- Email: [email protected]
In-vitro biomechanical evaluation of stress shielding and initial stability of a low-modulus hip stem made of β type Ti-33.6Nb-4Sn alloy
Medical Engineering & Physics 36 (2014) 1665–1671
Contents lists available at ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
In-vitro biomechanical evaluation of stress shielding and initial stability of a low-modulus hip stem made of  type Ti-33.6Nb-4Sn alloy Go Yamako a,∗ , Etsuo Chosa b , Koji Totoribe b , Shuji Hanada c , Naoya Masahashi c , Norikazu Yamada d , Eiji Itoi d a
Organization for Promotion of Tenure Track, University of Miyazaki, 1-1 Gakuen-kibanadai-nishi, Miyazaki 889-2192, Japan Department of Medicine of Sensory and Motor Organs, Division of Orthopedic Surgery, Faculty of Medicine, University of Miyazaki, Miyazaki 889-1692, Japan c Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan d Tohoku University Schoool of Medicine, Sendai 980-8574, Japan b
a r t i c l e
i n f o
Article history: Received 25 March 2014 Received in revised form 18 August 2014 Accepted 7 September 2014 Keywords: Total hip arthroplasty Bone resorption Low-modulus stem Stress-shielding Initial stability
a b s t r a c t Stress shielding-related proximal femoral bone loss after total hip arthroplasty occurs because of the different stiffness of metallic alloy stems and host bone. To overcome this, we fabricated a low-modulus cementless hip stem from -type Ti-33.6Nb-4Sn alloy (TNS). Then we evaluated its stiffness, stress shielding, and initial stability compared with a similar Ti-6Al-4V alloy stem. Stiffness was determined by axial compression and cantilever-bending tests. Thirteen triaxial strain gages measured cortical strain. Stress shielding was defined as the percentage of intact strain after stem insertion. To evaluate initial stability, displacement transducers measured axial relative displacement and rotation. Intact and implanted femurs underwent single-leg-stance loading. Axial stiffness was 56% lower in the TNS stem than in the Ti6Al-4V stem, and bending stiffness of the TNS stem decreased gradually from the proximal region to the distal region, being ≤53% that of the Ti-6Al-4V stem, indicating gradation of Young’s modulus. The TNS stem decreased stress shielding in the proximal calcar region (A1: 83%, B1: 85% relative to intact cortical strain) without affecting the proximal lateral region (B3: 53%). The initial stabilities of the stems were comparable. These findings indicate that the TNS stem with gradation of Young’s modulus minimizes proximal femoral bone loss and biological fixation, improving long-term stability. © 2014 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Total hip arthroplasty (THA) is one of the most beneficial procedures in orthopedic surgery [1]. However, subsequent femoral bone remodeling affects long-term clinical performance. Progressive proximal bone loss can decrease implant stability, causing subsidence and aseptic loosening, resulting in periprosthetic fracture and hindrance to revision surgery [2–5]. The etiology of proximal bone resorption is multifactorial [6], but it is mainly caused by a reduction in load transmitted from implant to bone, i.e., socalled “stress shielding” [7,8]. Consequently, the femur remodels to
Abbreviations: ANOVa, analysis of variance; ISO, International Organization for Standardization; THA, total hip arthroplasty; TNS, -type Ti-33.6Nb-4Sn alloy. ∗ Corresponding author. Tel.: +81 985 58 7332; fax: +81 985 58 7332. E-mail addresses: [email protected], [email protected] (G. Yamako). http://dx.doi.org/10.1016/j.medengphy.2014.09.002 1350-4533/© 2014 IPEM. Published by Elsevier Ltd. All rights reserved.
accommodate the new mechanical situation, obeying Wolff’s law [9–11]. To date, various types of femoral stems, including anatomic, straight, wedge-taper, customized, and short-type stems, have been introduced; nevertheless, stress shielding continues to occur in most cases [12–14]. The main reason for stress shielding is the difference in stiffness between metallic implants and host bone [15,16]. Stress is predominantly transferred through the implant, because the Young’s modulus of metallic implants is generally much higher than that of host bone. Therefore, to decrease stress shielding, low-modulus femoral stems, i.e., isoelastic stems, were first introduced 40 years ago [17–20]. However, inadequate bone fixation and fatigue failures of the plastic materials used led to high clinical failure rates [21–26]. Fixation failure associated with aseptic loosening is caused by the inferior surface properties or incompatibility with bone tissue of their constituent polymers [27]. The Epoch® stem (Zimmer, Inc., Warsaw, IN, USA) is the only successful low-stiffness stem for which long-term clinical performance has been reported [28,29]:
1666
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
proximal bone loss was less than that associated with a more rigid stem design [30,31]. However, the bone-sparing effect remains controversial. Dual X-ray absorptiometry demonstrated that the greatest decrease in mean bone density (27.5%) occurs in the calcar region at 2 years [32]. Development of femoral stems from alternative materials, more closely matched in stiffness to host bone, is desirable to prevent stress shielding. Ti-6Al-4 V alloy (␣-type and -type) is widely used in femoral stems and other orthopedic implants because of its properties, which include excellent biocompatibility, good corrosion resistance, and high strength. However, its Young’s modulus (110 GPa) is 3–4 times higher than that of human cortical bone (approximately 20–30 GPa). Therefore, the development of low-modulus Ti alloys for biomedical applications is ongoing [33–38]. Because Young’s modulus and strength are inversely proportional, the results of these studies have not met target values required for the femoral stem. The femoral neck region must be of high strength to support the patient’s body weight and enable reduction of the neck radius, ensuring a wide range of stem movement around the cup. Recently, Hanada et al. [39,40] and Jung et al. [41] developed a new, low-modulus stem to overcome these issues using -type Ti-33.6Nb-4Sn alloy (TNS). This stem demonstrates high tensile and fatigue strength in the femoral neck region because of local heat treatment, and a low modulus in the portion inserted into the femur. This design is based on the clinical observations that most stem fractures occur in the neck, and that bone loss due to stress shielding mainly occurs in the proximal femur. The Young’s modulus of the distal region of this stem is <55 GPa, approximately half that of Ti-6Al-4V alloy, and much closer to that of bone. In the femoral neck, the tensile and fatigue strengths are >1200 MPa and >800 MPa, respectively. Using optimized local heat treatment, they decreased the Young’s modulus of the TNS to 30 GPa, which is the lowest Young’s modulus of all -type Ti alloys developed for biomedical applications [38]. Moreover, the Ti–Nb–Sn alloy, which contains minimal cytotoxic elements, has a high degree of
bone tissue compatibility, similar to that of Ti-6Al-4V alloy [42]. This novel, low-modulus Ti alloy could supersede the conventional Ti-6Al-4V alloy as a biomaterial for femoral cementless stems; however, the biomechanical performance of this stem has never been investigated. In this study, we evaluated the stiffness, stress shielding, and initial stability of this low-modulus -type TNS hip stem compared with a similar Ti-6Al-4V alloy stem. We hypothesized that the TNS stem would decrease stress shielding and negatively influence initial stability compared with the Ti-6Al-4V stem. 2. Materials and methods 2.1. The low-modulus ˇ-type Ti-33.6Nb-4Sn stem The TNS stem has a fatigue strength of 850 MPa and a tensile strength of 1270 MPa in the femoral neck portion, and a Young’s modulus of 55 GPa in the distal region (Fig. 1). To increase its tensile and fatigue strengths, the femoral neck was inserted into a hole in a copper rod and locally heated to 798 K for 5 h, causing fine ␣ precipitation in the fiber structure. In the heat-treated neck region, the tensile strength of the TNS stem was higher than that of Ti-6Al-4V and other -type Ti alloys developed for biomedical applications [38]. The fatigue strength of the neck of the TNS stem was superior to that of the Ti-6Al-4V (700–800 MPa) [43]. Details of the material properties and fabrication methods of the TNS stem have been described previously [39,40]. The stem has a conically tapering shape. Proximally, bone fixation is supported by a grit-blasted surface in the proximal two-third. This stem design, i.e., straight, conically tapered, is widely used in clinical practice with excellent long-term results [12]. However, this design causes proximal stress shielding in 50–88% cases [44,45]. In addition, mild-to-moderate activity-related thigh pain was reported in 4% of 283 patients [44]. Therefore, this stem type seems to be effective in decreasing Young’s modulus and thus reducing both
Fig. 1. Photographic and X-ray images of a newly developed flexible cementless stem made of -type Ti-33.6Nb-4Sn alloy. The proximal region is grit blast-coated.
