The Journal of Arthroplasty 30 (2015) 135–140
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Correlation of Corrosion and Biomechanics in the Retrieval of a Single Modular Neck Total Hip Arthroplasty Design Modular Neck Total Hip Arthroplasty System Brent A. Lanting, MD, FRCSC a, 1, Matthew G. Teeter, PhD a, 1, Edward M. Vasarhelyi, MD, MSc, FRCSC a, Todor G. Ivanov, PhD b, James L. Howard, MD, MSc, FRCSC a, Douglas D.R. Naudie, MD, FRCSC a a b
Division of Orthopaedic Surgery, London Health Sciences Centre, London, Ontario, Canada Imaging Research Laboratories, Robarts Research Institute, London, Ontario, Canada
a r t i c l e
i n f o
Article history: Received 14 February 2014 Accepted 13 June 2014 Keywords: metallosis modular total hip modular neck pseudotumor
a b s t r a c t Increased modularity of total hip arthroplasty components has occurred, with theoretical advantages and disadvantages. Recent literature indicates the potential for elevated revision rates of modular neck systems and the potential for local pseudotumor and metallosis formation at the modular neck/stem site. Retrieval analysis of one modular neck implant design including SEM (SCANNING ELECTRON MICROSCOPY) assessment was done and correlated with FEA (finite element analysis) as well as clinical features of patient demographics, implant and laboratory analysis. Correlation of the consistent corrosion locations to FEA indicates that the material and design features of this system may result in a biomechanical reason for failure. The stem aspect of the modular neck/stem junction may be at particular risk. © 2014 Elsevier Inc. All rights reserved.
Total hip arthroplasty (THA) is a common procedure, with excellent longevity and patient satisfaction. However, advances continue to be proposed to improve implant longevity, patient satisfaction, and surgical technique. Implant innovations have led to increasing modularity from monoblock femoral components to hip systems with modular heads. In addition to modularity of the prosthetic head, several manufacturers have proposed modularity of the femoral neck. Modularity at this level was proposed to allow changes in version, length, offset and neck – shaft angle. This was theorized to be an attractive design innovation as it potentially allowed for better recreation of the patients' anatomy. Increased modularity also had surgical technique advantages, such as allowing increased opportunities for soft tissue balancing and leg length optimization, thereby potentially allowing for reduced surgical time. However, modular necks have been cause for some concern clinically, and there are reports of corrosion, metallosis [1] and modular neck fracture [2]. There have also been reports of higher revision rates, with pseudotumor and catastrophic mechanical failure being implicated in registry data [3]. This has led to the market withdrawal or recall of some implants. However, other products continue to be available. The purpose of this retrieval study was to examine modes of failure of a single implant design with a modular
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.06.009. Reprint requests: Brent A. Lanting, MD, FRCSC, Division of Orthopaedic Surgery, London Health Sciences Centre, London, Ontario, Canada. 1 Denotes equal contribution. http://dx.doi.org/10.1016/j.arth.2014.06.009 0883-5403/© 2014 Elsevier Inc. All rights reserved.
femoral neck. Visual classification of damage, scanning electron microscopy (SEM) assessment to assess for corrosion and metal transfer, and biomechanical assessment of the prosthesis using finite element (FE) modeling were performed. Methods Study Population All modular neck implants in our institutional implant retrieval lab were retrospectively reviewed. Approval for review of patient charts and implant retrieval analysis was obtained from the internal review board. A total of nineteen implants were identified to be of the same design of a dual taper Ti–Al–V (TMZF) femoral component, and a Co– Cr–Mo (Vitallium) modular femoral neck (Rejuvenate, Stryker, Mahwah, New Jersey) and examined after retrieval. All retrieved hips had Co–Cr heads (forged Vitallium) articulating on a highly crosslinked polyethylene liner. Details with regard to patient characteristics (Table 1), implant (Table 2), and laboratory analysis (Table 3) were collected. These implants were retrieved from fourteen females and five males, with a mean age of 65 years (range, 41 to 81 years) and a mean body mass index of 33.5 (range, 26.7 to 49.1 years). The diagnosis leading to the THA was avascular necrosis in two patients and osteoarthritis in the remainder of the patients. Ten of the twelve magnetic resonance imaging scans conducted prior to revision revealed cystic pseudotumors, with an average size of 8.2 × 5 × 2.7 cm (range, 12.7–1.8 cm). The implants were in situ for an average of 1.7 years (range, 0.8 to 3.1 years), and three of the retrieved implants were
136
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Table 1 Patient Demographics and Results of Cross Sectional Imaging of All Revised Implants. Case
Age
BMI
Time In Situ
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Average
81.0 57.6 60.1 64.5 66.7 78.3 81.1 75.2 60.8 68.9 60.3 65.9 41.4 60.5 41.1 71.3 72.6 66.6 60.8 65.0
32.4 33.8 33.9 26.0 33.7 30.8 34.0 26.1 52.0 28.2 43.1 26.7 32.1 49.1 31.1 30.7 38.1 34.9 31.4 34.1
1.3 1.8 2.2 1.2 2.4 3.1 1.9 2.0 2.1 1.8 2.0 1.5 1.6 1.1 1.0 1.3 1.1 0.8 3.0 1.7
Symptoms
Reason
MRI
Pain Pain Weakness Pain/Neurology Weakness Pain Weakness Pain Weakness Pain/Weakness Weakness Pain/Weakness Pain/Neurology Drainage Pain Infection Pain Pain Pain
metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis Pain Infection Infection Infection Subsidence Subsidence Subsidence Infection: 3 metallosis: 12 Subsidence: 2 Pain: 1
ALVAL N/A ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL N/A Neg N/A Neg N/A N/A N/A N/A
revised for infection. The average ESR was 25, CRP 12, chromium level of 0.7, cobalt 5.5, and titanium 3.1, with an average ratio of chromium to cobalt of 0.2:1. Visual Damage Scoring Each modular neck was examined visually using the method of Goldberg et al [4] for signs of corrosion and fretting by two of the authors (M.G.T. and B.A.L.). The necks were divided into four zones, corresponding to the superior, anterior, inferior, and posterior sides of the neck (Fig. 1). The marking arrow on the face of the trunion was taken as pointing to the superior zone, and all other zones followed in a consistent clockwise pattern. For each zone, corrosion and fretting were graded as none (grade 1), mild (grade 2), moderate (grade 3), or severe (grade 4) under stereomicroscope visualization (SZ-CTV, Olympus, Tokyo, Japan). Where differences in grading occurred, consensus was achieved after further discussion and evaluation. Table 2 Implant Specifications for All Retrieved Implants.
