Variations in the trunnion surface topography ... - Wiley Online Library

Variations in the Trunnion Surface Topography Between Different Commercially Available Hip Replacement Stems Selin Munir,1,2,3 William L. Walter,2 William Robert Walsh1 1 Prince of Wales Clinical School, The Surgical and Orthopaedic Research Laboratory, Sydney, New South Wales, Australia, 2The Specialist Orthopaedic Group, Sydney, New South Wales, Australia, 3The Graduate School of Biomedical Engineering, Sydney, New South Wales, Australia

Received 20 May 2014; accepted 26 August 2014 Published online 15 October 2014 in Wiley Online Library (wileyonlinelibrary.com). DOI 10.1002/jor.22741

ABSTRACT: Modular hip implants allows for the adjustment of leg length, offset, and the ability to remove the head for acetabular exposure during primary and revision surgery. The design of the Morse taper facilitates the intimate contact of the conical trunnion of the femoral stem (male component), with the conical bore of the femoral head (female component). Orthopaedic trunnion tapers are not standardized and vary in length, taper angle, and base dimension. Variations in the design and surface characteristics of the trunnion, will directly reflect on the interface at the taper junction and can influence the likelihood of subsequent wear, corrosion and longevity of the implant. The effect of surface topography of trunnions on commercially available hip stems has not yet been considered as a possible contributing factor in the corrosion observed at taper junctions. In this study we analyzed the surface topography and surface roughness of randomly selected commercially available femoral hip stem trunnions to obtain a greater insight into their surface characteristics. ß 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 33:98–105, 2015. Keywords: taper; corrosion; topography; trunnion; surface roughness

A taper connection within total hip replacements (THR) facilitates the intimate contact of the conical trunnion of the femoral stem (male component), with the conical bore of the femoral head (female component). The stress transferred from the trunnion imposes compressive stresses on the walls of the bore providing a self-locking mechanism, which is resistant to rotational forces.1,2 The interlocking mechanism is supported by dominance of the axial load in comparison to the radial load, which transmits a torque. The acute angle of the Morse taper helps the locking mechanism under axial loads and provides a secure fix at the connecting junction. The design consideration for tapers entails a thorough consideration of angle, length, diameter, surface texture (manufacturing finish), surface roughness, and length of contact. The taper junction within hip replacements is a potential site for corrosive attack, with retrieval studies showing corrosion damage.3–6 The severity of corrosion may vary depending on material choice, surrounding environment, and surface geometries. Surface analysis is critical for engineered components due to 90% of failures occur at the surface.7 A surface is machined in accordance to specific dimensions and tolerance limits, however, every specific manufacturing method produces a specific finish named the “fingerprint,” which is the surface topography of the finish component.7 The surface of any material viewed at the micro scale consists of peaks and valleys where deviations in the characteristics of the size and spacing of these features determine the variability between surface topography. There is diversity in techniques that can be employed to relate to surface characteristics of materials, with the common

Conflicts of interest: None. Correspondence to: Selin Munir (T: þ61-422803996; F: þ61 2 93822660; E-mail: [email protected]) # 2014 Orthopaedic Research Society. Published by Wiley Periodicals, Inc.

98

JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2015

two options revolving around contact and non-contact methods.8 The surface parameters, which are used to describe the topographical characteristics of a surface commonly, obtained from a stylus is represented in Table 1.9,10 These parameters are highly valuable in assessing the surface topography of a sample, by measuring the vertical characteristics of the surface deviations and are the 2D parameters.11 Optical profilometers have been used in industrial and scientific sectors to analyze the surface finish and performance of materials.12 The four categories which encapsulated all parameters are the amplitude, spatial, hybrid, and functional. A list of definitions provided with the measurement parameters are given in Table 2.8–10,13 The 2D arithmetic surface roughness provides the average roughness of a material across a single line of points (x,y) axis whilst in comparison the 3D arithmetic surface roughness considers the surface area of the region being analyzed (x,y) and corresponds these changes relative to the changes in vertical amplitude (z axis). The 3D parameters will determine the topographic changes within the specified region in comparison to the linear measurements (r parameters). The specific 3D amplitude parameters used in this study are highlighted in more detail below, which are used to analyze the topography of the trunnions. The definition of Sa is the arithmetic deviation of the surface evaluated over the complete 3D surface. The mathematical representation of Sa is shown in Equation (1), where Lx and Ly are the side lengths of the sampling area. Sa ¼