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
1667
mechanical vice. Vertical force of ≤500 N was applied 40, 65, 90, and 115 mm distally from the medial curve origin of the stem. The slope of the bending moment–deflection curve was recorded and used to calculate the bending stiffness of each loading point. 2.3. Assessment of stress shielding and initial stability
Fig. 2. Illustration of the axial compression and cantilever-bending tests.
stress shielding and thigh pain. To investigate its effect on stress shielding and initial stability, we fabricated five TNS stems and compared them with five similarly shaped Ti-6Al-4V stems. 2.2. Measurements of axial and bending stiffness Axial compression and cantilever-bending tests were conducted to evaluate the axial and bending stiffness of both the TNS and Ti6Al-4V stems using a material testing machine (Autograph AG-20 kNX Plus; Shimadzu Corporation, Kyoto, Japan; Fig. 2). In the axial compression test, samples of both stem types (n = 5 each) were held in a mechanical clamping vice for metal cutting. Two steel tapered V-shaped flute blocks were used to clamp the distal cylindrical tapered ends of the stems. The resulting working lengths of the hip stems from the top of the clamping block to the center of the stems were set at 100 mm. A vertical axial force was applied to each femoral head using load control (maximum load = 2100 N; loading rate = 600 N/min). The loading direction was parallel to the long axis of the stems. The slope of the axial load–displacement curve was recorded and used to estimate the axial stiffness of each stem. In the cantilever-bending test, the proximal region of the stem was securely clamped using metal molding blocks and a fixed
After the mechanical tests for axial and bending stiffness, both the TNS and Ti-6Al-4 V stems were implanted in composite femurs (Medium, Left, Fourth-Generation Composite Femur, Sawbones; Pacific Research Laboratories, Inc., Vashon, WA, USA) by an experienced orthopedic surgeon to assess stress shielding and initial stem stability. Stress shielding after femoral stem implantation was evaluated by measuring the alteration of cortical surface strain patterns on the intact femur and on femurs implanted with either a TNS or Ti-6Al-4 V stem (n = 5 each). For each femur, 13 triaxial strain-gage rosettes (KFG-3-120-D17; gage length = 3 mm; gage resistance = 120 ; Kyowa Electronic Instruments Co., Ltd., Tokyo, Japan) were bonded on to the femoral surface at five points in the proximal region, and four points in both the middle and distal regions (Fig. 3). To assess stress shielding at different locations around the stem, the equivalent strain (von Mises strain) was used to sum up the strain conditions into one scalar value. To evaluate stress shielding after stem insertion, relative strain was defined as the percentage of intact strain [46]. Thus, a relative strain of 0% represents the situation in which the bone is fully stress shielded, whereas 100% represents the situation in which physiological strain level are maintained. Insertion of a femoral stem changes both the magnitude of the principal strains and the principal strain direction. Thus, principal strains before and after THA are not necessarily comparable. Instead, von Mises equivalent strain was used to express the strain conditions at the strain gage as a scalar parameter. Similar approaches to simplifying the strain conditions into equivalent strain or octahedral shear strain have been used in previous biomechanical studies [46,47]. To evaluate initial stem stability, two linear displacement transducers (GT2; Keyence Corporation, Osaka, Japan) with a 0.1-m resolution and 1-m accuracy were applied to each femur to measure the relative displacement and rotation between stem and femur (Fig. 4). To assess axial displacement, a transducer (S1) was
Fig. 3. Positions of the 13 strain-gage rosettes on the composite femur.
1668
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
Fig. 4. Photograph and illustration of the testing apparatus for applying the load to the femur, showing the positions of the displacement transducers (S1 for axial displacement, S2 for axial rotation).
attached to the greater trochanter with a 4.5-mm cortical screw made of pure Ti to align the longitudinal axis of the stem with the displacement sensor. The relative rotation of the longitudinal axis was also recorded. The transducer (S2) was attached to the greater and lesser trochanters with cortical screws and the sensor head was set at the center of the femoral head and perpendicular to the neck of the stem. The relative rotation was calculated using the distance between the longitudinal axis of the stem and the center of the femoral head (the contact point of the S2 transducer) [48].
compared intact femurs with TNS stem-implanted and Ti-6Al-4V stem-implanted femurs. Statistical analysis was performed using one-way analysis of variance (ANOVA). When ANOVA indicated a significant difference among groups, the difference was evaluated using Fisher’s protected least significant difference test. P < 0.05 was considered statistically significant. 3. Results 3.1. Axial and bending stiffness
2.4. Loading procedure Before testing, each femur was cut 260 mm distally from the top of the greater trochanter and cemented 60 mm deep in a metal box containing dental die stone (New FujiRock® Fast Set; GC Corporation, Tokyo, Japan). To load the femurs, the specimens were fixed in a specially designed jig on a material testing machine (Autograph AGS-10 kNX Plus; Shimadzu Corporation; Fig. 4). Each femur was tilted laterally by 8◦ in the frontal plane and 6◦ dorsally in the sagittal plane to simulate a single-leg stance and create a loading situation, including torsional and bending moments [13,49]. Two orthogonally placed ball-bearing slides were placed above a polyethylene acetabular cup to neutralize critical moments during loading. Vertical load was then applied to the femoral heads through the cup at a rate of 600 N/min up to a maximum of 2100 N. To avoid experimental errors caused by altered femoral head positions after stem implantation, the results were not compared at applied axial load but were compared using a common bending moment [49,50]. The bending moments were calculated using the axial load and horizontal distances of the displaced center of the heads relative to the fixed support of the specimen. The moment reached by all femurs under maximal load was 47 Nm. The equivalent load where this moment was reached was calculated for each bone. The applied load versus strain and displacement data were recorded at a sampling rate of 10 Hz using a multi-input data recording system (NR-600; Keyence Corporation). 2.5. Statistical analysis To compare the axial and bending stiffness and initial stability, we used the Student’s t-test. To evaluate stress shielding, we
The linearity of the axial force–displacement data (TNS stem, R2 = 0.99; Ti-6Al-4V stem, R2 = 0.99) showed that the specimens remained within the linear elastic region, incurring no permanent deformation during the mechanical tests (Fig. 5a). The axial stiffness of the TNS stem (754.8 ± 32.8 N/mm) was 55.9% that of the Ti-6Al-4V stem (1349.4 ± 158.4 N/mm; Fig. 5b), and the difference was significant (P < 0.001). In the cantilever-bending test, the bending stiffness (Nm/mm) of both stems gradually decreased with increasing distance between the origin of the medial curve and the loading point (Table 1). There was a significant difference in bending stiffness between the stems at all loading points (all P < 0.001). The difference in bending stiffness between the stems also increased with increasing distance. The stiffness of the TNS stem as a percentage of that of the Ti-6Al4V stem at loading points of 40, 65, 90, and 115 mm were 72.1%, 61.9%, 55.9%, and 53.4%, respectively. 3.2. Stress shielding The von Mises equivalent strain varied from region to region (Fig. 6). The equivalent strains ranged from 122.8 to 3284.9 strains in intact and operated femurs. In the medial and lateral regions (A1, B1, B3, C1, C3, D1, D3), the equivalent strains of the Ti-6Al-4V stem were each significantly lower than those of the intact femur (all P < 0.05), indicating stress shielding (Table 2). In particular, pronounced stress shielding was evident with the Ti-6Al-4V stem in both the proximal medial and lateral regions (gage positions A1, B1, B3). Conversely, significant differences in the equivalent strains between the TNS stem and the intact femur were evident at B1 (P = 0.003), B3 (P < 0.001), and B4 (P < 0.001). In the proximal medial
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
1669
Fig. 5. Typical experimental raw data of force versus vertical displacement obtained from the axial stiffness test (a), and the axial stiffness of -type Ti-33.6Nb-4Sn and Ti-6Al-4V stems (b).
Fig. 6. Comparison of von Mises equivalent strain on the femoral surface at 13 measuring points for the intact femur, and for femurs implanted with the -type Ti-33.6Nb-4Sn and Ti-6Al-4V stems. *P < 0.05. Table 1 Bending stiffness of the TNS and Ti6Al4 V stem at each loading point.
Table 3 Relative axial displacement (m) and rotation (◦ ).