Case ID
Acetabular Size Head Size (mm) (length) (mm) Stem Size
Neck Length (mm)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
58 52 52 54 52 56 50 56 54 50 5 50 58 50 62 48 52 56a 58
30 30 34 30 30 34 30 34 30 34 30 30 34 30 34 34 34 34 34 Length 34: 10 Length 30: 10
a
36 36 36 (−5) 36 36 (−5) 32 32 (−4) 36 36 (−5) 32 32 (+4) 32 (+0) 40 32 36 32 (−4) 36 (+5) 36 (+5) 40 (+4) Size 32: 7 Size 36: 10 Size 40: 1
Porous 273 acetabular component.
9 7 8 7 9 7 7 9 9 7 8 7 8 7 8 7 7 7 8 Size 7: 10 Size 8: 5 Size 9: 4
Version (°) 0 0 8 16 0 8 0 0 8 8 0 0 0 8 0 0 0 8 0 0°: 12 8°: 6 16°: 1
Table 3 Laboratory Analysis of All Patients Prior to Revision.
Case ID
ESR (mm/h)
CRP (mg/L)
Chromium (μg/L)
Cobalt (μg/L)
Titanium (μg/L)
Ratio (Chr/Co)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 AVERAGE
34.0 38.0 25.0 90.0 9.0 16.0 32.0 6.0 14.0 1.0 20.0 7.0 5.0 38.0 27.0 44.0 16.0 28.0 7.0 24.1
13.3 5.7 7.2 28.2 2.9 1.7 8.1 6.4 10.1 9.0 6.0 1.2 1.2 54.4 16.0 35.8 4.3 4.9 2.6 11.5
0.4 0.2 1.0 N/A 1.9 0.8 0.9 1.3 1.0 0.4 1.2 N/A 0.4 N/A 0.9 0.6 0.3 N/A 0.4 0.8
3.2 2.8 2.7 N/A 5.7 15.7 9.5 9.9 6.3 9.8 4.6 N/A 2.4 N/A 4.2 3.6 0.5 N/A 2.4 5.5
1.6 2.3 5.0 N/A 3.8 2.3 3.2 4.3 5.8 1.9 2.2 N/A 2.5 N/A 1.7 3.5 1.9 N/A 3.5 3.0
0.1 0.1 0.4 N/A 0.3 0.1 0.1 0.1 0.2 0.0 0.3 N/A 0.2 N/A 0.2 0.2 0.6 N/A 0.2 0.2
Scanning Electron Microscopy The first six consecutively retrieved implants, along with two never-implanted reference specimens, were further analyzed with scanning electron microscopy (LEO 440 SEM, Carl Zeiss SMT Inc., Peabody, Massachusetts) and energy dispersive x-ray (Quartz Xone EDX system, Quartz Imaging Corporation, Vancouver, British Columbia) analysis (SEM/EDX). All six modular necks were examined using SEM/EDX for evidence of corrosion, fretting, and material transfer. In addition, one stem was sectioned in two planes at the neck–stem interface, and these sections also underwent SEM/EDX analysis. Finite Element Analysis The never-implanted reference specimens (size 7 and size 12 stems, 0 and 8 degree 30 mm necks, and 32 mm head) were laser scanned to generate 3D models for finite element analysis (FEA). The models were appropriately meshed and assembled in the FEA software (Abaqus, Dassault Systemes, Waltham, Massachusetts). The material properties applied to the components were a Poisson's ratio of 0.32 for the TMZF stems, and a Poisson's ratio of 0.30 for the Vitallium® necks and head [5–7]. The elastic moduli for TMZF and Vitallium vary in the literature, therefore we ran multiple models across these different moduli. These were 79 500, 100 000, and 110 000 MPa for the TMZF stem, and 200 000 and 240 000 MPa for the Vitallium neck and head [5,6,8]. The outer surface of the stem was fixed, and the coefficient of friction between the head and neck, and neck and stem was set as 1.00 to model zero motion between them. Simulating one body weight of an 80 kg person crossing the joint, an 800 N load was applied to the femoral head as a concentrated force, as has been done in other studies [7,9,10]. Von mises stress was measured from the simulations, virtual cross-sections between the segments were taken, and the regions of highest stress were probed for maximum values. Statistics Descriptive statistics (mean ± SD) were calculated for the corrosion and fretting damage scores. A D'Agostino and Pearson omnibus test for normality was used. As the scores were not normally distributed, Wilcoxon matched-pairs signed rank tests were used to compare the damage between zones.
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
137
Fig. 1. Three-dimensional model demonstrating the modular stem–neck junction (A), photograph of a modular neck with corrosion (B), and photograph of a cross-section through a stem with damage (C). Black arrows are the superior and inferior sides of the modular neck. White arrows are the damage scar within the neck–stem junction, along the edge that opposes the superior surface of the modular neck.