1 Lx Ly

Z 0

Ly

Z 0

Lx

jnðx; yÞjdx dy

ð1Þ

Ten-point height (Sz) is defined as the difference in the height between the average of the five highest peaks and the five deepest valleys within the sampling

TOPOGRAPHICAL VARIATIONS IN HIP TRUNNIONS

99

Table 1. Surface Characteristic Parameter for 2D Analysis Measurement Parameter

Definition

Ra

Average roughness for a given sample length

Rp Rq

Maximal profile peak height Root mean square roughness for given sample length

Rt Rz

Maximum height of the profile Average maximum height of the profile

Rsk

The measure of the third central profile amplitude within the assessment length[24] Determines the sharpness of the probability density of the profile

Rku

Equation Rl Ra ¼ 1l 0 jyðxÞjdx Rp ¼ maxjyj qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi RL Rq ¼ L1 0 z2 ðxÞdx Rt ¼ Rp Rv n n P P pi vi Rz ¼ 1z i¼1 i¼1 R1 Rsk ¼ R13 1 y3 pðyÞdy q

Rku ¼

R 1 1 R4q 1

y4 pðyÞdy

Table 2. Surface Characterization Parameters for a 3D Optical Profilometer Derived From 2D Parameters8,13 Measurement Parameter Definition Amplitude Sq

Ssk

Sku

Sz

Spatial Sds Str Sal Std Hybrid Sdq Ssc Sdr Functional Sbi Sci Svi

Root mean square deviation of the height distribution Degree of the asymmetry of the surface, relative to the height distribution Degree of the peakness of the surface, relative to the height distribution Average of the ten highest and lowest points along the scan

Equation sffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi N P M P 1 Sq MN n2 ðxi yi Þ j¼1 i¼1

Ssk ¼

R Ra 1 a S3q a a Ra Ra

Sku ¼ S14

a a

q

Sz ¼ 15

5 P

i¼1

n3 ðx; yÞpðnÞdxdy

n4 ðx; yÞpðnÞdxdy

npi þ

5 P i¼1

nvi

The density of the summits The texture aspect ratio The fastest decay of the autocorrelation length The direction of the texture of the surface

Ns Sds ¼ ðM1ÞðN1ÞDxDy

The root mean square of the surface slope Mean summit curvature

Sdq ¼

The developed surface area ratio The surface bearing index The core fluid retention index The valley fluid retention index

min Str ¼ rrmax

Sal ¼ min

pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi tx2 þ ty2

qffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi ffi R Ly R Lx 2 1 u ðx; yÞdxdy 0 Lx Ly 0 i n h 2 P d nðx;yÞ dn2 ðx;yÞ 1 Ssc ¼ 2n þ 2 2 dx dy k¼1 1 pffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi PN1 MP 2 2 j¼1

Sdr ¼

1þðnðxiþ1 ;yj Þnðxi ;yj Þ=ðDxÞÞ þððxi ;yjþi ÞÞnðxi ;yj Þ=ðDyÞÞ ððM1ÞðN1ÞÞ

i¼1

ðM1ÞðN1Þ

100

S

q Sbi ¼ n0:05

V v ðn0:05 ÞV v ðn0:05 Þ Sci ¼ S1q ðM1ÞðN1ÞDxDy V v ðn0:8 Þ Svi ¼ S1q ðM1ÞðN1ÞDxDy


100

MUNIR ET AL.

area. The mathematical implementation of Sz is defined in Equation (2) where npi and nvi is the highest peak and deepest valley, respectively. 5 5 X 1 X npi þ nvi Sz ¼ 5 i¼1 i¼1

