 108. 16. Meyer H, Mueller T, Goldau G, et al. Corrosion at the cone/taper interface leads to failure of large-diameter
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
17.
18.
19.
20.
97
metal-on-metal total hip arthroplasties. Clin Orthop Relat Res 2012; 470(11): 3101–3108. Chana R, Esposito C, Campbell PA, et al. Mixing and matching causing taper wear: corrosion associated with pseudotumour formation. J Bone Joint Surg Br 2012; 94(2): 281–286. Fricka KB, Ho H, Peace WJ, et al. Metal-on-metal local tissue reaction is associated with corrosion of the head taper junction. J Arthroplasty 2012; 27(8): 26–31. Whitehouse MR, Endo M and Masri BA. Adverse local tissue reaction associated with a modular hip hemiarthroplasty. Clin Orthop Relat Res 2013; 471(12): 4082–4086. Cooper HJ, Urban RM, Wixson RL, et al. Adverse local tissue reaction arising from corrosion at the femoral
neck-body junction in a dual-taper stem with a cobaltchromium modular neck. J Bone Joint Surg Am 2013; 95: 865–872. 21. Hart AJ, Quinn PD, Lali F, et al. Cobalt from metal-onmetal hip replacements may be the clinically relevant active agent responsible for periprosthetic tissue reactions. Acta Biomater 2012; 8: 3865–3873. 22. Shareef N and Levine D. Effect of manufacturing tolerances on the micromotion at the morse taper interface in modular hip implants using the finite element technique. Biomaterials 1996; 17: 623–630. 23. Kop AM and Swarts E. Corrosion of a hip stem with a modular neck taper junction. J Arthroplasty 2009; 24(7): 1019–1023.
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Influence of stem type on material loss at the metal-on-metal pinnacle taper junction
Proc IMechE Part H: J Engineering in Medicine 2015, Vol. 229(1) 91–97 Ó IMechE 2015 Reprints and permissions: sagepub.co.uk/journalsPermissions.nav DOI: 10.1177/0954411914567931 pih.sagepub.com
Harry S Hothi, Robert K Whittaker, Jay M Meswania, Gordon W Blunn, John A Skinner and Alister J Hart
Abstract The clinical importance of material loss at the head–stem junction is unknown. Comparison of retrievals with different stem types can provide the opportunity to understand the importance of the taper junction. This was a case–control study involving 20 retrieved 36 mm metal-on-metal Pinnacle (DePuy) hips that were paired with either a Corail (n = 10) or S-ROM (n = 10) stem. The median head taper material loss rate for the Corail group was 0.238 (0.0002–2.178) mm3/ year and was significantly greater than the S-ROM group (p = 0.042), which had a median material loss rate of 0.132 (0.015–0.518) mm3/year. The only significant difference between the groups was the stem taper roughness and length: this was rougher and shorter for the Corails. Long and smooth stem taper designs are preferred when used in conjunction with metal heads.