Results Visual Damage Scoring All of the retrieved stems had evidence of corrosion at the neck–stem interface (Fig. 1). There was no difference between the superior and inferior zones (P = 0.59) or between the anterior and posterior zones (P = 0.42) of the modular necks, therefore the superior and inferior zones were combined, as were the anterior and posterior zones. The corrosion was more severe (P = 0.01) on the superior/inferior zones (3.4 ± 0.8) than the anterior/posterior zones (3.0 ± 0.8). Fretting appeared mild or absent on the modular necks. There was no difference (P = 0.82) in the severity of the fretting between the superior/inferior zone (1.2 ± 0.5) and the anterior/posterior zone (1.2 ± 0.4). Scanning Electron Microscopy The SEM/EDX revealed the black corroded regions on the modular necks to contain chromium phosphate, a corrosion byproduct (Fig. 2). In some cases, transfer of titanium from the TMZF stem onto the modular necks was also seen (Fig. 2). The section taken from the stem revealed a marked shiny streak of metallic damage along the superior surface of the neck–stem interface (Fig. 3). Under SEM/EDX analysis, the streak contained higher than reference levels of iron, along with titanium and chromium phosphate (Fig. 3). Adjacent to this streak was a black ring of corrosion (running perpendicular to the metallic streak) that contained chromium phosphate (Fig. 3). Finite Element Analysis The pattern of stresses within FEA results was similar between the neutral and anteverted or retroverted necks, and between the two sizes of stems (Fig. 4). The maximum stress (Table 4) was consistently located on the superior and inferior sides of the neck, specifically at the superior-lateral and inferior-medial corners of the neck–stem interface. For the model of the small stem with straight neck, the mean maximum stress was 372.9 ± 14.0 MPa within the neck and 970.8 ± 60.4 MPa within the stem. Modeling an angled neck reduced the mean maximum stress in the neck to 167.9 ± 16.2 MPa, but increased the mean maximum stress within the stem to 1611.0 ±
118.1 MPa. For the model of the large stem with the straight neck, the mean maximum stress was 220.0 ± 2.4 MPa within the neck and 1611.7 ± 159.0 MPa within the stem.
Discussion Modularity at the neck–stem junction was initially designed and marketed to have patient and surgeon advantages. The goal was recreation of patient anatomy, and optimization of the THA position and balance with greater surgical ease. However, this additional level of modularity has been shown to have higher revision rates in some implants. The potential reasons for failure have not been well described in a clinical context. In this study, the retrieved implants were all of the same design, and were revised relatively early. Of all the retrieved implants, three were revised due to an infection. The relatively low chromium levels measured in these patients that required revision is notable. Literature has indicated that ion levels may be a way of monitoring metal on metal THA. In spite of the corrosion and metal transfer seen, the ion levels seen in these cases of modular neck THA that required revision are relatively low, with a maximum chromium level seen of 1.7 μg/L, cobalt 15.6 μg/L and titanium 5.8 μg/L. However, metallosis and significant soft tissue destruction were seen in twelve cases. Corrosion in the crevasse of the neck/stem junction may allow the low pH conditions favorable to the formation of hexavalent chromium [11]. Hexavalent chromium is known to be more toxic than trivalent chromium [11,12]. Although advanced techniques may allow for measurement of hexavalent chrome, these values were not available for this study [13]. The relatively lower values of chromium than cobalt seen are also notable, and have been seen in other studies [14]. The patient group described here had an average ratio of chromium to cobalt of 0.2:1. All retrieved implants that underwent SEM were consistent in demonstrating chromium phosphate, indicating corrosion as well as titanium transfer at the upper and lower edges of the neck. There were minimal changes on the sides of the modular neck. Stem analysis revealed more extensive changes (Fig. 3). Elevated iron, titanium and chromium phosphate demonstrate corrosion and metal transfer. The metal transfer suggests micro-motion at the
138
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Fig. 2. SEM/EDX results from a modular neck. The photograph of tested stem (A) is highlighted where the SEM images in B (yellow box) and C (red box) were obtained. The EDX spectrum from the first (yellow) region is shown in D, and the second (red) region is shown in E. Notable peaks of phosphorus (F) from chromium phosphate corrosion byproduct were found in the first region, and titanium transfer (G) from the stem to the modular neck was found in the second region.
neck–stem interface. These findings are in keeping with other reports in the literature [14,15]. Correlation of the FE model and the SEM analysis is of particular importance. The most severe corrosion was found along the superior and inferior surfaces of the modular necks, corresponding to the regions of greatest stress within the FE model. This is similar to FE modeling of other systems [16]. The concept of corrosion fatigue is not a new concept, but has been previously described with a resultant possibility of fracture [17]. It is not understood if the forces seen via the FE modeling have a role in the corrosion seen due to corrosion fatigue. Severe damage was seen within the neck–stem junction of the stem, along the superior surface, again corresponding to the greatest corrosion on the modular necks and the greatest stresses within the FE model. However, it is notable that the FE
model revealed stresses to be greater in the stem than in the neck. The elastic limits of Vitallium are approximately 500 to 850 MPa, whereas the FE models revealed the greatest stresses to be 372.9 MPa within the modular neck [5,6,8]. The elastic limit of TMZF is higher, at 1000 to 1060 MPa [5,6,8]. However stresses within the stem were also higher in the FE model, and at up to 1611.7 MPa, exceeding this elastic limit. This may explain the damage seen within the stem–neck junction of the stem, as well as the titanium transfer from the stem to the modular necks. The role of galvanic corrosion due to dissimilar metals is of interest, but the relative contribution of the dissimilar metals to the corrosion seen is not clear [17,18]. Notably, the FE model used in this study represents one specific loading scenario that has been used in other studies [7,9,10]. This loading scenario was applied to different
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
139
Fig. 3. SEM (A) and EDX (B) results from a cross-section of a stem (shown in Fig. 1C). The region had normal amounts of titanium (C), but high amounts of iron (D) as well as chromium (E) transferred from the modular neck.
geometries of neck and stem geometries. Different stresses may occur if different applied loads were modeled. The finding in this study of the stem aspect of the modular neck/ stem interface being potentially more impacted by the design is important. In cases of revision where the modular neck looks relatively undamaged, the potential for compromise of the stem
aspect of the junction must be considered, and warrants revision of this component as well. The strengths of this study are that it is a comprehensive study on a single implant design and examines the potential sources of failure of this implant. The use of advanced technologies allowed precise measurement of the retrieved implants included laser mapping and
Fig. 4. Cut-through of the FE model for the small stem with straight modular neck, using a color map to represent von Mises stress. Areas of peak stress are denoted with arrows, along the superior and inferior sides of the modular neck–stem junction.