! ð2Þ

Skewness (Ssk) is defined as degree of asymmetry of the surface height about the mean plane. The mathematical representation of the skewness is presented in Equation (3), where RMS roughness parameter is Sq (Table 2) and p(n) is the probability of the density function of the residual surface. Ssk ¼

1 S3 q

Z

a

a

Z

a

a

n3 ðx; yÞpðnÞdx dy

ð3Þ

Kurtosis (Rku) is a coefficient, which determines the sharpness of the probability density of the profile mathematically represented in Equation (4). Therefore a “spikey” surface has a high kurtosis whilst a “bumpy” surface has a low kurtosis. Sku ¼

1 4

S q

Z

a

a

Z

a a

n4 ðx; yÞpðnÞdx dy

ð4Þ

Trunnions on commercial hip stems, in particular if designed by different manufactures vary in dimensions such as length and circumference, to our knowledge; there have been no previous attempts to accurately quantify the surface properties of trunnions on different commercial femoral stems. It has been proposed that even small variations between the tapers of the trunnion and bore in terms of taper trunnion angle mismatch may alter the magnitude of micro-motion occurring at the modular junction.14,15 We propose that there may be variations found on the surface of trunnion tapers on commercially available femoral stems. If this is the case, these differences may have an influence on the likelihood of subsequent wear, corrosion and longevity of the implant. The authors believe there have not been

previous studies, which describe the surface finish of trunnions within THRs in a quantitative manner. Therefore this study provides insight to potential research in this area.

MATERIALS AND METHODS We studied 11 clean un-used commercial stems from five different companies DePuy (Warsaw, IN), Stryker (Mahwah, NJ), Biomet (Warsaw, IN), Wright Medical (Arlington, TN), Smith and Nephew (London, UK). Most of the analyzed stems are listed within the top 10 primary used stems (HT3), and additionally listed in the most commonly used cementless stems (HT5), within the AOA National Joint Registry (2013). The material compositions of the femoral stems analyzed are titanium alloy compositions. All femoral stems use the alloy composition of Ti6Al4V with the exception of the ABGII and the Accolade (Stryker), which uses the alloy composition Ti12Mo6Zr2Fe. The ABGII, Accolade and Securfit Max stems have a V40 taper. The Tri-Lock, Silent, Summit, Corail, Synergy, Profemur, and Taperloc are 12/14 tapers, whilst the SROM is an 11/13 taper. All the femoral stems analyzed are manufactured to conjoin with both a ceramic and a metal femoral head. Each trunnion surface was scanned with an optical profilometer (Bruker ContourGT-I 3D Optical Microscope, Karlsruhe, Germany), an interference microscope with the capability to analyze 3D topographical features of materials. For all the analyzed trunnion surfaces three scans were taken, one at the proximal, one at the mid-point and one at the distal segments. All eleven analyzed surfaces were assessed using the same scan parameters and adjusted using Vision software at a magnification of 20, scan length of 1,000 mm, a back scan of 500 mm, and a threshold of 0.05%. An inbuilt stitching function within the software was used to scan multiple segments to be combined to provide the overall scan of 1,000 mm by 200 mm (width height) including a 10% overlay. Spatial and amplitude parameters were obtained via Ranalysis and S-analysis to determine the topographical characteristics of all trunnions. The parameters collected for the analyses were the 2D parameters (Ra, Rp, Rq, Rt, Rv, Rpm, Rz), the 3D S parameters (Sa, Sp, Sq, Sz, Sku, Ssk), and the area of the scanned region (lateral area, and surface are). Within the study the S-parameters are used to identify the surface characteristics in 3D (Fig. 1). Pitch and thread height was obtained by measuring the captured image on Vision 64

Figure 1. Representation of the surface parameters in relation to a surface profile. JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2015


software and using a trace tool that determined the change in the X–Y coordinate planes, which determined the heights of the threads for each trunnion analyzed. The coordinate points were correlated with standard vertical and horizontal lines drawn on the image.