Distance between medial curve origin and loading point
TNS stem, Nm/mm Ti6Al4V stem, Nm/mm P-value
TNS stem Mean
40 mm
65 mm
90 mm
115 mm
370.2 ± 16.4 513.3 ± 16.3 P < 0.001
214.3 ± 9.7 346.3 ± 10.7 P < 0.001
86.5 ± 5.1 154.8 ± 9.9 P < 0.001
40.0 ± 2.1 74.9 ± 1.9 P < 0.001
Axial displacement (m) Axial rotation (deg.)
Ti-6Al-4V stem SD
Mean
73.0 12.7 0.298 0.083
P-value
SD
67.1 15.5 0.291 0.006
0.262 0.444
3.3. Initial stability
region (A1, B1), the strains of the TNS stem were significantly greater than those of the Ti-6Al-4V stem (P = 0.01 at A1, P = 0.006 at B1). At the anterior and posterior sides, strains in the middle and distal regions (C2, C4, D2, D4) were comparable. However, in the proximal posterior region (B4), a significant increase was observed after insertion of both stems (all P < 0.001). In this region, the average hoop tensile strains of the intact femur, TNS stem, and Ti-6Al-4 V stem were −142.7, 279.0 (versus intact femur, P < 0.001), and 84.2 strain (versus intact femur, P < 0.001), respectively.
With increasing applied load to the femoral head, both the TNS and Ti-6Al-4V stems were displaced distally and internally rotated against the femur. There were no significant differences in axial displacement (P = 0.26) or rotation (P = 0.44) between the TNS and Ti-6Al-4V stems (Table 3). 4. Discussion Stress-shielding-mediated femoral bone loss is inevitable following THA with intramedullary cementless femoral stems [12,14].
Table 2 Relative equivalent strains (%) at the positions of the strain gages. Position of the strain gauge on composite femur
TNS stem Ti6Al4V stem
A1
B1
B2
B3
B4
C1
C2
C3
C4
D1
D2
D3
D4
83.9 57.8
85.3 72.1
114.4 99.8
53.6 53.1
561.8 516.9
92.0 82.5
96.2 87.3
88.4 82.8
95.5 91.7
104.2 84.3
132.1 117.7
96.5 77.3
100.9 97.3
1670
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
In particular, proximal bone loss increases distal loading, and is considered to influence implant fixation. Thigh pain may also occur after THA, and its origin may be related to the relatively high bending stiffness of the implant within the femur [29,51]. The TNS stem was designed to decrease proximal bone loss and thigh pain through decreased bending stiffness. Our in vitro biomechanical study demonstrated that the axial stiffness of the TNS stem is 56% lower than that of the Ti-6Al-4V stem, and that its bending stiffness gradually decreased from the proximal region to the distal region by ≤53% compared with the Ti-6Al-4V stem. This difference in bending stiffness indicates the inhomogenous distribution of the Young’s modulus in the TNS stem: the Ti-6Al-4V stem is made of Ti-6Al-4V alloy with homogeneous material properties, and the stem shape of both implants is the same. Thus, the Young’s modulus of the TNS stem decreased gradually from the proximal region to the distal region. We found that this low-modulus TNS stem decreased stress shielding in the proximal calcar region (relative equivalent strains at A1 and B1 were 83% and 85%, respectively) without affecting the proximal lateral region (B3; 53%). Moreover, the initial stability of the TNS stem was comparable with that of the Ti-6Al-4V stem. This low-modulus TNS stem with a graded Young’s modulus could minimize proximal femoral bone loss and improve biological fixation for long-term stability. Implantation of either stem changed the cortical strain pattern on the femur. A significant reduction in cortical strains in the proximal and middle regions occurred following implantation of the Ti-6Al-4V stem. These findings confirmed those of previous studies on different types of femoral stems, such as anatomic, straight, custom-made, and short-type stems [13,52–55]. In the proximal medial region (A1, B1), higher cortical strains with the TNS stem may be caused by the lower bending stiffness of this stem relative to the Ti-6Al-4V stem. The significant rise in tensile hoop strain observed in the proximal posterior region at B4 after the implantation of either stem can be ascribed to the wedging of the stem; this result is consistent with that of previous studies [46,53]. Furthermore, the greatly increased von Mises strain following implantation of either stem compared with the intact femur may indicate a risk of periprosthetic fracture or secondary instability of the stem. Therefore, further optimization of the stem shape and the gradation of the Young’s modulus of the TNS stem may be necessary to recreate a more physiological load-transfer pattern after THA. Previous biomechanical studies have indicated that decreasing stem stiffness ameliorates stress shielding and prevents severe bone resorption. However, such changes also increase proximal stem–bone interface stresses, which may cause interface debonding and relative motion, and possibly inhibit biological fixation, resulting in loosening [15,56]. Therefore, two incompatible design goals exist with regard to the femoral stem. In fact, the classic, isoelastic RM stem made of polyacetyl resin had to be revised, because micromovements in the proximal part of the stem caused loosening [21]. Hence, Kuiper and Huiskes [57], using numerical design optimization methods, proposed a nonhomogeneous distribution of the Young’s modulus in the femoral stem to solve the two conflicting demands. In the simplified model of a hip stem, in which the distribution of the Young’s modulus is optimized with the aim of maintaining femoral bone, the interface stress was decreased by >50% compared with that of a homogeneously low-modulus stem. The bending test showed a gradual decrease in the Young’s modulus of the TNS stem from the proximal to the distal regions. This gradation was created by local heat treatment, which increases the tensile strength of the stem neck region [39]. Both the Young’s modulus and tensile strength of the TNS alloy can be increased by low-temperature aging treatment [41]. Therefore, local heat treatment of the femoral neck region generated a temperature gradient in the stem by low thermal conductivity [40], and the Young’s modulus was graded accordingly. Thus, this Young’s modulus gradient
decreases the stem–bone interface stress and relative motions, and prevents interface failure. The initial stability of the TNS stem was equal to that of the Ti6Al-4 V stem. Gradation of the Young’s modulus of the TNS stem may have a positive influence on initial stability. Initial stability of a cementless implant is vital for bone bonding to the proximal coated region of the stem, termed osseointegration, a process that guarantees long-term clinical success [58,59]. Interface micromotion exceeding 150 m inhibits bone ingrowth and, instead, fibrous tissue forms, resulting in aseptic loosening [60]. The axial relative displacement measured by the S1 transducer was <150 m for both the TNS and Ti-6Al-4 V stems, and no statistically significant difference was observed. In addition, the bonding strength of the TNS alloy was equal to that of the Ti-6Al-4V alloy [42]. In brief, the shear strength of the Ti-6Al-4V alloy was significantly greater than that of both Co–Cr alloy and stainless steel. This study aimed to demonstrate the biomechanical performance of the newly developed TNS stem. To verify the endurance properties of femoral stems, preclinical standard test methods are defined in International Organization for Standardization (ISO) 7206-4 (Third Edition, 2010-06-15) and ISO 7206-6 (Second Edition, 2013-11-15), which require them to resist dynamic cyclic loading without fracture. Thus, we are now conducting endurance tests of the TNS stem on the basis of the ISO tests to prevent unfavorable clinical conditions before its clinical introduction, and will report the fatigue results of the TNS stem as a separate work. There are several limitations to our study. First, in our test setup, the action of major muscle forces was not included, and only the effect of hip contact force, with one loading configuration, was taken into account. The magnitude and direction of the load acting on hip contact, and muscle forces during daily activities, can influence the stress-shielding pattern and initial stability. Second, in our in vitro preclinical experiments, we could not incorporate adaptive bone remodeling to determine whether progressive bone changes occur around the stem. However, any bone alteration could potentially influence the cortical strain pattern on the femur. Moreover, because we used a composite femur that mimics the properties and shape of bone, our results did not take into account variations in bone shape and quality. Long-term clinical studies are required to evaluate whether these promising in vitro results are applicable in vivo. In conclusion, although stress shielding was not eliminated completely, there was a tendency toward more physiological conditions in the proximal femur after the implantation of the newly developed TNS stem, without compromising initial stability for long-term fixation. The novel, low-modulus TNS alloy may represent a candidate biomaterial for manufacturing cementless femoral stems to replace the conventional Ti-6Al-4V alloy. Funding This work was financially supported by the Program to Disseminate Tenure Tracking System from the Ministry of Education, Culture, Sports, Science and Technology. This study was funded in part by Mizuho Corporation, Tokyo, Japan. Ethical approval Ethical approval not required. Conflict of interest statement The authors declare that they have no financial or personal relationships with any other people or organizations that could have inappropriately influenced or biased this work.