Keywords Wear analysis/testing, orthopaedic tribology, hip prostheses, medical biomaterials, biomedical devices
Date received: 16 October 2014; accepted: 18 December 2014
Introduction Of all the primary metal-on-metal total hip replacement (MOM-THR) procedures performed in the United Kingdom between 2003 and 2011, approximately 52% of these used modular cups consisting of a metal outer shell and metal liner; approximately, 80% of these were paired with a 36-mm femoral head.1 The latest annual report by The National Joint Registry for England, Wales and Northern Ireland2 shows that the most commonly used cup brand in the United Kingdom is the DePuy Pinnacle (DePuy Orthopaedics, Warsaw, IN, USA), which consists of a titanium shell and a polyethylene, ceramic or metal (cobalt–chromium) liner, which may be coupled with a metal or ceramic femoral head. However, the sale of the Food and Drug Administration (FDA)-approved MOM articulation with this cup system (DePuy Ultamet) was discontinued in 2013, reflecting the rapid global decline in the use of MOM hip implants. The Pinnacle Ultamet hip had two regions of modularity: (1) at the cup shell and liner and (2) at the head and stem. While the modular cup element consisted of a single shell/liner design combination, the femoral head could be paired with either a Corail (DePuy) or
S-ROM (DePuy) stem. Munir et al.3 found that the surface topography of the tapers of these two stems differs considerably, such that the S-ROM taper is relatively smooth whereas the Corail is distinctly threaded. Second, the geometries of the two tapers differ between an ‘11/13’ (S-ROM) and ‘12/14’ (Corail) taper. Nassif et al.4 reported that taper design can affect failure of MOM-THRs and Panagiotidou et al.5 presented evidence of enhanced corrosion and fretting of head taper surfaces when paired with roughened stem tapers. Langton et al.6 reported median wear rates of Pinnacle head tapers to be 2.7 times greater with paired with a Corail stem than with an S-ROM. However, Matthies et al.7 demonstrated that there are multiple factors that may contribute to an increase in taper wear that must be taken into consideration. We aimed to investigate
Institute of Orthopaedics and Musculoskeletal Science, University College London and Royal National Orthopaedic Hospital, Stanmore, UK Corresponding author: Harry S Hothi, Institute of Orthopaedics and Musculoskeletal Science, University College London and Royal National Orthopaedic Hospital, Brockley Hill, Stanmore, Middlesex HA7 4LP, UK. Email: [email protected]
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
92
Proc IMechE Part H: J Engineering in Medicine 229(1)
Table 1. Implant and patient data showing median (range) values with p-values indicating the significance of differences between the parameters.
Gender (male:female) Age at primary surgery (years) Time to revision (months) Cup inclination (°) Whole blood cobalt (ppb) Whole blood chromium (ppb) Co/Cr ratio
S-ROM group
Corail group
p-value
3:7 64.5 (53–74) 63.5 (40–84) 36 (31–53) 4.95 (1.06–24.6) 4.51 (1.51–90) 1.03 (0.27–2.38)
5:5 63 (26–73) 56 (50–77) 44 (31–55) 10.9 (2.3–130) 4.81 (1.45–42.4) 1.95 (0.98–8.25)
0.649 0.478 0.765 0.159 0.397 0.916 0.034
Figure 1. Examples of (a) S-ROM stem, (b) Corail stem, (c) metal Ultamet head and (d) Pinnacle cup.
the effect of stem taper design on head taper junction material loss in patients.
Method This was a retrieval study with a case–control design involving 20 MOM 36-mm Pinnacle Ultamet hips received at our retrieval laboratory which were used in conjunction with an S-ROM stem (n = 10) or Corail stem (n = 10) (Figure 1). In the majority of cases, the actual stem was retained during revision surgery; only one S-ROM and two Corail stems were retrieved in this study. The femoral heads used in the Pinnacle Ultamet articulation are designed with differing neck lengths between the S-ROM (11/13 taper) and Corail (12/14 taper). Therefore, to match the two designs as closely as possible, we required that the head neck length was + 6 when used with S-ROM stems and + 5 with Corail stems for inclusion in our study. Second, we required that the head taper was visible so that its surface could be assessed. The implants for the S-ROM group were retrieved from three male and seven female patients with a median age of 64.5 (53–74) years at primary surgery, while the Corail group consisted of five male and five female patients with a median age of 63 (26–73) years. Table 1 summarises all the patient and implant demographic data for the two stem designs considered. Mann– Whitney tests confirmed that the two groups were
Figure 2. Study design used.
statistically matched with respect to gender, age, time to revision, inclination and Cr and Cr blood metal ion levels. The Corail group had a median (range) Co/Cr ratio of 1.95 (0.95–8.25) and was significantly greater (p = 0.0341) than the S-ROM group, which had a median (range) Co/Cr ratio of 1.03 (0.27–2.38). The study design of this work is summarised in Figure 2.
