Dual Mobility Bearings Withstand Loading From Steeper Cup-inclinations Without Substantial Wear LaQuawn Loving,1 Lizeth Herrera,1 Samik Banerjee,2 Christopher Heffernan,1 Jim Nevelos,1 David C. Markel,3 Michael A. Mont2 1
Stryker Orthopaedics, Mahwah, New Jersey, 2Rubin Institute for Advanced Orthopedics, Center for Joint Preservation and Replacement, Sinai Hospital of Baltimore, Baltimore, Maryland, 3Providence Hospital, DMC-Providence Residency, Detroit, Michigan Received 27 February 2014; accepted 20 October 2014 Published online 24 November 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22774
ABSTRACT: Steep cup abduction angles with adverse joint loading may increase traditional polyethylene bearing wear in total hip arthroplasties. However, there have been few reports evaluating the effect of cup inclination on the wear of dual-mobility devices. In a hip joint simulation, we compared the short-term wear of two-sizes of modular highly cross-linked dual-mobility bearings (28 mm femoral head diameter/42 mm polyethylene insert outer diameter/54 mm acetabular shell diameter; 22.2 mm femoral head diameter/ 36 mm polyethylene insert outer diameter/48 mm acetabular shell diameter) at 50 and 65˚ of cup inclination with modular 28 mm femoral head on 54 mm cup diameter metal-on-highly cross-linked polyethylene bearings. Increasing inclination from 50–65˚ had no changes in volumetric wear of 28/42/54 mm (mean, 1.7 vs. 1.2 mm3/million cycles, respectively; p ¼ 0.50) and 22.2/36/48 mm (mean, 1.7 vs. 1.2 mm3/million cycles, respectively; p ¼ 0.48) dual mobility bearings. At 65˚, 22.2/36/48 mm dual-mobility bearings had lower volumetric loss (mean, 2.2 vs. 6.3 mm3; p ¼ 0.03) and wear rates (mean, 1.2 vs. 2.7 mm3/million cycles; p ¼ 0.02) compared to metal-onhighly cross-linked polyethylene bearings. Modern-generation dual-mobility designs with highly cross-linked polyethylenes may potentially withstand edge-loading from steeper cup-inclinations without substantial decreases in wear. ß 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 33:398–404, 2015. Keywords: dual-mobility; wear; arthroplasty; acetabular; inclination
Instability is a frustrating complication following total hip arthroplasty (THA) and has been reported to be the third most common cause of revision in the Swedish registry between the years 1979 and 2000.1 It is also reported as one of the primary causes of revision surgery in the US following THA.2 Dislocations are a high economic burden with the overall cost of revision surgery for instability reported to be as high as 150% of that of a primary total hip arthroplasty.3 Various strategies have been developed to reduce the risks of instability after primary and revision total hip arthroplasty such as the use of elevated rims, large diameter femoral heads, tripolar articulations,4,5 and constrained liners.5–7 Of these, use of large diameter femoral heads has been the most commonly used strategies in contemporary practice to augment stability. However, greater volumetric polyethylene wear may occur with the use of 36- and 40mm heads despite their improved efficacy in preventing dislocations.8 Moreover, the use of large diameter metal-on-metal articulations as an approach to prevent instability has also markedly diminished due to concerns about adverse local tissue reactions, pseudotumors, and metallosis.9–11 Bousquet and Grammont in 1972 introduced dual mobility bearings in total hip arthroplasty with the aim of reducing the incidence of dislocations.12 These bearings potentially minimize the risks of instability by using large diameter mobile polyethylene inserts with a captured small diameter head.13,14 The large Conflict of interest: None. Correspondence to: Michael A. Mont (T: 410-601-8500, F: 410601-8501; E-mail: [email protected] and [email protected]) # 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
398
JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2015
polyethylene insert functions as a large head, which increases the jump-distance13 needed to dislocate the femoral head from the acetabular cup. There has been limited global acceptance of dual mobility hip bearings, although there are multiple reports of improved stability and durable implant survivorships of these bearings from Europe, and especially from Lyon and Saint Etienne in France.15–19 Widespread acceptance of the dual mobility concept may have been stunted due to concerns of long-term wear and osteolysis from historic gamma-air sterilized ultrahigh molecular weight polyethylene (UHMWPE) bearings commonly employed with this design.20,21 With the advent of highly cross-linked polyethylene with improved wear properties, there has been renewed interest with these bearings in the past decade. However, there is limited evidence reported in literature on the wear properties of the currently available dual mobility designs, especially with potentially adverse joint loading conditions such as found with steep cup inclination angles. In vitro hip simulator studies have been previously reported to accurately predict the in vivo performance of different bearing materials and designs by closely reproducing the in vivo environment.22 Therefore, in this study, we aimed to assess the in vitro wear characteristics of a current generation dual mobility bearing design’ using a hip joint wear simulator. Specifically, we evaluated the effect of two different acetabular cup inclination angles (50 and 65˚) on the volumetric loss and the wear rates of: (1) 22.2/36/ 48 mm dual mobility bearings; (2) 28/42/54 mm dual mobility bearings; and (3) compared these with the wear properties of metal-on-highly cross-linked polyethylene articulations of 28 femoral head diameters on 54 mm shell sizes.