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
References [1] Learmonth ID, Young C, Rorabeck C. The operation of the century: total hip replacement. Lancet 2007;370:1508–19. [2] Engh Jr CA, Young AM, Engh Sr CA, Hopper Jr RH. Clinical consequences of stress shielding after porous-coated total hip arthroplasty. Clin Orthop Relat Res 2003:157–63. [3] Lindahl H. Epidemiology of periprosthetic femur fracture around a total hip arthroplasty. Injury 2007;38:651–4. [4] Taylor M, Tanner KE. Fatigue failure of cancellous bone: a possible cause of implant migration and loosening. J Bone Joint Surg Br 1997;79:181–2. [5] Wilkinson JM, Hamer AJ, Rogers A, Stockley I, Eastell R. Bone mineral density and biochemical markers of bone turnover in aseptic loosening after total hip arthroplasty. J Orthop Res 2003;21:691–6. [6] Sychterz CJ, Engh CA. The influence of clinical factors on periprosthetic bone remodeling. Clin Orthop Relat Res 1996;322:285–92. [7] Van Rietbergen B, Huiskes R, Weinans H, Sumner DR, Turner TM, Galante JO, ESB Research Award 1992. The mechanism of bone remodeling and resorption around press-fitted THA stems. J Biomech 1993;26:369–82. [8] Huiskes R. The various stress patterns of press-fit, ingrown, and cemented femoral stems. Clin Orthop Relat Res 1990;261:27–38. [9] Bobyn JD, Mortimer ES, Glassman AH, Engh CA, Miller JE, Brooks CE. Producing and avoiding stress shielding. Laboratory and clinical observations of noncemented total hip arthroplasty. Clin Orthop Relat Res 1992;274:79–96. [10] Engh CA, Bobyn JD. The influence of stem size and extent of porous coating on femoral bone resorption after primary cementless hip arthroplasty. Clin Orthop Relat Res 1988;231:7–28. [11] Engh Jr CA, Sychterz C, Engh Sr C. Factors affecting femoral bone remodeling after cementless total hip arthroplasty. J Arthroplasty 1999;14:637–44. [12] Khanuja HS, Vakil JJ, Goddard MS, Mont MA. Cementless femoral fixation in total hip arthroplasty. J Bone Joint Surg Am 2011;93:500–9. [13] Bieger R, Ignatius A, Decking R, Claes L, Reichel H, Durselen L. Primary stability and strain distribution of cementless hip stems as a function of implant design. Clin Biomech (Bristol, Avon) 2012;27:158–64. [14] Nysted M, Benum P, Klaksvik J, Foss O, Aamodt A. Periprosthetic bone loss after insertion of an uncemented, customized femoral stem and an uncemented anatomical stem. A randomized DXA study with 5-year follow-up. Acta Orthop 2011;82:410–6. [15] Huiskes R, Weinans H, van Rietbergen B. The relationship between stress shielding and bone resorption around total hip stems and the effects of flexible materials. Clin Orthop Relat Res 1992;274:124–34. [16] Head WC, Bauk DJ, Emerson Jr RH. Titanium as the material of choice for cementless femoral components in total hip arthroplasty. Clin Orthop Relat Res 1995;311:85–90. [17] Bombelli R, Mathys R. Cementless isoelastic RM total hip prosthesis. J R Soc Med 1982;75:588–97. [18] Katoozian H, Davy DT, Arshi A, Saadati U. Material optimization of femoral component of total hip prosthesis using fiber reinforced polymeric composites. Med Eng Phys 2001;23:503–9. [19] Skinner HB. Isoelasticity and total hip arthroplasty. Orthopedics 1991;14:323–8. [20] Morscher E, Bombelli R, Schenk R, Mathys R. The treatment of femoral neck fractures with an isoelastic endoprosthesis implanted without bone cement. Arch Orthop Trauma Surg 1981;98:93–100. [21] Jakim I, Barlin C, Sweet MB. RM isoelastic total hip arthroplasty. A review of 34 cases. J Arthroplasty 1988;3:191–9. [22] Morscher EW, Dick W. Cementless fixation of isoelastic hip endoprostheses manufactured from plastic materials. Clin Orthop Relat Res 1983;7:7–8, 7. [23] Niinimaki T, Puranen J, Jalovaara P. Total hip arthroplasty using isoelastic femoral stems. A seven- to nine-year follow-up in 108 patients. Bone Joint Surg Br 1994;76:413–8. [24] Adam F, Hammer DS, Pfautsch S, Westermann K. Early failure of a press-fit carbon fiber hip prosthesis with a smooth surface. J Arthroplasty 2002;17:217–23. [25] Trebse R, Milosev I, Kovac S, Mikek M, Pisot V. Poor results from the isoelastic total hip replacement: 14–17-year follow-up of 149 cementless prostheses. Acta Orthop 2005;76:169–76. [26] Runne WC, van Sambeek KJ, Stierum JL, van Tongerloo RB. Femoral endoprosthesis fixation with a soft, flexible low modulus stem coating. Four to six year clinical results. Orthopedics 1989;12:529–35. [27] Andrew TA, Flanagan JP, Gerundini M, Bombelli R. The isoelastic, noncemented total hip arthroplasty. Preliminary experience with 400 cases. Clin Orthop Relat Res 1986;206:127–38. [28] Thien TM, Thanner J, Karrholm J. Fixation and bone remodeling around a low-modulus stem seven-year follow-up of a randomized study with use of radiostereometry and dual-energy x-ray absorptiometer. J Arthroplasty 2012;27:134–42, e1. [29] Hartzband MA, Glassman AH, Goldberg VM, Jordan LR, Crowninshield RD, Fricka KB, et al. Survivorship of a low-stiffness extensively porous-coated femoral stem at 10 years. Clin Orthop Relat Res 2010;468:433–40. [30] Karrholm J, Anderberg C, Snorrason F, Thanner J, Langeland N, Malchau H, et al. Evaluation of a femoral stem with reduced stiffness. A randomized study with use of radiostereometry and bone densitometry. J Bone Joint Surg Am 2002;84A:1651–8.
1671
[31] Glassman AH, Crowninshield RD, Schenck R, Herberts P. A low stiffness composite biologically fixed prosthesis. Clin Orthop Relat Res 2001;393:128–36. [32] Akhavan S, Matthiesen MM, Schulte L, Penoyar T, Kraay MJ, Rimnac CM, et al. Clinical and histologic results related to a low-modulus composite total hip replacement stem. J Bone Joint Surg Am 2006;88:1308–14. [33] Wang KK, Gustavson LJ, Dumbleton JH. Microstructure and properties of a new beta titanium alloy, Ti-12MO-6ZR-2FE, developed for surgical implants. ASTM STP 1996;1272:76–86. [34] Kuroda D, Niinomi M, Morinaga M, Kato Y, Yashiro T. Design and mechanical properties of new  type titanium alloys for implant materials. Mater Sci Eng A 1998;243:244–9. [35] Hao YL, Li SJ, Sun SY, Zheng CY, Yang R. Elastic deformation behaviour of Ti24Nb-4Zr-7.9Sn for biomedical applications. Acta Biomater 2007;3:277–86. [36] Jung TK, Semboshi S, Masahashi N, Hanada S. Mechanical properties and microstructures of  Ti-25Nb-11Sn ternary alloy for biomedical applications. Mater Sci Eng C 2013;33:1629–35. [37] Matsumoto H, Watanabe S, Hanada S. Beta TiNbSn alloys with low Young’s modulus and high strength. Mater Trans 2005;46:1070–8. [38] Niinomi M, Nakai M, Hieda J. Development of new metallic alloys for biomedical applications. Acta Biomater 2012;8:3888–903. [39] Hanada S, Masahashi N, Jung TK, Yamada N, Yamako G, Itoi E. Fabrication of a high-performance hip prosthetic stem using beta Ti-33.6Nb-4Sn. J Mech Behav Biomed Mater 2013;30C:140–9. [40] Hanada S, Masahashi N, Jung TK. Effect of stress-induced ␣ martensite on Young’s modulus of  Ti-33.6Nb-4Sn alloy. Mater Sci Eng A 2013;588: 403–10. [41] Jung T-K, Lee H-S, Semboshi S, Masahashi N, Abumiya T, Hanada S. A new concept of hip joint stem and its