Visual assessment of corrosion Macroscopic and stereomicroscopic examinations of the taper surfaces of all 20 heads were performed by a single examiner experienced in retrieval analysis, to score for visual evidence of corrosion. A scale of 1 (none) to 4 (severe) was used as defined by Goldberg et al.8 and a single score was determined for each head, following assessment of the surface as a whole. This method has previously been demonstrated as being both repeatable and reproducible, and corrosion scores
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
93
have been shown to be correlated with measured material loss at the surface.9 Corroded regions were considered as those with discolouration or dullness or with black debris or signs of pitting or etching. A Leica M50 microscope (Leica Microsystems, Germany) at up to 403 magnification was used in examinations.
Measurement of head taper material loss A Talyrond 365 (Taylor Hobson, Leicester, UK) roundness-measuring machine was used to measure the volume of material loss at each head taper (n = 20), using methods previously published.10 The femoral head was positioned on a rotating air spindle, levelled and centred with respect to the spindle axis, and a series of 180 vertical traces were taken axially along the taper surface with a 5-mm diamond stylus. The vertical traces from each taper surface were combined into a rectangular surface map and the unworn taper surface was manually identified in order to level the data obtained. An Abbott–Firestone curve was then used to calculate the volume of material loss of the regenerated head taper surface, taking into account its conical shape.
Measurement of bearing surface material loss A Zeiss Prismo (Carl Zeiss Ltd, Rugby, UK) coordinate measuring machine (CMM) was used to measure the volume of material loss at the bearing surfaces of each component (n = 40), using methods previously published.11 In all, 400 polar scan lines were defined to record up to 300,000 data points using a 2-mm ruby stylus moving at a speed of 3 mm/s. The measurement data were analysed using an iterative least square fitting method and were such that only the geometry of the unworn bearing was used to map the distribution of material loss. The wear maps generated were used to determine whether edge loading of the hip had occurred by identifying the region of greatest wear on the cup surface.
Stem taper analysis A Contour GT-K 3D optical profilometer (Bruker, UK) was used to examine the retrieved stems (n = 3) to determine the height and pitch of the machined thread on the surface of the tapers. An objective lens of 53 was used to scan the surface with a backscan of 80 mm and length of 50 mm.
Statistical analysis An independent samples t-test was performed to assess the significance of taper material loss (mm3/year) differences between the two stem groups. Additionally, we examined the significance of differences between the two groups in relation to (1) taper engagement length,
(2) bearing surface wear, (3) cup edge wear and (4) increased vertical femoral offset.
Results The median vertical femoral offset with an S-ROM and Corail stem was 69 (63–82) mm and 70 (59–103) mm, respectively. There was no significant difference between the two stems (p = 0.44). Edge wearing of the cup occurred in five hips with an S-ROM and five with a Corail. The 11/13 S-ROM stem tapers had a median taper engagement length of 14 (10.5–14) mm at their longest point of engagement in the medial–lateral plane (Figure 3(a)) and were significantly greater (p \ 0.001) than the 12/14 Corail stem tapers, which had a median taper engagement length of 10.5 (10–11) mm. The median taper engagement length of the S-ROM tapers in the regions of the scallops was comparable to that of the Corail stems. Figure 3 illustrates the differences in taper engagement length and position between the two stem designs.
Visual assessment of corrosion All head tapers were found to have evidence of corrosion; statistical analysis revealed that the differences were not significant (p = 0.0769).
Measurement of bearing surface material loss The median material loss of the combined cup and head bearing surfaces of hips with an S-ROM stem it was 3.92 (1.20–7.81) mm3/year, while with a Corail stem, it was 3.21 (0.87–62.12) mm3/year (Table 2). There was no significant difference between the two groups (p = 0.3192).
Measurement of head taper material loss The median material loss rate of the head tapers with an S-ROM was 0.132 (0.015–0.518) mm3/year, while with a Corail it was 0.238 (0.0002–2.178) mm3/year (Figure 4). The Corail group was found to have significantly more material loss than the S-ROM group (p = 0.036).
