This article has Open Peer Review reports available.
Modular titanium alloy neck adapter failures in hip replacement - failure mode analysis and influence of implant material
© Grupp et al; licensee BioMed Central Ltd. 2010
Received: 27 February 2009
Accepted: 4 January 2010
Published: 4 January 2010
Modular neck adapters for hip arthroplasty stems allow the surgeon to modify CCD angle, offset and femoral anteversion intraoperatively. Fretting or crevice corrosion may lead to failure of such a modular device due to high loads or surface contamination inside the modular coupling. Unfortunately we have experienced such a failure of implants and now report our clinical experience with the failures in order to advance orthopaedic material research and joint replacement surgery.
The failed neck adapters were implanted between August 2004 and November 2006 a total of about 5000 devices. After this period, the titanium neck adapters were replaced by adapters out of cobalt-chromium. Until the end of 2008 in total 1.4% (n = 68) of the implanted titanium alloy neck adapters failed with an average time of 2.0 years (0.7 to 4.0 years) postoperatively. All, but one, patients were male, their average age being 57.4 years (36 to 75 years) and the average weight 102.3 kg (75 to 130 kg). The failures of neck adapters were divided into 66% with small CCD of 130° and 60% with head lengths of L or larger. Assuming an average time to failure of 2.8 years, the cumulative failure rate was calculated with 2.4%.
A series of adapter failures of titanium alloy modular neck adapters in combination with a titanium alloy modular short hip stem was investigated. For patients having received this particular implant combination risk factors were identified which were associated with the occurence of implant failure. A Kaplan-Meier survival-failure-analysis was conducted. The retrieved implants were analysed using microscopic and chemical methods. Modes of failure were simulated in biomechanical tests. Comparative tests included modular neck adapters made of titanium alloy and cobalt chrome alloy material.
Retrieval examinations and biomechanical simulation revealed that primary micromotions initiated fretting within the modular tapered neck connection. A continuous abrasion and repassivation process with a subsequent cold welding at the titanium alloy modular interface. Surface layers of 10 - 30 μm titanium oxide were observed. Surface cracks caused by fretting or fretting corrosion finally lead to fatigue fracture of the titanium alloy modular neck adapters. Neck adapters made of cobalt chrome alloy show significantly reduced micromotions especially in case of contaminated cone connection. With a cobalt-chromium neck the micromotions can be reduced by a factor of 3 compared to the titanium neck. The incidence of fretting corrosion was also substantially lower with the cobalt-chromium neck configuration.
Failure of modular titanium alloy neck adapters can be initiated by surface micromotions due to surface contamination or highly loaded implant components. In the present study, the patients at risk were men with an average weight over 100 kg. Modular cobalt chrome neck adapters provide higher safety compared to titanium alloy material.
Total Hip Arthroplasty (THA) has become a successful clinical treatment to restore the function of the joint, with a positive impact on the patient's quality of life. Modular connections for hip prostheses have been used since the early 70ies for heads with different neck sizes or diameters. Later in the 90ies, modular neck adapters have been introduced [1, 2] for intraoperative adjustment of collum-caput-diaphysis (CCD) angle and femoral anteversion to optimise offset and leg length, irrespective of the hip stem implanted. These solutions have been proven their relevance in total hip arthroplasty [3, 4]. Several failures of modular connections in hip replacement including primary and revision stem components [5–8] were reported. Long-term experiences with modular heads made of cobalt-chromium alloy (CoCr29Mo6) in combination with the cone of the stem out of titanium alloy (TiAl6V4) have revealed, apart from traces of fretting corrosion, no adverse events such as fractures of the tapers in clinical use [9–17]. Failures of modular neck adapters have been rarely documented [5–7]. In 2007, we reported three cases of a failure of a modular short hip stem . The purpose of this paper is to present the state-of-the-art research and the recent findings of the failure analysis.