fabrication using metastable TiNbSn alloy. J Alloys Compd 2012;536:S582–5. [42] Miura K, Yamada N, Hanada S, Jung TK, Itoi E. The bone tissue compatibility of a new Ti–Nb–Sn alloy with a low Young’s modulus. Acta Biomater 2011;7:2320–6. [43] Niinomi M, Nakai M, Akahori T. Frictional wear characteristics of biomedical Ti-29Nb-13Ta-4.6Zr alloy with various microstructures in air and simulated body fluid. Biomed Mater 2007;2:S167–74. [44] Bourne RB, Rorabeck CH, Patterson JJ, Guerin J. Tapered titanium cementless total hip replacements: a 10- to 13-year followup study. Clin Orthop Relat Res 2001;393:112–20. [45] Park MS, Choi BW, Kim SJ, Park JH. Plasma spray-coated Ti femoral component for cementless total hip arthroplasty. J Arthroplasty 2003;18:626–30. [46] Pettersen SH, Wik TS, Skallerud B. Subject specific finite element analysis of stress shielding around a cementless femoral stem. Clin Biomech (Bristol, Avon) 2009;24:196–202. [47] Kessler O, Lacatusu E, Sommers MB, Mayr E, Bottlang M. Malrotation in total knee arthroplasty: effect on tibial cortex strain captured by laser-based strain acquisition. Clin Biomech (Bristol, Avon) 2006;21:603–9. [48] Gotze C, Steens W, Vieth V, Poremba C, Claes L, Steinbeck J. Primary stability in cementless femoral stems: custom-made versus conventional femoral prosthesis. Clin Biomech (Bristol, Avon) 2002;17:267–73. [49] Decking R, Puhl W, Simon U, Claes LE. Changes in strain distribution of loaded proximal femora caused by different types of cementless femoral stems. Clin Biomech (Bristol, Avon) 2006;21:495–501. [50] Cristofolini L, Viceconti M. In vitro stress shielding measurements can be affected by large errors. J Arthroplasty 1999;14:215–9. [51] Brown TE, Larson B, Shen F, Moskal JT. Thigh pain after cementless total hip arthroplasty: evaluation and management. J Am Acad Orthop Surg 2002;10:385–92. [52] Ostbyhaug PO, Klaksvik J, Romundstad P, Aamodt A. An in vitro study of the strain distribution in human femora with anatomical and customised femoral stems. Bone Joint Surg Br 2009;91:676–82. [53] Aamodt A, Lund-Larsen J, Eine J, Andersen E, Benum P, Husby OS. Changes in proximal femoral strain after insertion of uncemented standard and customised femoral stems. An experimental study in human femora. Bone Joint Surg Br 2001;83:921–9. [54] Bieger R, Ignatius A, Reichel H, Durselen L. Biomechanics of a short stem: in vitro primary stability and stress shielding of a conservative cementless hip stem. J Orthop Res 2013;31:1180–6. [55] Kim YH, Kim JS, Cho SH. Strain distribution in the proximal human femur. An in vitro comparison in the intact femur and after insertion of reference and experimental femoral stems. Bone Joint Surg Br 2001;83:295–301. [56] Kuiper JH, Huiskes R. Friction and stem stiffness affect dynamic interface motion in total hip replacement. J Orthop Res 1996;14:36–43. [57] Kuiper JH, Huiskes R. Mathematical optimization of elastic properties: application to cementless hip stem design. J Biomech Eng 1997;119:166–74. [58] Schneider E, Kinast C, Eulenberger J, Wyder D, Eskilsson G, Perren SM. A comparative study of the initial stability of cementless hip prostheses. Clin Orthop Relat Res 1989;248:200–9. [59] Buhler DW, Berlemann U, Lippuner K, Jaeger P, Nolte LP. Three-dimensional primary stability of cementless femoral stems. Clin Biomech (Bristol, Avon) 1997;12:75–86. [60] Pilliar RM, Lee JM, Maniatopoulos C. Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin Orthop Relat Res 1986;208:108–13.
Contents lists available at ScienceDirect
Medical Engineering & Physics journal homepage: www.elsevier.com/locate/medengphy
In-vitro biomechanical evaluation of stress shielding and initial stability of a low-modulus hip stem made of  type Ti-33.6Nb-4Sn alloy Go Yamako a,∗ , Etsuo Chosa b , Koji Totoribe b , Shuji Hanada c , Naoya Masahashi c , Norikazu Yamada d , Eiji Itoi d a
Organization for Promotion of Tenure Track, University of Miyazaki, 1-1 Gakuen-kibanadai-nishi, Miyazaki 889-2192, Japan Department of Medicine of Sensory and Motor Organs, Division of Orthopedic Surgery, Faculty of Medicine, University of Miyazaki, Miyazaki 889-1692, Japan c Institute for Materials Research, Tohoku University, Sendai 980-8577, Japan d Tohoku University Schoool of Medicine, Sendai 980-8574, Japan b
a r t i c l e
i n f o
Article history: Received 25 March 2014 Received in revised form 18 August 2014 Accepted 7 September 2014 Keywords: Total hip arthroplasty Bone resorption Low-modulus stem Stress-shielding Initial stability
a b s t r a c t Stress shielding-related proximal femoral bone loss after total hip arthroplasty occurs because of the different stiffness of metallic alloy stems and host bone. To overcome this, we fabricated a low-modulus cementless hip stem from -type Ti-33.6Nb-4Sn alloy (TNS). Then we evaluated its stiffness, stress shielding, and initial stability compared with a similar Ti-6Al-4V alloy stem. Stiffness was determined by axial compression and cantilever-bending tests. Thirteen triaxial strain gages measured cortical strain. Stress shielding was defined as the percentage of intact strain after stem insertion. To evaluate initial stability, displacement transducers measured axial relative displacement and rotation. Intact and implanted femurs underwent single-leg-stance loading. Axial stiffness was 56% lower in the TNS stem than in the Ti6Al-4V stem, and bending stiffness of the TNS stem decreased gradually from the proximal region to the distal region, being ≤53% that of the Ti-6Al-4V stem, indicating gradation of Young’s modulus. The TNS stem decreased stress shielding in the proximal calcar region (A1: 83%, B1: 85% relative to intact cortical strain) without affecting the proximal lateral region (B3: 53%). The initial stabilities of the stems were comparable. These findings indicate that the TNS stem with gradation of Young’s modulus minimizes proximal femoral bone loss and biological fixation, improving long-term stability. © 2014 IPEM. Published by Elsevier Ltd. All rights reserved.
1. Introduction Total hip arthroplasty (THA) is one of the most beneficial procedures in orthopedic surgery [1]. However, subsequent femoral bone remodeling affects long-term clinical performance. Progressive proximal bone loss can decrease implant stability, causing subsidence and aseptic loosening, resulting in periprosthetic fracture and hindrance to revision surgery [2–5]. The etiology of proximal bone resorption is multifactorial [6], but it is mainly caused by a reduction in load transmitted from implant to bone, i.e., socalled “stress shielding” [7,8]. Consequently, the femur remodels to
Abbreviations: ANOVa, analysis of variance; ISO, International Organization for Standardization; THA, total hip arthroplasty; TNS, -type Ti-33.6Nb-4Sn alloy. ∗ Corresponding author. Tel.: +81 985 58 7332; fax: +81 985 58 7332. E-mail addresses: [email protected], [email protected] (G. Yamako). http://dx.doi.org/10.1016/j.medengphy.2014.09.002 1350-4533/© 2014 IPEM. Published by Elsevier Ltd. All rights reserved.