DUAL MOBILITY BEARINGS
399
MATERIALS AND METHODS Implants Tested The dual mobility components used were the Modular Dual Mobility (MDM) Mobile Bearing Hip System (Stryker, Mahwah, NJ). The outer diameters of the dual mobility highlycross linked polyethylene inserts were 42 and 36 mm which articulated against 42 and 36 mm polished CoCr acetabular liners, respectively. The inner diameter of the dual mobility polyethylene inserts with the smallest thickness that would articulate against Co-Cr femoral head diameters of 28 and 22.2 mm were chosen. The acetabular shell sizes were 54 and 48 mm, respectively for these femoral head sizes (see Table 1). The metal-on-highly cross-linked polyethylene control group used 28 mm CoCr heads with outer shell diameters of 54 mm (28/54 mm sizes in both comparison testing) and corresponding polyethylene diameters of 9.9 mm. All acetabular shells used were made of Titanium alloy which had a plasma-spray porous coat of commercially pure titanium and hydroxyapatite on the outer surface These sizes were chosen as they represented the most commonly used acetabular cup diameters in men and women in our patient population. The polyethylene components in the dual mobility and control group were made of highly cross-linked ultra-high molecular weight polyethylene made from compression molded GUR 1020 which had been sequentially irradiated and annealed three times at a dosage of 30 kGy to a total dose of 90 kGy (X3 Stryker, Mahwah, NJ).23 The current study evaluated polyethylene inserts in the dual mobility group that would potentially exhibit the highest stresses for a given femoral head size. Hence, our inserts have the smallest diametrical difference between the inner and outer diameters for a given head size; lower forces/ torques are necessary to initiate movement at the outer diameter of the polyethylene insert. This increased movement at the outer bearing may increase the total sliding distance thus increasing wear and enabling a better estimation of worst case wear behaviors.
Figure 1. Illustration depicting the hip simulator set-up.
Machine and Kinematics A comparative wear analysis between dual mobility and metal-on-highly cross-linked polyethylene bearings was performed using a multi-station MTS hip joint wear simulator (MTS, Eden Prairie, MN). In the hip simulator set-up, the acetabular component was mounted superiorly to a fixed block at inclination angle of 50 and 65˚ to the horizontal with neutral anteversion. The cup remained static during the entire experiment. In addition, the acetabular component was mounted in a fashion that it remained in the same vertical axis as the femoral component (see Fig. 1). The femoral component was fixed below to an inclined block at 23˚ which moved at a frequency of 1 Hz providing a composite motion of abduction-adduction, flexion-extension, and rotation. This angle was chosen to represent human level walking gait and
has been based on previously developed ISO 14242-3:2009 protocols.24 There is no anteversion with this hip simulator setup. Although this remains a limitation of the simulator set up, applying anteversion does not change the mode of bearing contact with this hip simulator. A cyclic compressive load was applied axially with a maximum of 2,450 N following the physiological walking gait profile determined by Paul et al.25 The testing method represented a standard gait cycle which included swing phase. The high angle of 65˚ represents an edge loading condition. This inclination has been consistently produces runaway wear associated with edge loaded metal-onmetal components. However, micro-separation has not been found to occur during this testing condition.