Overview about the average age, weight and time in vivo until failure of the titanium alloy neck adapter in 68 patients and distribution of neck adapter and combined head geometry
Ø 57.2 years (36 - 75 years)
Ø 102.3 kg (75 to 130 kg)
time in vivo
Ø 24 months (8 to 48 months)
130° ± 7.5°
135° ± 7.5°
140° ± 7.5°
There was no correlation between implant failures and a specific clinic or surgeon.
Implant component description
In the retrieval analysis, the failed neck adapters were subjected to a detailed examination to which most patients consented. The analysis is in every case based on the failed neck adapter and the hip stem including the remaining distal part of the neck adapter. The revised modular heads and cup inserts were in most of the cases not available and not subject of this examination. The present report is based on the retrieval analysis of 47 devices, the complete investigation data being compiled at the end of 2008. The fracture and modular connection surfaces were evaluated using light microscope and scanning electron microscope (SEM) (Zeiss EVO 50, Carl Zeiss NTS GmbH Oberkochen, Germany).
The fragment of the broken adapters which remained inside the shaft was cut through axially. One half of the adapters was cautiously removed from the shaft to examine the contact zone between the two components. The other half was left inside the shaft to enable the metallographic investigation of the interface and the microstructure of the materials. To enable the metallographic investigation of the neck adapter/stem interface the samples were prepared by mechanical cutting using a cut off-wheel (Secotom 10 Struers A/S, Denmark). After embedding into epoxy resin (Polyfast Struers A/S, Denmark) the cross section was prepared by grinding, polishing and etching according to the method by Kroll (composition 100 ml distilled H2O, 1 ml HF, 3 ml HNO3). The investigation was performed by light microscope (Wilde M3Z Herrenbrugg, Switzerland) with a magnification up to 500-times.
The chemical composition of the abrasive substances found in the crevice of the cone connection was determined using energy dispersive analysis (EDX) (Oxforth EDX-Analysensystem INCA 7059 Oberkochen, Germany) and wet chemical digestion. The surface of the cone was rinsed cautiously using dilute acid. After hot digestion and dilution the analysis of the elements was performed using inductively coupled plasma optical emission spectrometry (ICP-OES Horiba Ultima II Jobin Yvon Longjumeau, France).
Micromotion analysis of the modular neck stem interface
The stem was combined with neck adapters with 130° CCD angle, embedded in bone cement (Palacos R, Heraeus Medical GmbH Wehrheim, Germany) and tested on a servohydraulic testing machine (MTS 850.2, MTS Systems Corporation Eden Prairie MN, USA). A sinusoidal axial force between 50 and 2500 N was applied via a ceramic head with neck length L at a frequency of 1 Hz for 2000 cycles to measure the relative displacement between neck adapter and stem with regard to irreversible settling and micromotions. A statistical analysis was performed to distinguish between independent groups (clean and particle contaminated joining area) (paired Student's t test) for both neck adapter materials (SPSS 15.0).
Pre-clinical fatigue testing
To determine the endurance properties of the neck region comparative tests were performed according to ASTM F 2068-03 and ISO 7206-6:1992(E) (MTS Mini Bionix II, MTS Systems GmbH Berlin, Germany). The hip contact force was set at 5340 N with a sinusoidal loading mode (ratio Fmin/Fmax = 0.1). In a saline medium (0.9%, pH 7) the number of cycles was set at 10 million load alterations at a frequency of 15 Hz. In a worst case scenario using a 32 mm diameter XL head, neck fatigue was tested on the smallest modular stem in combination with neck components with 130° CCD angle and 7.5° retrotorsion out of titanium and cobalt-chromium alloy, respectively. A paired Student's t test was used to differentiate the fatigue behaviour of the two neck adapter materials (Statistica 7, StatSoft Europe GmbH, Hamburg, Germany).