accommodate the new mechanical situation, obeying Wolff’s law [9–11]. To date, various types of femoral stems, including anatomic, straight, wedge-taper, customized, and short-type stems, have been introduced; nevertheless, stress shielding continues to occur in most cases [12–14]. The main reason for stress shielding is the difference in stiffness between metallic implants and host bone [15,16]. Stress is predominantly transferred through the implant, because the Young’s modulus of metallic implants is generally much higher than that of host bone. Therefore, to decrease stress shielding, low-modulus femoral stems, i.e., isoelastic stems, were first introduced 40 years ago [17–20]. However, inadequate bone fixation and fatigue failures of the plastic materials used led to high clinical failure rates [21–26]. Fixation failure associated with aseptic loosening is caused by the inferior surface properties or incompatibility with bone tissue of their constituent polymers [27]. The Epoch® stem (Zimmer, Inc., Warsaw, IN, USA) is the only successful low-stiffness stem for which long-term clinical performance has been reported [28,29]:
1666
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
proximal bone loss was less than that associated with a more rigid stem design [30,31]. However, the bone-sparing effect remains controversial. Dual X-ray absorptiometry demonstrated that the greatest decrease in mean bone density (27.5%) occurs in the calcar region at 2 years [32]. Development of femoral stems from alternative materials, more closely matched in stiffness to host bone, is desirable to prevent stress shielding. Ti-6Al-4 V alloy (␣-type and -type) is widely used in femoral stems and other orthopedic implants because of its properties, which include excellent biocompatibility, good corrosion resistance, and high strength. However, its Young’s modulus (110 GPa) is 3–4 times higher than that of human cortical bone (approximately 20–30 GPa). Therefore, the development of low-modulus Ti alloys for biomedical applications is ongoing [33–38]. Because Young’s modulus and strength are inversely proportional, the results of these studies have not met target values required for the femoral stem. The femoral neck region must be of high strength to support the patient’s body weight and enable reduction of the neck radius, ensuring a wide range of stem movement around the cup. Recently, Hanada et al. [39,40] and Jung et al. [41] developed a new, low-modulus stem to overcome these issues using -type Ti-33.6Nb-4Sn alloy (TNS). This stem demonstrates high tensile and fatigue strength in the femoral neck region because of local heat treatment, and a low modulus in the portion inserted into the femur. This design is based on the clinical observations that most stem fractures occur in the neck, and that bone loss due to stress shielding mainly occurs in the proximal femur. The Young’s modulus of the distal region of this stem is <55 GPa, approximately half that of Ti-6Al-4V alloy, and much closer to that of bone. In the femoral neck, the tensile and fatigue strengths are >1200 MPa and >800 MPa, respectively. Using optimized local heat treatment, they decreased the Young’s modulus of the TNS to 30 GPa, which is the lowest Young’s modulus of all -type Ti alloys developed for biomedical applications [38]. Moreover, the Ti–Nb–Sn alloy, which contains minimal cytotoxic elements, has a high degree of
bone tissue compatibility, similar to that of Ti-6Al-4V alloy [42]. This novel, low-modulus Ti alloy could supersede the conventional Ti-6Al-4V alloy as a biomaterial for femoral cementless stems; however, the biomechanical performance of this stem has never been investigated. In this study, we evaluated the stiffness, stress shielding, and initial stability of this low-modulus -type TNS hip stem compared with a similar Ti-6Al-4V alloy stem. We hypothesized that the TNS stem would decrease stress shielding and negatively influence initial stability compared with the Ti-6Al-4V stem. 2. Materials and methods 2.1. The low-modulus ˇ-type Ti-33.6Nb-4Sn stem The TNS stem has a fatigue strength of 850 MPa and a tensile strength of 1270 MPa in the femoral neck portion, and a Young’s modulus of 55 GPa in the distal region (Fig. 1). To increase its tensile and fatigue strengths, the femoral neck was inserted into a hole in a copper rod and locally heated to 798 K for 5 h, causing fine ␣ precipitation in the fiber structure. In the heat-treated neck region, the tensile strength of the TNS stem was higher than that of Ti-6Al-4V and other -type Ti alloys developed for biomedical applications [38]. The fatigue strength of the neck of the TNS stem was superior to that of the Ti-6Al-4V (700–800 MPa) [43]. Details of the material properties and fabrication methods of the TNS stem have been described previously [39,40]. The stem has a conically tapering shape. Proximally, bone fixation is supported by a grit-blasted surface in the proximal two-third. This stem design, i.e., straight, conically tapered, is widely used in clinical practice with excellent long-term results [12]. However, this design causes proximal stress shielding in 50–88% cases [44,45]. In addition, mild-to-moderate activity-related thigh pain was reported in 4% of 283 patients [44]. Therefore, this stem type seems to be effective in decreasing Young’s modulus and thus reducing both
Fig. 1. Photographic and X-ray images of a newly developed flexible cementless stem made of -type Ti-33.6Nb-4Sn alloy. The proximal region is grit blast-coated.
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
1667
mechanical vice. Vertical force of ≤500 N was applied 40, 65, 90, and 115 mm distally from the medial curve origin of the stem. The slope of the bending moment–deflection curve was recorded and used to calculate the bending stiffness of each loading point. 2.3. Assessment of stress shielding and initial stability
Fig. 2. Illustration of the axial compression and cantilever-bending tests.
stress shielding and thigh pain. To investigate its effect on stress shielding and initial stability, we fabricated five TNS stems and compared them with five similarly shaped Ti-6Al-4V stems. 2.2. Measurements of axial and bending stiffness Axial compression and cantilever-bending tests were conducted to evaluate the axial and bending stiffness of both the TNS and Ti6Al-4V stems using a material testing machine (Autograph AG-20 kNX Plus; Shimadzu Corporation, Kyoto, Japan; Fig. 2). In the axial compression test, samples of both stem types (n = 5 each) were held in a mechanical clamping vice for metal cutting. Two steel tapered V-shaped flute blocks were used to clamp the distal cylindrical tapered ends of the stems. The resulting working lengths of the hip stems from the top of the clamping block to the center of the stems were set at 100 mm. A vertical axial force was applied to each femoral head using load control (maximum load = 2100 N; loading rate = 600 N/min). The loading direction was parallel to the long axis of the stems. The slope of the axial load–displacement curve was recorded and used to estimate the axial stiffness of each stem. In the cantilever-bending test, the proximal region of the stem was securely clamped using metal molding blocks and a fixed
After the mechanical tests for axial and bending stiffness, both the TNS and Ti-6Al-4 V stems were implanted in composite femurs (Medium, Left, Fourth-Generation Composite Femur, Sawbones; Pacific Research Laboratories, Inc., Vashon, WA, USA) by an experienced orthopedic surgeon to assess stress shielding and initial stem stability. Stress shielding after femoral stem implantation was evaluated by measuring the alteration of cortical surface strain patterns on the intact femur and on femurs implanted with either a TNS or Ti-6Al-4 V stem (n = 5 each). For each femur, 13 triaxial strain-gage rosettes (KFG-3-120-D17; gage length = 3 mm; gage resistance = 120 ; Kyowa Electronic Instruments Co., Ltd., Tokyo, Japan) were bonded on to the femoral surface at five points in the proximal region, and four points in both the middle and distal regions (Fig. 3). To assess stress shielding at different locations around the stem, the equivalent strain (von Mises strain) was used to sum up the strain conditions into one scalar value. To evaluate stress shielding after stem insertion, relative strain was defined as the percentage of intact strain [46]. Thus, a relative strain of 0% represents the situation in which the bone is fully stress shielded, whereas 100% represents the situation in which physiological strain level are maintained. Insertion of a femoral stem changes both the magnitude of the principal strains and the principal strain direction. Thus, principal strains before and after THA are not necessarily comparable. Instead, von Mises equivalent strain was used to express the strain conditions at the strain gage as a scalar parameter. Similar approaches to simplifying the strain conditions into equivalent strain or octahedral shear strain have been used in previous biomechanical studies [46,47]. To evaluate initial stem stability, two linear displacement transducers (GT2; Keyence Corporation, Osaka, Japan) with a 0.1-m resolution and 1-m accuracy were applied to each femur to measure the relative displacement and rotation between stem and femur (Fig. 4). To assess axial displacement, a transducer (S1) was
Fig. 3. Positions of the 13 strain-gage rosettes on the composite femur.
1668
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
Fig. 4. Photograph and illustration of the testing apparatus for applying the load to the femur, showing the positions of the displacement transducers (S1 for axial displacement, S2 for axial rotation).
attached to the greater trochanter with a 4.5-mm cortical screw made of pure Ti to align the longitudinal axis of the stem with the displacement sensor. The relative rotation of the longitudinal axis was also recorded. The transducer (S2) was attached to the greater and lesser trochanters with cortical screws and the sensor head was set at the center of the femoral head and perpendicular to the neck of the stem. The relative rotation was calculated using the distance between the longitudinal axis of the stem and the center of the femoral head (the contact point of the S2 transducer) [48].