Table 1. Details of Sizes of Dual Mobility Bearings Used in the Experiment
Experiment Phase 1 Phase 2
Size of Inner Diameter of Femoral Head (mm)
Size of Outer Diameter of Polyethylene Insert & Inner Diameter of CoCr Liner (mm)
Size of Hemispherical Outer Acetabular Shell (mm)
Polyethylene Thickness (mm)
28 22.2
42 36
54 48
6.8 6.7
JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2015
400
LOVING ET AL.
A specimen chamber surrounded this construct and allowed for lubricant submersion (approx. 450 ml). The lubricant used was alpha calf serum (Hyclone Alpha Calf Fraction, Logan, UT), diluted to 50% using a pH-balanced 20 mmol solution of deionized water and ethylenediaminetetraacetic acid (EDTA) to obtain a physiological relevant protein level (approx. 20 g/dl). EDTA was added to retard serum decomposition and the solution was filtered through a 20 mm filter before use. All fixtures used in the study were made of non-corrosive constituents and all components and fixtures were cleaned ultrasonically prior to testing. Experimental Conditions and Measurement Techniquestesting Was Conducted in Two Comparison Groups Three sets of dual mobility implants of size 28/42/54 mm were tested at acetabular inclination angles of 50 and 65˚, and three sets metal-on-highly cross-linked bearings were tested at 50 and 65˚ during 28 mm dual mobility component evaluation. Four sets of dual mobility implants of size 22.2/ 36/48 mm were tested at acetabular inclination angles of 50 and 65˚ during 22 mm dual mobility component evaluation. Additionally, one set of each of the four conditions from the 28 mm dual mobility component evaluation were repeated during 22 mm dual mobility component testing. Therefore, there was a sample size of four for each testing cohort. One metal-on-highly cross-linked polyethylene 50˚ control sample was lost from the analysis due to a mechanical failure of the setup; all other components completed the duration of testing. Based on an institutional database of over two hundred wear tested highly cross-linked polyethylene samples against cobalt-chrome heads with a standard deviation of 1.5 mm3/mc, a sample size of three is needed to obtain a power of 80% for the effect size of 8 mm3/mc. The true power is 99% for a sample size of three. All polyethylene components were pre-soaked in deionized water for a minimum of 7 days at 37˚C. The femoral heads were captured inside the polyethylene liner by pressing the liner with a clamp and slowly inserting the femoral head into the insert. The vice-grip was slowly closed until full seating occurred and the polyethylene liner freely rotated about the femoral head. All tools that were in direct contact with the samples were coated with a multi-purpose rubber coating or equivalent to prevent damage to the samples. During each of the two components wear analysis, testing was stopped after every 500,000 cycles for cleaning of components, changing of lubricant solution, and gravimetric wear analysis of acetabular inserts. Weight readings were also obtained at beginning of testing. The testing results presented are representative of those weight readings. Soak specimens were used to correct for net weight gain due to fluid absorption in the polyethylene liners and the soak control absorption data used to calculate reported wear
values in this study.26 Soak-corrected weight loss data was converted to volumetric data by dividing by material density of polyethylene. Volumetric loss was then plotted as a function of cycle count and linear regression was used to determine wear rates based on previously described ISO14242-2 and ASTM F2025 protocols.26,27 The samples were loaded with the same axial compression waveform applied to the wear testing samples. Fixturing provided for rotational and translational alignment to eliminate off-axis loading or motion such that dynamic loading was applied with no motion. Specimen chambers allowed for lubricant submersion of the component assemblies. One sample from each testing group served as the statically heat-controlled soak. All data collected was integrated into an Excel spreadsheet (Excel, Microsoft Corporation, Redmond, WA) for the final analysis. A Student t-test was used to determine significance between the mean wear measurements obtained using a statistical program (GraphPad 6.0, Inc., La Jolla, CA). A p-value