For these customised test, the hip contact force was set at 2300 N for 5 million cycles according ISO 7206-8:1995(E) followed by a stepwise load increase (Locati method 500 N, 1 million cycles) until failure. These customised test setup was introduced to analyse if the mode of fatigue failure occurs in the modularity or in the stem region outside of the neck/stem connection.
In vitro model to simulate the clinical failure modes with various parameters
Hip contact force
1 Hz/15 Hz
Stress and rest phases
1,000 cycles, 1.5 minutes rest, 1,000 cycles, 1.5 minutes rest repeat
Dry, Blood, Serum, Cortical bone particles
Saline 0.9% pH 7 Saline 1.8% pH 2 CaCl2 solution pH 2 FeCl3 solution pH 0.5
The resultant hip contact force was set at 3800 N to simulate a higher strain as in overweight or active patients. The test frequency was decreased from 15 Hz to 1 Hz to detect a possible influence on fretting corrosion in the modular neck interface. To simulate the patient's situation in daily life more realistically frequent stress phases alternated with rest phases. Taking into account the clinical conditions of hip replacement assembly, the connection was contaminated with blood, serum and cortical bone particles (approximately 1 mm in diameter) to provoke a mechanical disturbance in the interface. Additionally, the surrounding chemical conditions were altered by modifying pH values and enhancing chloride concentrations with the use of NaCl, CaCl2 and FeCl3 to accelerate any eventual tribochemical processes. To determine the continuous abrasion and repassivation of the titanium alloy surfaces in the neck/stem connection the redox potential was measured (InLab® 301 Mettler Toledo Balingen, Germany).
Findings of the retrieval examination in clinical failures
Additionally in some cases this process accelerated through crevice corrosion as the reduction of the pH to values of about 2-3 indicates.
The surface of the cone adapter which was not cleaned after revision was examined using a scanning electron microscope. No calcium or phosphor could be detected. There was no indication of a contamination of the cone connection by bone particles. Besides mild abrasion on the surface, no signs of fretting corrosion could be found.
Influence of micromotions on the modular neck stem interface
In the case of the cobalt-chromium alloy neck adapter, the contamination of the joining area did not influence the micromotions significantly (p = 0.43).
In vitro fatigue behaviour of the stem/neck modularity
The neck adapters made of titanium and of cobalt-chromium alloy showed significantly different (p = 0.009) endurance behaviour. The Ti6Al4V adapters combined with a 28 L Biolox forte head failed after 2.45 million cycles (range 0.189 to 4.43). No failure occurred with the cobalt-chromium adapters (32 XL Biolox option) up to 20 million cycles.
This study confirmes that the failure of the adapter cannot be attributed to any material or processing deviation or incorrect dimensioning. A twisted assembling of the adapter can also be excluded.
Due to mechanical perturbation in the modular connection, micro-movements caused fretting on the surface. Microcracks developed in the fretting zone, ultimately leading to the dynamic fatigue fracture of the implant. In 87% of the cases, fretting was accompanied by corrosion [17, 20–25] as it initiates crevice corrosion [26, 27]. The combination of fretting and crevice corrosion destroys the passive layer permanently. The electrochemical reactions provoke a shift in the pH value into the acidic range. The corrosion process generates a fissured surface of the connecting cone and can also generate microcracks. The examinations revealed that corrosion does not trigger the implant failure but does accelerate the process . Cone adapters free from corrosive attack last the longest period up to failure (Figure 10).
In general, the titanium alloy provides excellent corrosion resistance and high biocompatibility because it quickly develops a thin passive layer with a thickness under 1 μm. Under inadequate conditions of cone assembling and in case of micro-movements within the modular connection, the protective passive layer is abraded initiating a continued abrasion and repassivation process which depletes the oxygen inside the crevice . The metallic surface in the crevice becomes anodic relative to the outer surface, thus triggering an electrolytic process. The presence of chloride ions which disturb the passivity, induces pitting corrosion in the crevice leading to a quick dissolution of the metal accompanied by an acidification of the electrolyte [12, 20, 21, 28].