compared intact femurs with TNS stem-implanted and Ti-6Al-4V stem-implanted femurs. Statistical analysis was performed using one-way analysis of variance (ANOVA). When ANOVA indicated a significant difference among groups, the difference was evaluated using Fisher’s protected least significant difference test. P < 0.05 was considered statistically significant. 3. Results 3.1. Axial and bending stiffness
2.4. Loading procedure Before testing, each femur was cut 260 mm distally from the top of the greater trochanter and cemented 60 mm deep in a metal box containing dental die stone (New FujiRock® Fast Set; GC Corporation, Tokyo, Japan). To load the femurs, the specimens were fixed in a specially designed jig on a material testing machine (Autograph AGS-10 kNX Plus; Shimadzu Corporation; Fig. 4). Each femur was tilted laterally by 8◦ in the frontal plane and 6◦ dorsally in the sagittal plane to simulate a single-leg stance and create a loading situation, including torsional and bending moments [13,49]. Two orthogonally placed ball-bearing slides were placed above a polyethylene acetabular cup to neutralize critical moments during loading. Vertical load was then applied to the femoral heads through the cup at a rate of 600 N/min up to a maximum of 2100 N. To avoid experimental errors caused by altered femoral head positions after stem implantation, the results were not compared at applied axial load but were compared using a common bending moment [49,50]. The bending moments were calculated using the axial load and horizontal distances of the displaced center of the heads relative to the fixed support of the specimen. The moment reached by all femurs under maximal load was 47 Nm. The equivalent load where this moment was reached was calculated for each bone. The applied load versus strain and displacement data were recorded at a sampling rate of 10 Hz using a multi-input data recording system (NR-600; Keyence Corporation). 2.5. Statistical analysis To compare the axial and bending stiffness and initial stability, we used the Student’s t-test. To evaluate stress shielding, we
The linearity of the axial force–displacement data (TNS stem, R2 = 0.99; Ti-6Al-4V stem, R2 = 0.99) showed that the specimens remained within the linear elastic region, incurring no permanent deformation during the mechanical tests (Fig. 5a). The axial stiffness of the TNS stem (754.8 ± 32.8 N/mm) was 55.9% that of the Ti-6Al-4V stem (1349.4 ± 158.4 N/mm; Fig. 5b), and the difference was significant (P < 0.001). In the cantilever-bending test, the bending stiffness (Nm/mm) of both stems gradually decreased with increasing distance between the origin of the medial curve and the loading point (Table 1). There was a significant difference in bending stiffness between the stems at all loading points (all P < 0.001). The difference in bending stiffness between the stems also increased with increasing distance. The stiffness of the TNS stem as a percentage of that of the Ti-6Al4V stem at loading points of 40, 65, 90, and 115 mm were 72.1%, 61.9%, 55.9%, and 53.4%, respectively. 3.2. Stress shielding The von Mises equivalent strain varied from region to region (Fig. 6). The equivalent strains ranged from 122.8 to 3284.9 strains in intact and operated femurs. In the medial and lateral regions (A1, B1, B3, C1, C3, D1, D3), the equivalent strains of the Ti-6Al-4V stem were each significantly lower than those of the intact femur (all P < 0.05), indicating stress shielding (Table 2). In particular, pronounced stress shielding was evident with the Ti-6Al-4V stem in both the proximal medial and lateral regions (gage positions A1, B1, B3). Conversely, significant differences in the equivalent strains between the TNS stem and the intact femur were evident at B1 (P = 0.003), B3 (P < 0.001), and B4 (P < 0.001). In the proximal medial
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
1669
Fig. 5. Typical experimental raw data of force versus vertical displacement obtained from the axial stiffness test (a), and the axial stiffness of -type Ti-33.6Nb-4Sn and Ti-6Al-4V stems (b).
Fig. 6. Comparison of von Mises equivalent strain on the femoral surface at 13 measuring points for the intact femur, and for femurs implanted with the -type Ti-33.6Nb-4Sn and Ti-6Al-4V stems. *P < 0.05. Table 1 Bending stiffness of the TNS and Ti6Al4 V stem at each loading point.
Table 3 Relative axial displacement (m) and rotation (◦ ).
Distance between medial curve origin and loading point
TNS stem, Nm/mm Ti6Al4V stem, Nm/mm P-value
TNS stem Mean
40 mm
65 mm
90 mm
115 mm
370.2 ± 16.4 513.3 ± 16.3 P < 0.001
214.3 ± 9.7 346.3 ± 10.7 P < 0.001
86.5 ± 5.1 154.8 ± 9.9 P < 0.001
40.0 ± 2.1 74.9 ± 1.9 P < 0.001
Axial displacement (m) Axial rotation (deg.)
Ti-6Al-4V stem SD
Mean
73.0 12.7 0.298 0.083
P-value
SD
67.1 15.5 0.291 0.006
0.262 0.444
3.3. Initial stability
region (A1, B1), the strains of the TNS stem were significantly greater than those of the Ti-6Al-4V stem (P = 0.01 at A1, P = 0.006 at B1). At the anterior and posterior sides, strains in the middle and distal regions (C2, C4, D2, D4) were comparable. However, in the proximal posterior region (B4), a significant increase was observed after insertion of both stems (all P < 0.001). In this region, the average hoop tensile strains of the intact femur, TNS stem, and Ti-6Al-4 V stem were −142.7, 279.0 (versus intact femur, P < 0.001), and 84.2 strain (versus intact femur, P < 0.001), respectively.
With increasing applied load to the femoral head, both the TNS and Ti-6Al-4V stems were displaced distally and internally rotated against the femur. There were no significant differences in axial displacement (P = 0.26) or rotation (P = 0.44) between the TNS and Ti-6Al-4V stems (Table 3). 4. Discussion Stress-shielding-mediated femoral bone loss is inevitable following THA with intramedullary cementless femoral stems [12,14].
Table 2 Relative equivalent strains (%) at the positions of the strain gages. Position of the strain gauge on composite femur
TNS stem Ti6Al4V stem
A1
B1
B2
B3
B4
C1
C2
C3
C4
D1
D2
D3
D4
83.9 57.8
85.3 72.1
114.4 99.8
53.6 53.1
561.8 516.9
92.0 82.5
96.2 87.3
88.4 82.8
95.5 91.7
104.2 84.3
132.1 117.7
96.5 77.3
100.9 97.3
1670
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
In particular, proximal bone loss increases distal loading, and is considered to influence implant fixation. Thigh pain may also occur after THA, and its origin may be related to the relatively high bending stiffness of the implant within the femur [29,51]. The TNS stem was designed to decrease proximal bone loss and thigh pain through decreased bending stiffness. Our in vitro biomechanical study demonstrated that the axial stiffness of the TNS stem is 56% lower than that of the Ti-6Al-4V stem, and that its bending stiffness gradually decreased from the proximal region to the distal region by ≤53% compared with the Ti-6Al-4V stem. This difference in bending stiffness indicates the inhomogenous distribution of the Young’s modulus in the TNS stem: the Ti-6Al-4V stem is made of Ti-6Al-4V alloy with homogeneous material properties, and the stem shape of both implants is the same. Thus, the Young’s modulus of the TNS stem decreased gradually from the proximal region to the distal region. We found that this low-modulus TNS stem decreased stress shielding in the proximal calcar region (relative equivalent strains at A1 and B1 were 83% and 85%, respectively) without affecting the proximal lateral region (B3; 53%). Moreover, the initial stability of the TNS stem was comparable with that of the Ti-6Al-4V stem. This low-modulus TNS stem with a graded Young’s modulus could minimize proximal femoral bone loss and improve biological fixation for long-term stability. Implantation of either stem changed the cortical strain pattern on the femur. A significant reduction in cortical strains in the proximal and middle regions occurred following implantation of the Ti-6Al-4V stem. These findings confirmed those of previous studies on different types of femoral stems, such as anatomic, straight, custom-made, and short-type stems [13,52–55]. In the proximal medial region (A1, B1), higher cortical strains with the TNS stem may be caused by the lower bending stiffness of this stem relative to the Ti-6Al-4V stem. The significant rise in tensile hoop strain observed in the proximal posterior region at B4 after the implantation of either stem can be ascribed to the wedging of the stem; this result is consistent with that of previous studies [46,53]. Furthermore, the greatly increased von Mises strain following implantation of either stem compared with the intact femur may indicate a risk of periprosthetic fracture or secondary instability of the stem. Therefore, further optimization of the stem shape and the gradation of the Young’s modulus of the TNS stem may be necessary to recreate a more physiological load-transfer pattern after THA. Previous biomechanical studies have indicated that decreasing stem stiffness ameliorates stress shielding and prevents severe bone resorption. However, such changes also increase proximal stem–bone interface stresses, which may cause