Stem taper analysis Figures 5(a) and 5(b) presents the taper surfaces of the S-Rom and Corail stems, respectively, as viewed under the optical microscope. The median (range) thread heights for the single S-ROM stem and two Corail stems retrieved in this study were 1 (1–1.1) mm and 11.5 (11.3–11.7) mm, respectively (using three different scan regions on each taper). The median (range) thread spacing for the S-ROM and Corail stems was 0.099 (0.955–0.102) mm and 0.2 (0.193–0.228) mm,
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
94
Proc IMechE Part H: J Engineering in Medicine 229(1)
Figure 3. Schematic illustrations presenting the taper engagement of the S-ROM stem in the (a) medial–lateral and (b) anterior– posterior views. The shorter engagement length of the scalloped regions can be seen. The Corail stem is presented in the (c) distal– proximal and (d) anterior–posterior views (examples of points of elevated stress due to toggling are circled).
Table 2. Median (range) material loss values, taper engagement length, presence of edge wearing and vertical femoral offset.
Taper engagement length (mm) Vertical femoral offset (mm) Cup edge worn? Yes:no Bearing surface material loss (mm3/year) Taper surface material loss (mm3/year)
respectively. This revealed that Corail taper had a distinctly rougher surface topography than the S-ROM stem.
Discussion The release of cobalt and chromium metal ions from the taper junction of large diameter (536 mm) MOMTHRs is speculated to contribute to the high revision rates of these hip systems. A number of retrieval studies have reported on a wide range of material loss at the
S-ROM group
Corail group
14 (10.5–14) 69 (63–82) 5:5 3.92 (1.20–7.81) 0.132 (0.015–0.518)
10.5 (10–11) 70 (59–103) 5:5 3.21 (0.87–62.12) 0.238 (0.0002–2.178)
surfaces of modular junctions10,12,13 together with evidence of substantial corrosion;14–16 this has been associated with periprosthetic soft tissue damage in patients.17–20 It is speculated that the underlying cause of increased mechanical wear and corrosion at the head taper is multifactorial. In order to better understand the clinical impact of an individual factor, there is considerable benefit in employing a case–control design in retrieval analysis whereby hips are selected such that they are statistically matched in relation to other influencing
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
95
factors. In this study, we reported on the effect of stem type (longer, smoother S-ROM taper vs shorter, rougher Corail taper) on head taper material loss. This study was limited by the comparatively low numbers of hips in each group; however, our inclusion criteria meant that the two groups were closely matched with respect to gender, age, time to revision, neck length and inclination. Furthermore, no significant differences in bearing surface wear, cup edge wearing or vertical femoral offset were found, which are thought to contribute to differences in taper material loss. As such we are able to draw clinically significant conclusions from our study regarding the impact of stem design. We found that head tapers that had been paired with Corail stems had significantly greater rates of material loss per year than heads paired with S-ROM stems. Additionally, we found a statistically significant difference in the taper engagement length: a median of 10.5 mm for the Corail group compared with 14 mm for the S-ROM group at its greatest length of engagement. While the S-ROM taper was engaged up to the opening
Figure 4. Distribution of head taper material loss values for the two stem groups.
of the head taper (excluding scalloped regions), the Corail stem taper in all cases was found to be seated fully within the head taper, such that at the base of the stem, taper was in contact with the head taper surface at a distance from the taper edge, Figure 3. This positioning of the stem taper may increase the susceptibility of it toggling within the head taper, resulting in localised regions of increased contact stresses (circled in Figure 3(b)), suggesting a pattern of mechanical wear not observed with the longer S-ROM taper. The significantly greater pre-revision Co/Cr ratios in the Corail group suggest that the taper junction in these hips would have corroded more, with much of the Co ions being retained on the surface of the surface, while much of the Cr ions would be released into the blood. The comparable volumes of material loss measured at the bearing surfaces of each group support differences in Co/Cr ratios as being due to corrosion at the taper junction. Cooper et al.20 reported high Co/Cr ratios in hips with severely corroded modular junctions and nonMOM bearings, while Hart et al.21 found implant corrosion to be associated with higher Co than Cr in periprosthetic tissue. This, however, was not reflected in the visual corrosion scoring of head tapers in this study; however, this may be due to the subjective nature of this scoring method. The 12/14 head tapers were found to have evidence of imprinting of the rougher thread of the Corail stem taper; this was not observed in S-ROM tapers. It has been suggested that this is due to galvanic corrosion in which the head taper, rather than the stem taper, is preferentially corroded.10 This corrosion mechanism coupled with an increase in the risk of toggling and edge loading of the shorter stem taper (when fully seated inside the head taper) may explain the greater material loss observed at these junctions.