Inside the layer, the elements calcium and phosphor could be detected and verified by different analytical methods. The crystalline composition of the calcium phosphate compound seems to indicate that the wear debris contained bone particles. Impurities which get inside the cone during intraoperative assembly may contribute to a mechanical perturbation which manifests itself as micro-movements which initiates fretting.
The study suggests the following hypothesis as to the cause of the damage. Fretting occurs when two surfaces in contact experience small amplitude oscillary relative motion; damage is induced on the fretting region. If the fretting fatigue strength of the material is exceeded, microcracks develope on the surface. In addition, tribochemically activated particles can discharge their content from the surface. These particles react with oxygen spontaneously, thus leading to fretting corrosion.
The severity of corrosion depends mainly from the frequency of the fretting action. For lower cyclic rates such as in a hip implant, the debris is usually oxidized if the environment is chemically active. The repeated removal of oxide films due to continued abrasion and repassivation produces thick oxide layer's. Such oxides are normally harder than the virgin titanium alloy leading to greater surface damage [29–31] These damages then accelerate the crack nucleation [32, 33].
Assessment of the cone adapter made of the different materials - (ooo = excellent oo = good o = moderate)
stiffness/modulus of elasticity
The surface damage of the titanium alloy adapters caused by the microcracks or by corrosive deterioration accelerates the propagation of cracks by the cyclic loads bringing about the dynamic fatigue failure of the adapters. Micro-movements cause fretting in the cone connection. They can be increased by contamination of the cone connection through tissue or other particles intraoperatively. To anticipate this process any contamination of the connection should be avoided and the components dried before assembling. For this purpose abrasion-resistant cleaning rods are supplied together with the implants.
The change of the material of the adapter from titanium alloy to a cobalt-based alloy (CoCr29Mo6) increases the safety of the cone connection significantly. The combination of the cobalt-based alloy and the titanium alloy of the shaft shows a considerably higher rigidity. The smaller micro-movements reduce abrasion. Furthermore, the highly stable passive layer of the cobalt-based alloy provides an improved resistance against fretting. Due to its structure, the cobalt alloy has a much lower notch sensitivity compared to the titanium alloy. This enhances fatigue strength.
Among patients treated with the titanium alloy neck adapter, a combination of different parameters was identified as risk factors of implant failure. The parameters are intraoperative particle contamination of the cone connection, excessive loading due to a patient weight above 100 kg or high activity level, and male gender. In addition, the risk for failure rises with CCD angles of the cone adapter of 135° and smaller.
The authors would like to thank Elisa Hoenig, M.Sc. and Michael M. Morlock, Ph.D. for the examination of micromotions in the neck adapter/stem interface, Thomas Hermle, M.Sc., MBA for the performance of the survival-failure analysis and Christoph Schilling, M.Sc. for the statistical analysis of the in vitro tests.