Stryker Orthopaedics, Mahwah, New Jersey, 2Rubin Institute for Advanced Orthopedics, Center for Joint Preservation and Replacement, Sinai Hospital of Baltimore, Baltimore, Maryland, 3Providence Hospital, DMC-Providence Residency, Detroit, Michigan Received 27 February 2014; accepted 20 October 2014 Published online 24 November 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22774
ABSTRACT: Steep cup abduction angles with adverse joint loading may increase traditional polyethylene bearing wear in total hip arthroplasties. However, there have been few reports evaluating the effect of cup inclination on the wear of dual-mobility devices. In a hip joint simulation, we compared the short-term wear of two-sizes of modular highly cross-linked dual-mobility bearings (28 mm femoral head diameter/42 mm polyethylene insert outer diameter/54 mm acetabular shell diameter; 22.2 mm femoral head diameter/ 36 mm polyethylene insert outer diameter/48 mm acetabular shell diameter) at 50 and 65˚ of cup inclination with modular 28 mm femoral head on 54 mm cup diameter metal-on-highly cross-linked polyethylene bearings. Increasing inclination from 50–65˚ had no changes in volumetric wear of 28/42/54 mm (mean, 1.7 vs. 1.2 mm3/million cycles, respectively; p ¼ 0.50) and 22.2/36/48 mm (mean, 1.7 vs. 1.2 mm3/million cycles, respectively; p ¼ 0.48) dual mobility bearings. At 65˚, 22.2/36/48 mm dual-mobility bearings had lower volumetric loss (mean, 2.2 vs. 6.3 mm3; p ¼ 0.03) and wear rates (mean, 1.2 vs. 2.7 mm3/million cycles; p ¼ 0.02) compared to metal-onhighly cross-linked polyethylene bearings. Modern-generation dual-mobility designs with highly cross-linked polyethylenes may potentially withstand edge-loading from steeper cup-inclinations without substantial decreases in wear. ß 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 33:398–404, 2015. Keywords: dual-mobility; wear; arthroplasty; acetabular; inclination
Instability is a frustrating complication following total hip arthroplasty (THA) and has been reported to be the third most common cause of revision in the Swedish registry between the years 1979 and 2000.1 It is also reported as one of the primary causes of revision surgery in the US following THA.2 Dislocations are a high economic burden with the overall cost of revision surgery for instability reported to be as high as 150% of that of a primary total hip arthroplasty.3 Various strategies have been developed to reduce the risks of instability after primary and revision total hip arthroplasty such as the use of elevated rims, large diameter femoral heads, tripolar articulations,4,5 and constrained liners.5–7 Of these, use of large diameter femoral heads has been the most commonly used strategies in contemporary practice to augment stability. However, greater volumetric polyethylene wear may occur with the use of 36- and 40mm heads despite their improved efficacy in preventing dislocations.8 Moreover, the use of large diameter metal-on-metal articulations as an approach to prevent instability has also markedly diminished due to concerns about adverse local tissue reactions, pseudotumors, and metallosis.9–11 Bousquet and Grammont in 1972 introduced dual mobility bearings in total hip arthroplasty with the aim of reducing the incidence of dislocations.12 These bearings potentially minimize the risks of instability by using large diameter mobile polyethylene inserts with a captured small diameter head.13,14 The large Conflict of interest: None. Correspondence to: Michael A. Mont (T: 410-601-8500, F: 410601-8501; E-mail: [email protected] and [email protected]) # 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.