140
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Table 4 Maximum Von Mises Stress (MPa) in the Stem and Neck Across the Different Models and Combination of Properties. Maximum Stress (MPa) Straight Neck, Small Stem
Angled Neck, Small Stem
Straight Neck, Large Stem
Assigned Head and Neck/Stem Elastic Moduli (MPa)
Stem
Neck
Stem
Neck
Stem
Neck
200 000/79 500 200 000/100 000 200 000/110 000 240 000/79 500 240 000/100 000 240 000/110 000 Mean Standard deviation
898.1 989.4 1029.0 894.6 986.8 1027.0 970.8 60.4
352.5 370.4 377.7 362.8 382.7 391.2 372.9 14.0
1692.0 1535.0 1459.0 1791.0 1621.0 1568.0 1611.0 118.1
154.2 182.7 185.2 150.9 154.7 179.8 167.9 16.2
1831.0 1699.0 1648.0 1634.0 1462.0 1396.0 1611.7 159.0
217.8 217.7 218.0 221.8 222.3 222.6 220.0 2.4
SEM analysis, and conservative loads of 800 N were applied in FE modeling. However, assumptions regarding of model geometries and fit between neck/stem may limit the accuracy of the absolute values reported for the stress. Conclusions In conclusion, SEM analysis of this particular design of a modular neck total hip system revealed corrosion at this additional junction of modularity. FE analysis indicates that the material and design features of this system may indicate a biomechanical reason for failure; particularly at the stem aspect of the modular neck/stem junction as the stem's material yield strength of loading may be exceeded under physiologic loads. Acknowledgments The authors thank Ross Davidson at Surface Science Western.
References 1. Gill IP, Webb J, Sloan K, et al. Corrosion at the neck–stem junction as a cause of metal ion release and pseudotumor formation. J Bone Joint Surg Br 2012;94(7):895. 2. Murugappan KS, Graves S. Outcome after modular neck hip arthroplasty: lessons from the registry. J Bone Joint Surg Br 2012;94-B(sXLI):112. 3. Atwood SA, Patten EW, Bozic KJ, et al. Corrosion-induced fracture of a doublemodular hip prosthesis: a case report. J Bone Joint Surg 2010;92(6):1522. 4. Goldberg JR, Gilbert JL, Jacobs JJ, et al. A multicenter retrieval study of the taper interfaces of modular hip prostheses. Clin Orthop Relat Res 2002;401:149. 5. Gracia L, Ibarz E, Puertolas S, et al. Study of bone remodelling of two models of femoral cementless stems by means of DEXA and finite elements. Biomed Eng Online 2010;9:22. 6. Gore D, Frazer RQ, Kovarik RE, et al. Vitallium. J Long Term Eff Med Implants 2005;15(6):673. 7. Theodorou EG, Provatidis CG, Megas PD. Investigating a total hip finite element model with respect to the offset and version changes, according to the modular neck type. J Serbian Soc Comput Mech 2009;3(2):1. 8. Moratal D. Finite element analysis—from biomedical applications to industrial developments. Croatia: InTech; 2012. 9. Krischak GD, Augat P, Beck A, et al. Biomechanical comparison of two side plate fixation techniques in an unstable intertrochanteric osteotomy model: sliding hip screw and percutaneous compression plate. Clin Biomech 2007;22(10):1112. 10. Nunn D, Freeman MA, Tanner KE, et al. Torsional stability of the femoral component of hip arthroplasty. Response to an anteriorly applied load. J Bone Joint Surg 1989;71(3):452. 11. Kotas J, Stasicka Z. Chromium occurrence in the environment and methods of its speciation. Environ Pollut 2000;107(3):263. 12. Levis AG, Majone F. Cytotoxic and clastogenic effects of soluble and insoluble compounds containing hexavalent and trivalent chromium. Br J Cancer 1981;44(2):219. 13. Sidaginamale RP, Joyce TJ, Lord JK, et al. Blood metal ion testing is an effective screening tool to identify poorly performing metal-on-metal bearing surfaces. Bone Joint Res 2013;2(5):84 [Vol. 16]. 14. 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 cobalt– chromium modular neck. J Bone Joint Surg Am 2013;95:865. 15. Panagiotidou A, Meswania J, Hua J, et al. Enhanced wear and corrosion in modular tapers in total hip replacement is associated with contact area and surface topography. J Orthop Res 2013. 16. Gill IPS, Webb J, Sloan K, et al. Corrosion at the neck–stem junction as a cause of metal ion release and pseudotumour formation. J Bone Joint Surg 2012;94-B(7). 17. Bolton JD, Hayden J, Humphreys M. A study of corrosion fatigue in cast cobalt– chrome–molybdenum alloys. Eng Med 1982;11(2):58. 18. Collier JP, Surprenant VA, Jensen RE, et al. Corrosion between the components of modular femoral hip prosthesis. J Bone Joint Surg (Br) 1992;74(4):511.