- Toni A, Paderni S, Sudanese A, Guerra E, Traina F, Antonietti B, Giunti A: Anatomic cementless total hip arthroplasty with ceramic bearings and modular necks: 3 to 5 years follow-up. Hip International. 2001, 11 (1): 1-17.Google Scholar
- Traina F, Baleani M, Viceconti M, Toni A: Modular Neck Primary Prosthesis: Experimental and Clinical Outcomes. Scientific Exhibit at the 71st AAOS Annual Meeting, San Francisco. 2004Google Scholar
- Toni A, Sudanese A, Paderni S, Guerra E, Bianchi G, Antonietti B, Giunti A: Artroprotesi d'anca non cementata con collo modulare - Cementless hip arthroplasty with a modular neck. Chir Organi Mov. 2001, LXXXVI: 73-85.Google Scholar
- Braun A, Lazovic D, Zigan R: Modular short stem prosthesis- Implant positioning and the impact of navigation. Orthopedics Suppl. 2007, 30 (Suppl 10): 144-147.Google Scholar
- Keggi KJ: My experience with proximal modular stems. Symposium "Cutting-edge developments on proximal modularity in THA" Annual AAHKS Meeting. 2008, 16-22.Google Scholar
- Keggi KJ, Kennon RE, Keggi JM: Hip reconstruction - primary total hip arthroplasty with modular stems. Current Orthopaedic Practice. 2008, 19 (2): 124-126. 10.1097/BCO.0b013e3282f53eca.View ArticleGoogle Scholar
- Cameron HU, Leslie CJ, McTighe T: Design considerations and results for a modular neck in cemented THA. Abstract, Australian Orthopaedic Association 67th Annual Scientific Meeting. 2007Google Scholar
- Hermle T, Zeller R, Grupp TM, Blömer W: Metha short stem hip prosthesis - Examination of the modular cone connection. 2007, Springer Medizin Verlag, Heidelberg, implant, 12-17. 1/07Google Scholar
- Bobyn JD, Tanzer M, Krygier JJ, Dujovne AR, Brooks CE: Concerns with modularity in total hip arthroplasty. Clin Orth Rel Res. 1994, 27-38. 298Google Scholar
- Brown SA, Flemming CA, Kawalec JS, Placko HE, Vassaux C, Merritt K, Payer JH, Kraay MJ: Fretting corrosion accelerates crevice corrosion of modular hip tapers. J Appl Biomater. 1995, 6 (1): 19-26. 10.1002/jab.770060104.View ArticlePubMedGoogle Scholar
- Lieberman JR, Rimnac CM, Garvin KL, Klein RW, Salvati EA: An analysis of the head-neck taper interface in retrieved hip prostheses. Clin Orth Rel Res. 1994, 162-167. 300Google Scholar
- Gilbert JL, Buckley CA, Jacobs JJ: In vivo corrosion of modular hip prosthesis components in mixed and similar metal combinations. The effect of crevice, stress, motion, and alloy coupling. Journal of Biomedical Materials Research. 1993, 27: 1533-1544.PubMedGoogle Scholar
- Goldberg JR, Gilbert JL, Jacobs JJ, Bauer TW, Paprosky W, Leurgans S: A multicenter retrieval study of the taper interfaces of modular hip prostheses. Clin Orth Rel Res. 2002, 149-161. 10.1097/00003086-200208000-00018. 401Google Scholar
- Windler M: Korrosionsverhalten von modularen Verbindungen bei Hüftprothesen. Dissertation Eidgenössisch TechnischeHochschule Zürich No. 15284. 2003Google Scholar
- Collier JP, Surprenant VA, Jensen RE, Mayor MB: Corrosion at the interface of cobalt-alloy heads on titanium-alloy stems. Clin Orth Rel Res. 1991, 305-312. 271Google Scholar
- McKellop H, Sarmiento A, Brien W, Park SH: Interface corrosion of a modular head total hip prosthesis. J Arthroplasty. 1992, 7 (3): 291-294. 10.1016/0883-5403(92)90051-Q.View ArticlePubMedGoogle Scholar
- Jacobs JJ, Gilbert JL, Urban RM: Current concepts review - Corrosion of metal orthopaedic implants. J Bone Joint Surg [Am]. 1998, 80-A: 268-282.Google Scholar