Figure 5. Image generated via the optical profilometer of the (a) smooth S-ROM taper and (b) rough Corail taper.
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
96
Proc IMechE Part H: J Engineering in Medicine 229(1)
The findings of our study are in agreement with those reported by Langton et al.,6 and the magnitude of head taper material loss is comparable. While notably larger volumes of material loss were measured at the bearing surface than the head taper in this study, it has been speculated that the metal debris from the taper junction may be more potent in causing damage to soft tissue, suggesting that the comparatively low volumes of material loss may still be clinically relevant.
Future work While the two stems in this study demonstrate clear differences in their taper surface topography and geometry, future work will examine the significance of more subtle variations in surface roughness and size on head taper material loss. Furthermore, there is some debate regarding the clinical impact of the mixing and matching of components from different manufacturers in THRs. It is speculated that stem–head taper incompatibility due to varying manufacturing tolerances may impact on taper junction performance22 and future work will examine this issue. Corrosion of the modular neck junction in THRs has been reported previously23 and we will examine the significance of this in relation to the stem–head junction.
Conclusion This was a closely matched case–control study involving a single femoral head design paired with two distinctly different stem tapers from the same manufacturer. We showed that shorter and rougher stems can increase the volume of taper material loss; the consequential release of metal ions from this junction has been implicated in implant failure. When using a metal head, we recommend the use of stems with smooth and long tapers to minimise the clinical impact of damage at the taper junction. Acknowledgements We are grateful for the support of Gwynneth Lloyd, Charlotte Page and Elizabeth Ellis for their coordination of the retrieval centre; Bob Skinner and Siva Mahindan for their support in metrology; Erica Cook for statistical analysis and Christian Klemt for assistance with the analysis of plain radiographs. Declaration of conflicting interests The authors declare that there is no conflict of interest. Funding Two authors received funding from the British Orthopaedic Association through an industry consortium of nine manufacturers: DePuy International Ltd, Zimmer GmbH, Smith & Nephew UK Ltd, Biomet UK Ltd, JRI Ltd, Finsbury Orthopaedics Ltd, Corin
Group PLC, Mathys Orthopaedics Ltd and Stryker UK Ltd. References 1. Smith AJ, Dieppe P, Vernon K, et al. Failure rates of stemmed metal-on-metal hip replacements: analysis of data from the National Joint Registry of England and Wales. Lancet 2012; 379(9822): 1199–1204. 2. National Joint Registry for England and Wales (NJR) 10th annual report, 2013, www.njrcentre.org.uk 3. Munir S, Imbuldeniya A and Walter WL. Variations in the taper surface topography between 11 different commercially available hip replacement stems. Presented at the annual meeting of the American academy of orthopaedic surgeons, New Orleans, LA, 11–15 March 2014. 4. Nassif NA, Nawabi DH, Stoner K, et al. Taper design affects failure of large-head metal-on-metal total hip replacements. Clin Orthop Relat Res 2013; 472: 564–571. 5. Panagiotidou A, Meswania J, Hua J, et al. Enhanced wear and corrosion in modular tapers in total hip replacement is associated with the contact area and surface topography. J Orthop Res 2013; 31(12): 2031–2039. 