Contents lists available at ScienceDirect
The Journal of Arthroplasty journal homepage: www.arthroplastyjournal.org
Correlation of Corrosion and Biomechanics in the Retrieval of a Single Modular Neck Total Hip Arthroplasty Design Modular Neck Total Hip Arthroplasty System Brent A. Lanting, MD, FRCSC a, 1, Matthew G. Teeter, PhD a, 1, Edward M. Vasarhelyi, MD, MSc, FRCSC a, Todor G. Ivanov, PhD b, James L. Howard, MD, MSc, FRCSC a, Douglas D.R. Naudie, MD, FRCSC a a b
Division of Orthopaedic Surgery, London Health Sciences Centre, London, Ontario, Canada Imaging Research Laboratories, Robarts Research Institute, London, Ontario, Canada
a r t i c l e
i n f o
Article history: Received 14 February 2014 Accepted 13 June 2014 Keywords: metallosis modular total hip modular neck pseudotumor
a b s t r a c t Increased modularity of total hip arthroplasty components has occurred, with theoretical advantages and disadvantages. Recent literature indicates the potential for elevated revision rates of modular neck systems and the potential for local pseudotumor and metallosis formation at the modular neck/stem site. Retrieval analysis of one modular neck implant design including SEM (SCANNING ELECTRON MICROSCOPY) assessment was done and correlated with FEA (finite element analysis) as well as clinical features of patient demographics, implant and laboratory analysis. Correlation of the consistent corrosion locations to FEA indicates that the material and design features of this system may result in a biomechanical reason for failure. The stem aspect of the modular neck/stem junction may be at particular risk. © 2014 Elsevier Inc. All rights reserved.
Total hip arthroplasty (THA) is a common procedure, with excellent longevity and patient satisfaction. However, advances continue to be proposed to improve implant longevity, patient satisfaction, and surgical technique. Implant innovations have led to increasing modularity from monoblock femoral components to hip systems with modular heads. In addition to modularity of the prosthetic head, several manufacturers have proposed modularity of the femoral neck. Modularity at this level was proposed to allow changes in version, length, offset and neck – shaft angle. This was theorized to be an attractive design innovation as it potentially allowed for better recreation of the patients' anatomy. Increased modularity also had surgical technique advantages, such as allowing increased opportunities for soft tissue balancing and leg length optimization, thereby potentially allowing for reduced surgical time. However, modular necks have been cause for some concern clinically, and there are reports of corrosion, metallosis [1] and modular neck fracture [2]. There have also been reports of higher revision rates, with pseudotumor and catastrophic mechanical failure being implicated in registry data [3]. This has led to the market withdrawal or recall of some implants. However, other products continue to be available. The purpose of this retrieval study was to examine modes of failure of a single implant design with a modular
The Conflict of Interest statement associated with this article can be found at http:// dx.doi.org/10.1016/j.arth.2014.06.009. Reprint requests: Brent A. Lanting, MD, FRCSC, Division of Orthopaedic Surgery, London Health Sciences Centre, London, Ontario, Canada. 1 Denotes equal contribution. http://dx.doi.org/10.1016/j.arth.2014.06.009 0883-5403/© 2014 Elsevier Inc. All rights reserved.
femoral neck. Visual classification of damage, scanning electron microscopy (SEM) assessment to assess for corrosion and metal transfer, and biomechanical assessment of the prosthesis using finite element (FE) modeling were performed. Methods Study Population All modular neck implants in our institutional implant retrieval lab were retrospectively reviewed. Approval for review of patient charts and implant retrieval analysis was obtained from the internal review board. A total of nineteen implants were identified to be of the same design of a dual taper Ti–Al–V (TMZF) femoral component, and a Co– Cr–Mo (Vitallium) modular femoral neck (Rejuvenate, Stryker, Mahwah, New Jersey) and examined after retrieval. All retrieved hips had Co–Cr heads (forged Vitallium) articulating on a highly crosslinked polyethylene liner. Details with regard to patient characteristics (Table 1), implant (Table 2), and laboratory analysis (Table 3) were collected. These implants were retrieved from fourteen females and five males, with a mean age of 65 years (range, 41 to 81 years) and a mean body mass index of 33.5 (range, 26.7 to 49.1 years). The diagnosis leading to the THA was avascular necrosis in two patients and osteoarthritis in the remainder of the patients. Ten of the twelve magnetic resonance imaging scans conducted prior to revision revealed cystic pseudotumors, with an average size of 8.2 × 5 × 2.7 cm (range, 12.7–1.8 cm). The implants were in situ for an average of 1.7 years (range, 0.8 to 3.1 years), and three of the retrieved implants were
136
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Table 1 Patient Demographics and Results of Cross Sectional Imaging of All Revised Implants. Case
Age
BMI
Time In Situ
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 Average
81.0 57.6 60.1 64.5 66.7 78.3 81.1 75.2 60.8 68.9 60.3 65.9 41.4 60.5 41.1 71.3 72.6 66.6 60.8 65.0
32.4 33.8 33.9 26.0 33.7 30.8 34.0 26.1 52.0 28.2 43.1 26.7 32.1 49.1 31.1 30.7 38.1 34.9 31.4 34.1
1.3 1.8 2.2 1.2 2.4 3.1 1.9 2.0 2.1 1.8 2.0 1.5 1.6 1.1 1.0 1.3 1.1 0.8 3.0 1.7
Symptoms
Reason
MRI
Pain Pain Weakness Pain/Neurology Weakness Pain Weakness Pain Weakness Pain/Weakness Weakness Pain/Weakness Pain/Neurology Drainage Pain Infection Pain Pain Pain
metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis metallosis Pain Infection Infection Infection Subsidence Subsidence Subsidence Infection: 3 metallosis: 12 Subsidence: 2 Pain: 1
ALVAL N/A ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL ALVAL N/A Neg N/A Neg N/A N/A N/A N/A
revised for infection. The average ESR was 25, CRP 12, chromium level of 0.7, cobalt 5.5, and titanium 3.1, with an average ratio of chromium to cobalt of 0.2:1. Visual Damage Scoring Each modular neck was examined visually using the method of Goldberg et al [4] for signs of corrosion and fretting by two of the authors (M.G.T. and B.A.L.). The necks were divided into four zones, corresponding to the superior, anterior, inferior, and posterior sides of the neck (Fig. 1). The marking arrow on the face of the trunion was taken as pointing to the superior zone, and all other zones followed in a consistent clockwise pattern. For each zone, corrosion and fretting were graded as none (grade 1), mild (grade 2), moderate (grade 3), or severe (grade 4) under stereomicroscope visualization (SZ-CTV, Olympus, Tokyo, Japan). Where differences in grading occurred, consensus was achieved after further discussion and evaluation. Table 2 Implant Specifications for All Retrieved Implants.