398
JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2015
polyethylene insert functions as a large head, which increases the jump-distance13 needed to dislocate the femoral head from the acetabular cup. There has been limited global acceptance of dual mobility hip bearings, although there are multiple reports of improved stability and durable implant survivorships of these bearings from Europe, and especially from Lyon and Saint Etienne in France.15–19 Widespread acceptance of the dual mobility concept may have been stunted due to concerns of long-term wear and osteolysis from historic gamma-air sterilized ultrahigh molecular weight polyethylene (UHMWPE) bearings commonly employed with this design.20,21 With the advent of highly cross-linked polyethylene with improved wear properties, there has been renewed interest with these bearings in the past decade. However, there is limited evidence reported in literature on the wear properties of the currently available dual mobility designs, especially with potentially adverse joint loading conditions such as found with steep cup inclination angles. In vitro hip simulator studies have been previously reported to accurately predict the in vivo performance of different bearing materials and designs by closely reproducing the in vivo environment.22 Therefore, in this study, we aimed to assess the in vitro wear characteristics of a current generation dual mobility bearing design’ using a hip joint wear simulator. Specifically, we evaluated the effect of two different acetabular cup inclination angles (50 and 65˚) on the volumetric loss and the wear rates of: (1) 22.2/36/ 48 mm dual mobility bearings; (2) 28/42/54 mm dual mobility bearings; and (3) compared these with the wear properties of metal-on-highly cross-linked polyethylene articulations of 28 femoral head diameters on 54 mm shell sizes.
DUAL MOBILITY BEARINGS
399
MATERIALS AND METHODS Implants Tested The dual mobility components used were the Modular Dual Mobility (MDM) Mobile Bearing Hip System (Stryker, Mahwah, NJ). The outer diameters of the dual mobility highlycross linked polyethylene inserts were 42 and 36 mm which articulated against 42 and 36 mm polished CoCr acetabular liners, respectively. The inner diameter of the dual mobility polyethylene inserts with the smallest thickness that would articulate against Co-Cr femoral head diameters of 28 and 22.2 mm were chosen. The acetabular shell sizes were 54 and 48 mm, respectively for these femoral head sizes (see Table 1). The metal-on-highly cross-linked polyethylene control group used 28 mm CoCr heads with outer shell diameters of 54 mm (28/54 mm sizes in both comparison testing) and corresponding polyethylene diameters of 9.9 mm. All acetabular shells used were made of Titanium alloy which had a plasma-spray porous coat of commercially pure titanium and hydroxyapatite on the outer surface These sizes were chosen as they represented the most commonly used acetabular cup diameters in men and women in our patient population. The polyethylene components in the dual mobility and control group were made of highly cross-linked ultra-high molecular weight polyethylene made from compression molded GUR 1020 which had been sequentially irradiated and annealed three times at a dosage of 30 kGy to a total dose of 90 kGy (X3 Stryker, Mahwah, NJ).23 The current study evaluated polyethylene inserts in the dual mobility group that would potentially exhibit the highest stresses for a given femoral head size. Hence, our inserts have the smallest diametrical difference between the inner and outer diameters for a given head size; lower forces/ torques are necessary to initiate movement at the outer diameter of the polyethylene insert. This increased movement at the outer bearing may increase the total sliding distance thus increasing wear and enabling a better estimation of worst case wear behaviors.
Figure 1. Illustration depicting the hip simulator set-up.
Machine and Kinematics A comparative wear analysis between dual mobility and metal-on-highly cross-linked polyethylene bearings was performed using a multi-station MTS hip joint wear simulator (MTS, Eden Prairie, MN). In the hip simulator set-up, the acetabular component was mounted superiorly to a fixed block at inclination angle of 50 and 65˚ to the horizontal with neutral anteversion. The cup remained static during the entire experiment. In addition, the acetabular component was mounted in a fashion that it remained in the same vertical axis as the femoral component (see Fig. 1). The femoral component was fixed below to an inclined block at 23˚ which moved at a frequency of 1 Hz providing a composite motion of abduction-adduction, flexion-extension, and rotation. This angle was chosen to represent human level walking gait and
has been based on previously developed ISO 14242-3:2009 protocols.24 There is no anteversion with this hip simulator setup. Although this remains a limitation of the simulator set up, applying anteversion does not change the mode of bearing contact with this hip simulator. A cyclic compressive load was applied axially with a maximum of 2,450 N following the physiological walking gait profile determined by Paul et al.25 The testing method represented a standard gait cycle which included swing phase. The high angle of 65˚ represents an edge loading condition. This inclination has been consistently produces runaway wear associated with edge loaded metal-onmetal components. However, micro-separation has not been found to occur during this testing condition.