RESULTS Graphical analyses from the optical profilometer of the analyzed trunnion surfaces confirm different trunnions are either threaded or non-threaded. Of the 11 scanned trunnions, four, namely the ABGII (Stryker), Accolade (Stryker), Taper-lock (Zimmer), and SROM (DePuy) presented a non-threaded surface characteristic, identified by a featureless pattern (Fig. 2). The remaining seven analyzed trunnions, shown in Figure 3 have a repetitive distinct threaded surface finish. A closer examination of Figure 3 revealed a periodic difference in the repetition, and heights of the threads for different hip implants. The Summit, Silent, and TriLock presented thread patterns, which show a variation between the thread heights used on the same trunnion surface. All the trunnions analyzed, which presented an altering thread pattern were DePuy products. The graphical analyses and the profiles of the scanned regions additionally identified differences in the thread angles used in the threaded trunnions. The thread height for the different trunnions was obtained from their profile by selecting a window of

101

multiple peaks and valleys to obtain an average thread height for each trunnion (Table 3). Comparing the dominant thread heights across all patterns, the results yielded a large variation with the exception of the Summit and Profemur, which both had an average dominant thread height of 12.90 mm. The Silent, the Summit, and the Tri-Lock revealed multiple thread heights within their profile with the former presenting two threads. The Summit showed the most unique thread pattern with five significantly different peak heights (Table 3). The dominant peak (12.91 mm) is positioned at the outer most section of the profile followed by the second thread at a height of 10.88 mm, with the thread heights decreasing as the profile is moved to the center where the thread height is approximately half the dominant value. The smaller threads showed a thread height of 5.908 mm, and 4.173 and 2.311 mm (Fig. 3e, Table 3). The pitch for all the trunnions with threads varied between 0.304 and 0.13 mm (Table 3). The silent had the widest pitch at 0.304 mm while the Profemur had the narrowest pitch at 0.13 mm resulting in the most frequent occurrence of the threads along the axial dimension of the trunnion (Fig. 3d and a, respectively and Table 3). The analyses of the thread heights and correlating pitch for each hip are shown in Table 3. The 3D surface roughness parameters Sa and Sz for all the analyzed trunnions are represented in Figure 4.

Figure 2. 3D surface topography images from the optical profilometer of the three non-threaded ABGII (a), the Accolade (b), the SROM (c), and the Taperloc (d) trunnions. JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2015

102

MUNIR ET AL.

Figure 3. 3D surface topography images from the optical profilometer of the seven threaded trunnions the Profemur (a), Secure-fit Max (b), Tri-Lock (c), Silent (d), Summit (e), Synergy (f), and the Corail (g).



103

Table 3. The Results of the Surface Geometry of the Eleven Commercial Trunnions Analyzed Hip Profemur Synergy Trilock Summit Silent Secur-fit Max Corail ABG II Taper-lock Accolade SROM

Sa

Sz

2.595 2.532 2.828 2.23 2.09 0.515 3.35 0.472 0.469 0.365 0.320

16.02 16.493 15.131 16.74 10.967 7.234 17.375 7.132 7.472 6.069 7.215

Thread height

Pitch

12.91 7.2368

0.13 0.22 0.15 3.77 0.30 2.38 2.51 0.20 N/A N/A N/A N/A

12.91 10.41 3.24 13.49 N/A N/A N/A N/A

10.88

5.91

4.17

2.31

7.91

5.80

1.97

The first column represents the Sa values, column two represents the Sz, third column is the thread height and the final column identifies the pitch for threaded trunnion.

The average surface roughness (Sa) is significantly smaller (p < 0.05) for non-threaded trunnions in comparison to the seven threaded trunnions. Similarly the Sz parameter is significantly smaller (p < 0.05) for the non-threaded trunnions in comparison to the threaded trunnions aside from the Secur-fit Max, which has a similar Sz parameter (Fig. 4). However, the Silent (threaded) and Secur-fit Max (threaded) stem both have a significantly smaller Sz parameter indicated in Figure 4 compared to the other threaded surfaces. The profiles of the Corail (threaded) and the SROM (non-threaded), which have a threaded and non-threaded surface finish, respectively, represent the largest difference of thread height identified between two trunnions (Fig. 5). This profile shows that the SROM does not have a distinct repeated thread height, where the asperities on the profile are related to the machine finishes of the trunnions. This finish could contribute to different contact length against the bore resulting in a difference of force distribution. The Corail profile shows the threaded surface of the trunnion would relate to periodic point contact along the engagement length between the trunnion and the bore. Figure 6 shows the x, z profile of x, y, z trunnions to demonstrate the alterations and non-

unique pattern of thread repetition found on different manufactured trunnion surfaces.