Case ID
Acetabular Size Head Size (mm) (length) (mm) Stem Size
Neck Length (mm)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19
58 52 52 54 52 56 50 56 54 50 5 50 58 50 62 48 52 56a 58
30 30 34 30 30 34 30 34 30 34 30 30 34 30 34 34 34 34 34 Length 34: 10 Length 30: 10
a
36 36 36 (−5) 36 36 (−5) 32 32 (−4) 36 36 (−5) 32 32 (+4) 32 (+0) 40 32 36 32 (−4) 36 (+5) 36 (+5) 40 (+4) Size 32: 7 Size 36: 10 Size 40: 1
Porous 273 acetabular component.
9 7 8 7 9 7 7 9 9 7 8 7 8 7 8 7 7 7 8 Size 7: 10 Size 8: 5 Size 9: 4
Version (°) 0 0 8 16 0 8 0 0 8 8 0 0 0 8 0 0 0 8 0 0°: 12 8°: 6 16°: 1
Table 3 Laboratory Analysis of All Patients Prior to Revision.
Case ID
ESR (mm/h)
CRP (mg/L)
Chromium (μg/L)
Cobalt (μg/L)
Titanium (μg/L)
Ratio (Chr/Co)
1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 AVERAGE
34.0 38.0 25.0 90.0 9.0 16.0 32.0 6.0 14.0 1.0 20.0 7.0 5.0 38.0 27.0 44.0 16.0 28.0 7.0 24.1
13.3 5.7 7.2 28.2 2.9 1.7 8.1 6.4 10.1 9.0 6.0 1.2 1.2 54.4 16.0 35.8 4.3 4.9 2.6 11.5
0.4 0.2 1.0 N/A 1.9 0.8 0.9 1.3 1.0 0.4 1.2 N/A 0.4 N/A 0.9 0.6 0.3 N/A 0.4 0.8
3.2 2.8 2.7 N/A 5.7 15.7 9.5 9.9 6.3 9.8 4.6 N/A 2.4 N/A 4.2 3.6 0.5 N/A 2.4 5.5
1.6 2.3 5.0 N/A 3.8 2.3 3.2 4.3 5.8 1.9 2.2 N/A 2.5 N/A 1.7 3.5 1.9 N/A 3.5 3.0
0.1 0.1 0.4 N/A 0.3 0.1 0.1 0.1 0.2 0.0 0.3 N/A 0.2 N/A 0.2 0.2 0.6 N/A 0.2 0.2
Scanning Electron Microscopy The first six consecutively retrieved implants, along with two never-implanted reference specimens, were further analyzed with scanning electron microscopy (LEO 440 SEM, Carl Zeiss SMT Inc., Peabody, Massachusetts) and energy dispersive x-ray (Quartz Xone EDX system, Quartz Imaging Corporation, Vancouver, British Columbia) analysis (SEM/EDX). All six modular necks were examined using SEM/EDX for evidence of corrosion, fretting, and material transfer. In addition, one stem was sectioned in two planes at the neck–stem interface, and these sections also underwent SEM/EDX analysis. Finite Element Analysis The never-implanted reference specimens (size 7 and size 12 stems, 0 and 8 degree 30 mm necks, and 32 mm head) were laser scanned to generate 3D models for finite element analysis (FEA). The models were appropriately meshed and assembled in the FEA software (Abaqus, Dassault Systemes, Waltham, Massachusetts). The material properties applied to the components were a Poisson's ratio of 0.32 for the TMZF stems, and a Poisson's ratio of 0.30 for the Vitallium® necks and head [5–7]. The elastic moduli for TMZF and Vitallium vary in the literature, therefore we ran multiple models across these different moduli. These were 79 500, 100 000, and 110 000 MPa for the TMZF stem, and 200 000 and 240 000 MPa for the Vitallium neck and head [5,6,8]. The outer surface of the stem was fixed, and the coefficient of friction between the head and neck, and neck and stem was set as 1.00 to model zero motion between them. Simulating one body weight of an 80 kg person crossing the joint, an 800 N load was applied to the femoral head as a concentrated force, as has been done in other studies [7,9,10]. Von mises stress was measured from the simulations, virtual cross-sections between the segments were taken, and the regions of highest stress were probed for maximum values. Statistics Descriptive statistics (mean ± SD) were calculated for the corrosion and fretting damage scores. A D'Agostino and Pearson omnibus test for normality was used. As the scores were not normally distributed, Wilcoxon matched-pairs signed rank tests were used to compare the damage between zones.
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
137
Fig. 1. Three-dimensional model demonstrating the modular stem–neck junction (A), photograph of a modular neck with corrosion (B), and photograph of a cross-section through a stem with damage (C). Black arrows are the superior and inferior sides of the modular neck. White arrows are the damage scar within the neck–stem junction, along the edge that opposes the superior surface of the modular neck.