6. Langton D, Nargol A, Sayginer O, et al. Taper failure: wear is the corrosion? Presented at the annual meeting of the British hip society, Exeter, 5–7 March 2014. 7. Matthies AK, Cro S, Bills P, et al. Which factors determine the volume of material lost from the taper junction of metal-on-metal hip replacements? Presented at the annual meeting of the American academy of orthopaedic surgeons, New Orleans, LA, 11–15 March 2014. 8. Goldberg JR, Gilbert JL, Jacobs JJ, et al. A multicentre retrieval study of the taper interfaces of modular hip prostheses. Clin Orthop Relat Res 2002; 401: 149–161. 9. Hothi HS, Matthies AK, Berber R, et al. The reliability of a scoring system for corrosion and fretting, and its relationship to material loss of tapered, modular junctions of retrieved hip implant. J Arthroplasty 2014; 29(6): 1313–1317. 10. Matthies AK, Racasan R, Bills P, et al. Material loss at the taper junction of retrieved large head metal-on-metal total hip replacements. J Orthop Res 2013; 31(11): 1677– 1685. 11. Bills PJ, Racasan R, Underwood RJ, et al. Volumetric wear assessment of retrieved metal-on-metal hip prostheses and the impact of measurement uncertainty. Wear 2012; 274: 212–219. 12. Langton DJ, Sidaginamale R, Lord JK, et al. Taper junction failure in large-diameter metal-on-metal bearings. Bone Joint Res 2012; 1: 56–63. 13. Bolland BJ, Culliford DJ, Langton DJ, et al. High failure rate with a large diameter hybrid metal-on-metal total hip replacement: clinical, radiological and retrieval analysis. J Bone Joint Surg Br 2011; 93–B: 608–615. 14. Cross MB, Esposito C, Sokolova A, et al. Fretting and corrosion changes in modular total hip arthroplasty. Bone Joint J 2013; 95-B(Suppl. 15): 127. 15. Higgs G, Kurtz S, Hanzlik J, et al. Retrieval analysis of metal-on-metal hip prostheses: characterising fretting and corrosion at modular interfaces. Bone Joint J 2013; 95-B (Suppl. 15): 108. 16. Meyer H, Mueller T, Goldau G, et al. Corrosion at the cone/taper interface leads to failure of large-diameter
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
Hothi et al.
17.
18.
19.
20.
97
metal-on-metal total hip arthroplasties. Clin Orthop Relat Res 2012; 470(11): 3101–3108. Chana R, Esposito C, Campbell PA, et al. Mixing and matching causing taper wear: corrosion associated with pseudotumour formation. J Bone Joint Surg Br 2012; 94(2): 281–286. Fricka KB, Ho H, Peace WJ, et al. Metal-on-metal local tissue reaction is associated with corrosion of the head taper junction. J Arthroplasty 2012; 27(8): 26–31. Whitehouse MR, Endo M and Masri BA. Adverse local tissue reaction associated with a modular hip hemiarthroplasty. Clin Orthop Relat Res 2013; 471(12): 4082–4086. Cooper HJ, Urban RM, Wixson RL, et al. Adverse local tissue reaction arising from corrosion at the femoral
neck-body junction in a dual-taper stem with a cobaltchromium modular neck. J Bone Joint Surg Am 2013; 95: 865–872. 21. Hart AJ, Quinn PD, Lali F, et al. Cobalt from metal-onmetal hip replacements may be the clinically relevant active agent responsible for periprosthetic tissue reactions. Acta Biomater 2012; 8: 3865–3873. 22. Shareef N and Levine D. Effect of manufacturing tolerances on the micromotion at the morse taper interface in modular hip implants using the finite element technique. Biomaterials 1996; 17: 623–630. 23. Kop AM and Swarts E. Corrosion of a hip stem with a modular neck taper junction. J Arthroplasty 2009; 24(7): 1019–1023.
Downloaded from pih.sagepub.com at NORTH CAROLINA STATE UNIV on April 28, 2015