DISCUSSION The quantified analyses of the surface topography of 11 different commercial THR stems, revealed four with a smooth surface finish and seven with a distinct threaded pattern. A threaded surface is designed to accommodate multiple bearing surfaces, mainly to allow the conjunction of metal stem on a ceramic femoral head. The impaction of a ceramic head flattens

Figure 5. Profile of the Corail (green) and the SROM (blue) trunnions.

Figure 4. A graphical representation of the Sa and Sz parameters for all the analyzed trunnion surfaces.

Figure 6. Profile of three different threaded trunnions; namely the Synergy (blue), Silent (green), and Summit (purple). JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2015

104

MUNIR ET AL.

the threads creating a customized tight fixed taper junction. The periodic contact points of the threads on the trunnion against the bore allow for the frictional forces to be dispersed evenly within the taper junction. The variation in the surface topography of the trunnions may compromise the interlock strength of the taper due to the difference in the resistance to the distractive and rotator forces. A standard threaded surface incorporates multiple diameters, which allow for the measure of the thread heights, pitch, and valley depths. The outer diameter is the perpendicular distance between two polar opposite peaks, and the inner diameter is specified as the perpendicular distance between two polar opposite troughs. The pitch of the threads is obtained from the distance between two adjacent peaks/troughs. The unique threaded patterns and geometries observed on all the analyzed trunnions are believed to create different stress distribution patterns along the axial length of the taper. All the analyzed trunnion surfaces resulted in a significantly smaller Sa value in relation to the Sz.1 This indicates the trunnion surfaces consist of repetitive peak heights and valley depths spanning across the scanned surface area. The Sa parameter is obtained by taking an arbitrary line as a reference origin, through the surface, and summing all the displacements of the peaks and valleys which deviate from this line, then dividing this by the number of sample points. The Sz parameter is obtained by taking the vertical displacements from the aforementioned origin line, of the five highest peaks and five deepest valleys, and then averaging them. This provides a measure of the maximal peak–peak displacement of the surface. Therefore the smaller Sa in comparison to Sz indicates the valleys and peaks are repeating. Non-threaded surfaces all showed significantly smaller Sa in comparison to the threaded surfaces indicating the overall surface of the trunnion is less rough. The similarity in the average roughness of some non-threaded and threaded stems such as the ABGII (non-threaded), Profemur (threaded), and SROM (nonthreaded) has shown that the machine finishing on the surfaces of the trunnions causes similar surface topographies, which result in similar average roughness values (Fig. 4). The Sz analyzes of the non-threaded surfaces were significantly higher than the average roughness revealed that the asperities on the surface of the non-threaded stems were significant. Although this was not defined as an evenly threaded pattern, the significant difference in the Sa and Sz contribute to the large asperities for non-threaded finishes on the surfaces. The difference of the peaked asperities in the nonthreaded surface may be eliminated during the insertion of the femoral head during primary and revision surgery. The higher Sa parameter for the threaded trunnions indicate a rougher overall surface due to the reoccurring