Results Visual Damage Scoring All of the retrieved stems had evidence of corrosion at the neck–stem interface (Fig. 1). There was no difference between the superior and inferior zones (P = 0.59) or between the anterior and posterior zones (P = 0.42) of the modular necks, therefore the superior and inferior zones were combined, as were the anterior and posterior zones. The corrosion was more severe (P = 0.01) on the superior/inferior zones (3.4 ± 0.8) than the anterior/posterior zones (3.0 ± 0.8). Fretting appeared mild or absent on the modular necks. There was no difference (P = 0.82) in the severity of the fretting between the superior/inferior zone (1.2 ± 0.5) and the anterior/posterior zone (1.2 ± 0.4). Scanning Electron Microscopy The SEM/EDX revealed the black corroded regions on the modular necks to contain chromium phosphate, a corrosion byproduct (Fig. 2). In some cases, transfer of titanium from the TMZF stem onto the modular necks was also seen (Fig. 2). The section taken from the stem revealed a marked shiny streak of metallic damage along the superior surface of the neck–stem interface (Fig. 3). Under SEM/EDX analysis, the streak contained higher than reference levels of iron, along with titanium and chromium phosphate (Fig. 3). Adjacent to this streak was a black ring of corrosion (running perpendicular to the metallic streak) that contained chromium phosphate (Fig. 3). Finite Element Analysis The pattern of stresses within FEA results was similar between the neutral and anteverted or retroverted necks, and between the two sizes of stems (Fig. 4). The maximum stress (Table 4) was consistently located on the superior and inferior sides of the neck, specifically at the superior-lateral and inferior-medial corners of the neck–stem interface. For the model of the small stem with straight neck, the mean maximum stress was 372.9 ± 14.0 MPa within the neck and 970.8 ± 60.4 MPa within the stem. Modeling an angled neck reduced the mean maximum stress in the neck to 167.9 ± 16.2 MPa, but increased the mean maximum stress within the stem to 1611.0 ±
118.1 MPa. For the model of the large stem with the straight neck, the mean maximum stress was 220.0 ± 2.4 MPa within the neck and 1611.7 ± 159.0 MPa within the stem.
Discussion Modularity at the neck–stem junction was initially designed and marketed to have patient and surgeon advantages. The goal was recreation of patient anatomy, and optimization of the THA position and balance with greater surgical ease. However, this additional level of modularity has been shown to have higher revision rates in some implants. The potential reasons for failure have not been well described in a clinical context. In this study, the retrieved implants were all of the same design, and were revised relatively early. Of all the retrieved implants, three were revised due to an infection. The relatively low chromium levels measured in these patients that required revision is notable. Literature has indicated that ion levels may be a way of monitoring metal on metal THA. In spite of the corrosion and metal transfer seen, the ion levels seen in these cases of modular neck THA that required revision are relatively low, with a maximum chromium level seen of 1.7 μg/L, cobalt 15.6 μg/L and titanium 5.8 μg/L. However, metallosis and significant soft tissue destruction were seen in twelve cases. Corrosion in the crevasse of the neck/stem junction may allow the low pH conditions favorable to the formation of hexavalent chromium [11]. Hexavalent chromium is known to be more toxic than trivalent chromium [11,12]. Although advanced techniques may allow for measurement of hexavalent chrome, these values were not available for this study [13]. The relatively lower values of chromium than cobalt seen are also notable, and have been seen in other studies [14]. The patient group described here had an average ratio of chromium to cobalt of 0.2:1. All retrieved implants that underwent SEM were consistent in demonstrating chromium phosphate, indicating corrosion as well as titanium transfer at the upper and lower edges of the neck. There were minimal changes on the sides of the modular neck. Stem analysis revealed more extensive changes (Fig. 3). Elevated iron, titanium and chromium phosphate demonstrate corrosion and metal transfer. The metal transfer suggests micro-motion at the
138
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Fig. 2. SEM/EDX results from a modular neck. The photograph of tested stem (A) is highlighted where the SEM images in B (yellow box) and C (red box) were obtained. The EDX spectrum from the first (yellow) region is shown in D, and the second (red) region is shown in E. Notable peaks of phosphorus (F) from chromium phosphate corrosion byproduct were found in the first region, and titanium transfer (G) from the stem to the modular neck was found in the second region.
neck–stem interface. These findings are in keeping with other reports in the literature [14,15]. Correlation of the FE model and the SEM analysis is of particular importance. The most severe corrosion was found along the superior and inferior surfaces of the modular necks, corresponding to the regions of greatest stress within the FE model. This is similar to FE modeling of other systems [16]. The concept of corrosion fatigue is not a new concept, but has been previously described with a resultant possibility of fracture [17]. It is not understood if the forces seen via the FE modeling have a role in the corrosion seen due to corrosion fatigue. Severe damage was seen within the neck–stem junction of the stem, along the superior surface, again corresponding to the greatest corrosion on the modular necks and the greatest stresses within the FE model. However, it is notable that the FE
model revealed stresses to be greater in the stem than in the neck. The elastic limits of Vitallium are approximately 500 to 850 MPa, whereas the FE models revealed the greatest stresses to be 372.9 MPa within the modular neck [5,6,8]. The elastic limit of TMZF is higher, at 1000 to 1060 MPa [5,6,8]. However stresses within the stem were also higher in the FE model, and at up to 1611.7 MPa, exceeding this elastic limit. This may explain the damage seen within the stem–neck junction of the stem, as well as the titanium transfer from the stem to the modular necks. The role of galvanic corrosion due to dissimilar metals is of interest, but the relative contribution of the dissimilar metals to the corrosion seen is not clear [17,18]. Notably, the FE model used in this study represents one specific loading scenario that has been used in other studies [7,9,10]. This loading scenario was applied to different
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
139
Fig. 3. SEM (A) and EDX (B) results from a cross-section of a stem (shown in Fig. 1C). The region had normal amounts of titanium (C), but high amounts of iron (D) as well as chromium (E) transferred from the modular neck.
geometries of neck and stem geometries. Different stresses may occur if different applied loads were modeled. The finding in this study of the stem aspect of the modular neck/ stem interface being potentially more impacted by the design is important. In cases of revision where the modular neck looks relatively undamaged, the potential for compromise of the stem
aspect of the junction must be considered, and warrants revision of this component as well. The strengths of this study are that it is a comprehensive study on a single implant design and examines the potential sources of failure of this implant. The use of advanced technologies allowed precise measurement of the retrieved implants included laser mapping and
Fig. 4. Cut-through of the FE model for the small stem with straight modular neck, using a color map to represent von Mises stress. Areas of peak stress are denoted with arrows, along the superior and inferior sides of the modular neck–stem junction.