Table 1. Details of Sizes of Dual Mobility Bearings Used in the Experiment
Experiment Phase 1 Phase 2
Size of Inner Diameter of Femoral Head (mm)
Size of Outer Diameter of Polyethylene Insert & Inner Diameter of CoCr Liner (mm)
Size of Hemispherical Outer Acetabular Shell (mm)
Polyethylene Thickness (mm)
28 22.2
42 36
54 48
6.8 6.7
JOURNAL OF ORTHOPAEDIC RESEARCH MARCH 2015
400
LOVING ET AL.
A specimen chamber surrounded this construct and allowed for lubricant submersion (approx. 450 ml). The lubricant used was alpha calf serum (Hyclone Alpha Calf Fraction, Logan, UT), diluted to 50% using a pH-balanced 20 mmol solution of deionized water and ethylenediaminetetraacetic acid (EDTA) to obtain a physiological relevant protein level (approx. 20 g/dl). EDTA was added to retard serum decomposition and the solution was filtered through a 20 mm filter before use. All fixtures used in the study were made of non-corrosive constituents and all components and fixtures were cleaned ultrasonically prior to testing. Experimental Conditions and Measurement Techniquestesting Was Conducted in Two Comparison Groups Three sets of dual mobility implants of size 28/42/54 mm were tested at acetabular inclination angles of 50 and 65˚, and three sets metal-on-highly cross-linked bearings were tested at 50 and 65˚ during 28 mm dual mobility component evaluation. Four sets of dual mobility implants of size 22.2/ 36/48 mm were tested at acetabular inclination angles of 50 and 65˚ during 22 mm dual mobility component evaluation. Additionally, one set of each of the four conditions from the 28 mm dual mobility component evaluation were repeated during 22 mm dual mobility component testing. Therefore, there was a sample size of four for each testing cohort. One metal-on-highly cross-linked polyethylene 50˚ control sample was lost from the analysis due to a mechanical failure of the setup; all other components completed the duration of testing. Based on an institutional database of over two hundred wear tested highly cross-linked polyethylene samples against cobalt-chrome heads with a standard deviation of 1.5 mm3/mc, a sample size of three is needed to obtain a power of 80% for the effect size of 8 mm3/mc. The true power is 99% for a sample size of three. All polyethylene components were pre-soaked in deionized water for a minimum of 7 days at 37˚C. The femoral heads were captured inside the polyethylene liner by pressing the liner with a clamp and slowly inserting the femoral head into the insert. The vice-grip was slowly closed until full seating occurred and the polyethylene liner freely rotated about the femoral head. All tools that were in direct contact with the samples were coated with a multi-purpose rubber coating or equivalent to prevent damage to the samples. During each of the two components wear analysis, testing was stopped after every 500,000 cycles for cleaning of components, changing of lubricant solution, and gravimetric wear analysis of acetabular inserts. Weight readings were also obtained at beginning of testing. The testing results presented are representative of those weight readings. Soak specimens were used to correct for net weight gain due to fluid absorption in the polyethylene liners and the soak control absorption data used to calculate reported wear
values in this study.26 Soak-corrected weight loss data was converted to volumetric data by dividing by material density of polyethylene. Volumetric loss was then plotted as a function of cycle count and linear regression was used to determine wear rates based on previously described ISO14242-2 and ASTM F2025 protocols.26,27 The samples were loaded with the same axial compression waveform applied to the wear testing samples. Fixturing provided for rotational and translational alignment to eliminate off-axis loading or motion such that dynamic loading was applied with no motion. Specimen chambers allowed for lubricant submersion of the component assemblies. One sample from each testing group served as the statically heat-controlled soak. All data collected was integrated into an Excel spreadsheet (Excel, Microsoft Corporation, Redmond, WA) for the final analysis. A Student t-test was used to determine significance between the mean wear measurements obtained using a statistical program (GraphPad 6.0, Inc., La Jolla, CA). A p-value