threads, accumulating more displacement from the origin line and resulting in the higher Sa value (Fig. 4). All the threaded trunnions, present different pitch and thread heights, however, the Sz analysis of the material surface has shown no significant difference between the Profemur, Synergy, Summit, or Corail (all threaded). The Silent has the largest thread angle (Fig. 3d). This result in wide peaks and long pitch has shown to be significantly different from all the other trunnion Sz parameters. The shorter thread height correlates directly with smaller Sz values. The Secur-fit Max presented a threaded surface, however, the average roughness is similar to the trunnion of the SROM (non-threaded). The similarity of the average roughness values is a result of the numerous surface asperities on the SROM trunnion, which create a surface texture that is similar to the specific threaded surface of the Secur-fit Max (threaded). The Sz analysis for both the stems has revealed that the average maximum height of the surface geometries is very similar therefore regardless of the fact that one of the surfaces is threaded the overall surface analyses of these two trunnions present similar roughness. Despite their perceived commonality as shown by the Sa and Sz parameters, the threads of the Secur-fit Max result in different contact points with the walls of the bore compared to the SROM trunnion, which should result in a larger contact area. This highlights the shortcoming of the Sa, Sz parameters, and necessitates the introduction of measures of the skewness and kurtosis2 of the surface. Ssk and Sku parameters represent the skewness and kurtosis of the surfaces. For the SROM (non-threaded) and Secur-fit Max (threaded) their Ssk parameters were 0.07 and 0.22, respectively. The negative Ssk value for the SROM indicates that valleys dominate the surface, but its close proximity to zero indicates that dominance of valleys is minor. This correlates with the unthreaded, machine-finished surface of the SROM. Conversely the larger magnitude and positive value of the Ssk for the Secur-fit Max show that its surface is dominated by peaks, as expected for a threaded surface. The lower Sku parameters for the Secur-fit Max at 2.3 compared to 3.5 for the SROM indicates a more periodic surface, which correlates with a threaded finish when compared to the ad-hoc nature of the un-threaded machine finish. Repeated impaction and removal of a modular femoral head may rupture the passivated layer of the trunnion. Retrieval analysis has shown corrosion damage at the trunnion due to the crevice environment, which provides oxygen-starved conditions.15–20 It is also believed that the differences in the compressive loads on the bore surface may explain the variation of fretting damage seen in retrieved implant.16,20–22 The variations of the trunnion surface topography seen in the selected clean commercial stems identify the differences present at the taper interface. The insertion of the trunnions within the bore will result

1

2

Also referred to as the 10-point Height S10z.


Kurtosis is a measured of the “peakedness.”


in different contact points hence force distributions from the bore to the trunnion. The interaction of the bore and trunnion at the taper connection for the threaded stems all have unique contact points and interface due to having different thread height and pitch. For example, the Corail has high threads with a wide pitch, in comparison to a stem such as the SecurFit Max, which has small thread height corresponding to a narrow pitch. The Summit and Trilock have complex repeating patterns and will therefore correspond to their own unique contact points. For example, all the threads on the Summit trunnion will not create a contact point along the bore due to altering heights. This variation in threads will create a unique microenvironment, which can affect the interaction at the interface in comparison to other total hips analyzed. All these trunnions will provide sequential contact points against the bore, however, the concentration of contact points varies due to the variation of the thread pitch. In comparison a non-threaded stem will have area contact against the bore rather than point contact. However, if the asperities are not removed during initial impaction then the contact area of the trunnion against the bore will result in a smaller contact region in comparison to the apparent area. The difference in the contact regions between threaded and non-threaded stems may lead to the severity of micro movement (fretting), which can be experience within the taper junction. The difference in thread height will also provide a difference micro surface topography seen at the interface of the taper junction. This may lead to difference in corrosion severity and wear of the material surface during constant cyclic motion. Commonly other variables such as the offset and contact length also contribute to the total mechanical environment created at the taper junction, however, the surface is a important consideration; where this study has shown with the exception of the offset and contact length, the topography between trunnions vary and can be a valuable consideration in understanding the interface, and ultimately corrosion.

CONCLUSION The use of the optical profilometer demonstrates that commercial trunnions consist of either a threaded or a smooth surface finish. More detailed analysis revealed that the threaded surface for all the trunnions varied in pitch and thread height. The surface roughness parameters showed that the threaded trunnions have a significantly higher roughness in comparison to the smooth. The unique variation in the surface topography of the trunnions may be a contributing factor to the severity of the corrosion damage observed at the trunnion surface. LIMITATION A limitation within this study is that only one stem of each type was analyzed with the assumption that the surface finish are assumed to be uniform within each specific implant time.