140
B.A. Lanting et al. / The Journal of Arthroplasty 30 (2015) 135–140
Table 4 Maximum Von Mises Stress (MPa) in the Stem and Neck Across the Different Models and Combination of Properties. Maximum Stress (MPa) Straight Neck, Small Stem
Angled Neck, Small Stem
Straight Neck, Large Stem
Assigned Head and Neck/Stem Elastic Moduli (MPa)
Stem
Neck
Stem
Neck
Stem
Neck
200 000/79 500 200 000/100 000 200 000/110 000 240 000/79 500 240 000/100 000 240 000/110 000 Mean Standard deviation
898.1 989.4 1029.0 894.6 986.8 1027.0 970.8 60.4
352.5 370.4 377.7 362.8 382.7 391.2 372.9 14.0
1692.0 1535.0 1459.0 1791.0 1621.0 1568.0 1611.0 118.1
154.2 182.7 185.2 150.9 154.7 179.8 167.9 16.2
1831.0 1699.0 1648.0 1634.0 1462.0 1396.0 1611.7 159.0
217.8 217.7 218.0 221.8 222.3 222.6 220.0 2.4
SEM analysis, and conservative loads of 800 N were applied in FE modeling. However, assumptions regarding of model geometries and fit between neck/stem may limit the accuracy of the absolute values reported for the stress. Conclusions In conclusion, SEM analysis of this particular design of a modular neck total hip system revealed corrosion at this additional junction of modularity. FE analysis indicates that the material and design features of this system may indicate a biomechanical reason for failure; particularly at the stem aspect of the modular neck/stem junction as the stem's material yield strength of loading may be exceeded under physiologic loads. Acknowledgments The authors thank Ross Davidson at Surface Science Western.
References 1. Gill IP, Webb J, Sloan K, et al. Corrosion at the neck–stem junction as a cause of metal ion release and pseudotumor formation. J Bone Joint Surg Br 2012;94(7):895. 2. Murugappan KS, Graves S. Outcome after modular neck hip arthroplasty: lessons from the registry. J Bone Joint Surg Br 2012;94-B(sXLI):112. 3. Atwood SA, Patten EW, Bozic KJ, et al. Corrosion-induced fracture of a doublemodular hip prosthesis: a case report. J Bone Joint Surg 2010;92(6):1522. 4. Goldberg JR, Gilbert JL, Jacobs JJ, et al. A multicenter retrieval study of the taper interfaces of modular hip prostheses. Clin Orthop Relat Res 2002;401:149. 5. Gracia L, Ibarz E, Puertolas S, et al. Study of bone remodelling of two models of femoral cementless stems by means of DEXA and finite elements. Biomed Eng Online 2010;9:22. 6. Gore D, Frazer RQ, Kovarik RE, et al. Vitallium. J Long Term Eff Med Implants 2005;15(6):673. 7. Theodorou EG, Provatidis CG, Megas PD. Investigating a total hip finite element model with respect to the offset and version changes, according to the modular neck type. J Serbian Soc Comput Mech 2009;3(2):1. 8. Moratal D. Finite element analysis—from biomedical applications to industrial developments. Croatia: InTech; 2012. 9. Krischak GD, Augat P, Beck A, et al. Biomechanical comparison of two side plate fixation techniques in an unstable intertrochanteric osteotomy model: sliding hip screw and percutaneous compression plate. Clin Biomech 2007;22(10):1112. 10. Nunn D, Freeman MA, Tanner KE, et al. Torsional stability of the femoral component of hip arthroplasty. Response to an anteriorly applied load. J Bone Joint Surg 1989;71(3):452. 11. Kotas J, Stasicka Z. Chromium occurrence in the environment and methods of its speciation. Environ Pollut 2000;107(3):263. 12. Levis AG, Majone F. Cytotoxic and clastogenic effects of soluble and insoluble compounds containing hexavalent and trivalent chromium. Br J Cancer 1981;44(2):219. 13. Sidaginamale RP, Joyce TJ, Lord JK, et al. Blood metal ion testing is an effective screening tool to identify poorly performing metal-on-metal bearing surfaces. Bone Joint Res 2013;2(5):84 [Vol. 16]. 14. 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 cobalt– chromium modular neck. J Bone Joint Surg Am 2013;95:865. 15. Panagiotidou A, Meswania J, Hua J, et al. Enhanced wear and corrosion in modular tapers in total hip replacement is associated with contact area and surface topography. J Orthop Res 2013. 16. Gill IPS, Webb J, Sloan K, et al. Corrosion at the neck–stem junction as a cause of metal ion release and pseudotumour formation. J Bone Joint Surg 2012;94-B(7). 17. Bolton JD, Hayden J, Humphreys M. A study of corrosion fatigue in cast cobalt– chrome–molybdenum alloys. Eng Med 1982;11(2):58. 18. Collier JP, Surprenant VA, Jensen RE, et al. Corrosion between the components of modular femoral hip prosthesis. J Bone Joint Surg (Br) 1992;74(4):511.