- Hoenig E, Morlock MM: "Bestimmung der Relativbewegung an der Konusverbindung der modularen Metha-Kurzschaftprothese". Study to determine the relative motion between the neck adapter and the stem Biomechanics Section - Hamburg University of Technology, Aesculap test report No. V1039. 2008Google Scholar
- Hoenig E, Morlock MM: Estimation of the relative motion in the taper lock interface of a modular hip prosthesis. Abstract Orthopedic Research Society, Las Vegas, Nevada. 2009Google Scholar
- Kaesche H: Die Korrosion der Metalle - Physikalisch-chemische Prinzipien und aktuelle Probleme. 1979, Springer-Verlag Berlin Heidelberg New York, 274-276.View ArticleGoogle Scholar
- Gilbert JL, Jacobs JJ: The mechanical and electrochemical processes associated with taper fretting crevice corrosion - a review. Modularity of orthopedic implants, ASTM STP 1301. Edited by: Parr JE, Mayor MB. 1997, J ASTM International, 45-59. full_text.Google Scholar
- Urban RM, Gilbert JL, Jacobs JJ: Corrosion of modular titanium alloy stems in cementless hip replacement. 2005, J ASTM International, 2 (10): 1-10.Google Scholar
- Rodrigues DC, Urban RM, Jacobs JJ, Gilbert JL: In vivo severe corrosion and hydrogen embrittlement of retrieved modular body titanium alloy hip-implants. J Biomed Mater Res Part B Appl Biomater. 2009, 88B (1): 206-219. 10.1002/jbm.b.31171.View ArticleGoogle Scholar
- Collier JP, Surprenant VA, Jensen RE, Mayor MB, Surprenant HP: Corrosion between the components of modular femoral hip prostheses. J Bone Joint Surg [Br]. 1992, 74-B: 511-517.Google Scholar
- Cook SD, Barrack RL, Baffes GC, Clemow AJT, Serekian P, Dong N, Kester MA: Wear and corrosion of modular interfaces in total hip replacements. Clin Orth Rel Res. 1994, 80-88. 298Google Scholar
- Goldberg JR, Buckley CA, Jacobs JJ, Gilbert JL: Corrosion testing of modular hip implants. Modularity of orthopedic implants, ASTM STP 1301. Edited by: Parr JE, Mayor MB. 1997, J ASTM International, 157-176. full_text.Google Scholar
- Goldberg JR, Gilbert JL: In vitro corrosion testing of modular hip tapers. J Biomed Mater Res Part B: Appl Biomater. 2003, 64B: 78-93. 10.1002/jbm.b.10526.View ArticleGoogle Scholar
- Duisabeau L, Combrade P, Forest B: Environmental effect on fretting of metallic materials for orthopaedic implants. Wear. 2004, 256: 805-816. 10.1016/S0043-1648(03)00522-2.View ArticleGoogle Scholar
- Hoeppner DW, Gates FL: Fretting Fatigue Consideration in Engineering Design, Wear. 1981, 70: 155-164.Google Scholar
- Viceconti M, Baleani M, Squarzoni S, Toni A: Fretting wear in modular neck hip prosthesis. Journal of Biomedical Materials Research. 1997, 35: 207-216. 10.1002/(SICI)1097-4636(199705)35:2<207::AID-JBM9>3.0.CO;2-R.View ArticlePubMedGoogle Scholar
- Viceconti M, Ruggeri O, Toni A, Giunti A: Design-related fretting wear in modular neck hip prosthesis. Journal of Biomedical Materials Research. 1996, 30: 181-186. 10.1002/(SICI)1097-4636(199602)30:2<181::AID-JBM7>3.0.CO;2-N.View ArticlePubMedGoogle Scholar
- Hoeppner DW, Chandrasekaran V: Fretting in orthopaedic implants: a review, Wear. 1994, 173: 189-197.Google Scholar
- Martinez SA, Sathish S, Mall S, Blodgett MP: Evolution of fretting fatigue damage and relaxation of residual stress in shot-peened Ti6Al4V. Metallurgical and Materials Trans A. 2005, 36A: 3385-3396. 10.1007/s11661-005-0012-8.View ArticleGoogle Scholar
- Forschungskuratorium Maschinenbau EV: Reibkorrosion - Vorhaben Nr. 6. Maschinenbau-Verlag GmbH, 6000 Frankfurt/M, Forschungsheft. 1976, 110-113. 56Google Scholar
- The pre-publication history for this paper can be accessed here:http://www.biomedcentral.com/1471-2474/11/3/prepub
This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.