105

REFERENCES 1. Skinner HB. 1992. Current biomaterial problems in implants. Instr Course Lect 41:137–144. 2. Fessler H, Fricker DC. 1989. Friction in femoral prosthesis and photoelastic model cone taper joints. Proc Inst Mech Eng H 3:1–14. 3. Erik B, Mathiesen JUL. 1991. Corrosion of modular hip prostheses. Br Ed Soc Bone Joint Surg 73-B:569–575. 4. Collier JP, Surprenant VA, Jensen RE, et al. 1992. Corrosion between the components of modular femoral hip prostheses. Br Ed Soc Bone Joint Surg 74-B:511–517. 5. Caitlin Huot Carlson J, Van Citters DW, Currier JH, et al. 2011. Femoral stem fracture and in vivo corrosion of retrieved modular femoral hips. J Arthrop 00:1–9. 6. Gilbert JL, Mali S, Urban RM, et al. 2011. In vivo oxideinduced stress corrosion cracking of Ti-6Al-4V in a neck– stem modular taper: emergent behavior in a new mechanism of in vivo corrosion. J Biomed Mater Res 100B:584–594. 7. Blunt L, Jiang X. 2003. Advanced techniques for assessment surface topography: development of a asis for 3D surface texture standards. Elsevier. p. 340. 8. Lonardo PMHT, De Chiffre L. 1996. Progress in 3D surface microtopography characterization. Keynote Papers 589–598. 9. Dong WP, Sullivan PJ, Stout KJ. 1994. Comprehensive study of parameters for characterizing three-dimensional surface topography III: parameters for characterizing amplitude and some functional properties. Wear 178:29–43. 10. Dong WP, Sullivan PJ, Stout KJ. 1994. Comprehensive study of parameters for characterizing three-dimensional surface topography IV: parameters for characterizing spatial and hybrid properties. Wear 178:45–60. 11. Gadelmawlaa ES, Kourab MM, Maksoudc TMA, et al. 2002. Roughness parameters. J Mater Process Technol 123:133–145. 12. Petheram TG, Bone M, Joyce TJ, et al. 2013. Surface finish of the Exeter trauma stem: a cause for concern? Bone Joint J 95-B:173–176. 13. Zecchino M. 2003. Characterizing surface quality: why average roughness is not enough. Veeco Intruments, Inc. 14. Nassif NA, Nawabi DH, Stoner K, et al. 2014. Taper design affects failure of large-head metal-on-metal total hip replacements. Clin Orthop Relat Res 472:564–541. 15. Brown SA, Flemming CAC, Merritt K, et al. 1992. Fretting corrosion testing of modular hip designs. Fourth World Biomater Congr (Berl) 268 Abstract. 16. Gilbert JL, Buckley CA, Jacobs JJ. 1993. In vivo corrosion of modular hip prosthesis components in mixed and similar metal combinations. The effect of crevice, stress, motion, and alloy coupling. J Biomed Mater Res 27:1533–1544. 17. Cook SD, Barrack RL, Baffes GC, et al. 1994. Wear and corrosion of the modular interfaces in total hip replacements. Clin Orthop Relat Res. 298:80–88. 18. Kop AM, Swarts E. 2009. Corrosion of a hip stem with a modular neck taper junction: a retrieval study of 16 cases. J Arthrop 24:1019–1023. 19. Kop A, Keogh C, Swarts E. 2012. Proximal component modularity in THA—at what cost? An implant retrieval study. Clin Orthop Relat Res 470:1885–1894. 20. Brown SA, Flemming CAC, Kawalec JS, et al. 1995. Fretting corrosion accelerates crevice corrosion of modular hip tapers. J Appl Biomater 6:19–26. 21. Collier JP, Surprenant VA, Jensen RE, et al. 1992. Corrosion between the components of modular femoral hip prostheses. J Bone Joint Surg Br 74:511–517. 22. Fraitzl CR, Moya LE, Castellani L, et al. 2011. Corrosion at the stem–sleeve interface of a modular titanium alloy femoral component as a reason for impaired disengagement. J Arthrop 26:113–119.e1. JOURNAL OF ORTHOPAEDIC RESEARCH JANUARY 2015