interface debonding and relative motion, and possibly inhibit biological fixation, resulting in loosening [15,56]. Therefore, two incompatible design goals exist with regard to the femoral stem. In fact, the classic, isoelastic RM stem made of polyacetyl resin had to be revised, because micromovements in the proximal part of the stem caused loosening [21]. Hence, Kuiper and Huiskes [57], using numerical design optimization methods, proposed a nonhomogeneous distribution of the Young’s modulus in the femoral stem to solve the two conflicting demands. In the simplified model of a hip stem, in which the distribution of the Young’s modulus is optimized with the aim of maintaining femoral bone, the interface stress was decreased by >50% compared with that of a homogeneously low-modulus stem. The bending test showed a gradual decrease in the Young’s modulus of the TNS stem from the proximal to the distal regions. This gradation was created by local heat treatment, which increases the tensile strength of the stem neck region [39]. Both the Young’s modulus and tensile strength of the TNS alloy can be increased by low-temperature aging treatment [41]. Therefore, local heat treatment of the femoral neck region generated a temperature gradient in the stem by low thermal conductivity [40], and the Young’s modulus was graded accordingly. Thus, this Young’s modulus gradient
decreases the stem–bone interface stress and relative motions, and prevents interface failure. The initial stability of the TNS stem was equal to that of the Ti6Al-4 V stem. Gradation of the Young’s modulus of the TNS stem may have a positive influence on initial stability. Initial stability of a cementless implant is vital for bone bonding to the proximal coated region of the stem, termed osseointegration, a process that guarantees long-term clinical success [58,59]. Interface micromotion exceeding 150 m inhibits bone ingrowth and, instead, fibrous tissue forms, resulting in aseptic loosening [60]. The axial relative displacement measured by the S1 transducer was <150 m for both the TNS and Ti-6Al-4 V stems, and no statistically significant difference was observed. In addition, the bonding strength of the TNS alloy was equal to that of the Ti-6Al-4V alloy [42]. In brief, the shear strength of the Ti-6Al-4V alloy was significantly greater than that of both Co–Cr alloy and stainless steel. This study aimed to demonstrate the biomechanical performance of the newly developed TNS stem. To verify the endurance properties of femoral stems, preclinical standard test methods are defined in International Organization for Standardization (ISO) 7206-4 (Third Edition, 2010-06-15) and ISO 7206-6 (Second Edition, 2013-11-15), which require them to resist dynamic cyclic loading without fracture. Thus, we are now conducting endurance tests of the TNS stem on the basis of the ISO tests to prevent unfavorable clinical conditions before its clinical introduction, and will report the fatigue results of the TNS stem as a separate work. There are several limitations to our study. First, in our test setup, the action of major muscle forces was not included, and only the effect of hip contact force, with one loading configuration, was taken into account. The magnitude and direction of the load acting on hip contact, and muscle forces during daily activities, can influence the stress-shielding pattern and initial stability. Second, in our in vitro preclinical experiments, we could not incorporate adaptive bone remodeling to determine whether progressive bone changes occur around the stem. However, any bone alteration could potentially influence the cortical strain pattern on the femur. Moreover, because we used a composite femur that mimics the properties and shape of bone, our results did not take into account variations in bone shape and quality. Long-term clinical studies are required to evaluate whether these promising in vitro results are applicable in vivo. In conclusion, although stress shielding was not eliminated completely, there was a tendency toward more physiological conditions in the proximal femur after the implantation of the newly developed TNS stem, without compromising initial stability for long-term fixation. The novel, low-modulus TNS alloy may represent a candidate biomaterial for manufacturing cementless femoral stems to replace the conventional Ti-6Al-4V alloy. Funding This work was financially supported by the Program to Disseminate Tenure Tracking System from the Ministry of Education, Culture, Sports, Science and Technology. This study was funded in part by Mizuho Corporation, Tokyo, Japan. Ethical approval Ethical approval not required. Conflict of interest statement The authors declare that they have no financial or personal relationships with any other people or organizations that could have inappropriately influenced or biased this work.
G. Yamako et al. / Medical Engineering & Physics 36 (2014) 1665–1671
References [1] Learmonth ID, Young C, Rorabeck C. The operation of the century: total hip replacement. Lancet 2007;370:1508–19. [2] Engh Jr CA, Young AM, Engh Sr CA, Hopper Jr RH. Clinical consequences of stress shielding after porous-coated total hip arthroplasty. Clin Orthop Relat Res 2003:157–63. [3] Lindahl H. Epidemiology of periprosthetic femur fracture around a total hip arthroplasty. Injury 2007;38:651–4. [4] Taylor M, Tanner KE. Fatigue failure of cancellous bone: a possible cause of implant migration and loosening. J Bone Joint Surg Br 1997;79:181–2. [5] Wilkinson JM, Hamer AJ, Rogers A, Stockley I, Eastell R. Bone mineral density and biochemical markers of bone turnover in aseptic loosening after total hip arthroplasty. J Orthop Res 2003;21:691–6. [6] Sychterz CJ, Engh CA. The influence of clinical factors on periprosthetic bone remodeling. Clin Orthop Relat Res 1996;322:285–92. [7] Van Rietbergen B, Huiskes R, Weinans H, Sumner DR, Turner TM, Galante JO, ESB Research Award 1992. The mechanism of bone remodeling and resorption around press-fitted THA stems. J Biomech 1993;26:369–82. [8] Huiskes R. The various stress patterns of press-fit, ingrown, and cemented femoral stems. Clin Orthop Relat Res 1990;261:27–38. [9] Bobyn JD, Mortimer ES, Glassman AH, Engh CA, Miller JE, Brooks CE. Producing and avoiding stress shielding. Laboratory and clinical observations of noncemented total hip arthroplasty. Clin Orthop Relat Res 1992;274:79–96. [10] Engh CA, Bobyn JD. The influence of stem size and extent of porous coating on femoral bone resorption after primary cementless hip arthroplasty. Clin Orthop Relat Res 1988;231:7–28. [11] Engh Jr CA, Sychterz C, Engh Sr C. Factors affecting femoral bone remodeling after cementless total hip arthroplasty. J Arthroplasty 1999;14:637–44. [12] Khanuja HS, Vakil JJ, Goddard MS, Mont MA. Cementless femoral fixation in total hip arthroplasty. J Bone Joint Surg Am 2011;93:500–9. [13] Bieger R, Ignatius A, Decking R, Claes L, Reichel H, Durselen L. Primary stability and strain distribution of cementless hip stems as a function of implant design. Clin Biomech (Bristol, Avon) 2012;27:158–64. [14] Nysted M, Benum P, Klaksvik J, Foss O, Aamodt A. Periprosthetic bone loss after insertion of an uncemented, customized femoral stem and an uncemented anatomical stem. A randomized DXA study with 5-year follow-up. Acta Orthop 2011;82:410–6. [15] Huiskes R, Weinans H, van Rietbergen B. The relationship between stress shielding and bone resorption around total hip stems and the effects of flexible materials. Clin Orthop Relat Res 1992;274:124–34. [16] Head WC, Bauk DJ, Emerson Jr RH. Titanium as the material of choice for cementless femoral components in total hip arthroplasty. Clin Orthop Relat Res 1995;311:85–90. [17] Bombelli R, Mathys R. Cementless isoelastic RM total hip prosthesis. J R Soc Med 1982;75:588–97. [18] Katoozian H, Davy DT, Arshi A, Saadati U. Material optimization of femoral component of total hip prosthesis using fiber reinforced polymeric composites. Med Eng Phys 2001;23:503–9. [19] Skinner HB. Isoelasticity and total hip arthroplasty. Orthopedics 1991;14:323–8. [20] Morscher E, Bombelli R, Schenk R, Mathys R. The treatment of femoral neck fractures with an isoelastic endoprosthesis implanted without bone cement. Arch Orthop Trauma Surg 1981;98:93–100. [21] Jakim I, Barlin C, Sweet MB. RM isoelastic total hip arthroplasty. A review of 34 cases. J Arthroplasty 1988;3:191–9. [22] Morscher EW, Dick W. Cementless fixation of isoelastic hip endoprostheses manufactured from plastic materials. Clin Orthop Relat Res 1983;7:7–8, 7. [23] Niinimaki T, Puranen J, Jalovaara P. Total hip arthroplasty using isoelastic femoral stems. A seven- to nine-year follow-up in 108 patients. Bone Joint Surg Br 1994;76:413–8. [24] Adam F, Hammer DS, Pfautsch S, Westermann K. Early failure of a press-fit carbon fiber hip prosthesis with a smooth surface. J Arthroplasty 2002;17:217–23. [25] Trebse R, Milosev I, Kovac S, Mikek M, Pisot V. Poor results from the isoelastic total hip replacement: 14–17-year follow-up of 149 cementless prostheses. Acta Orthop 2005;76:169–76. [26] Runne WC, van Sambeek KJ, Stierum JL, van Tongerloo RB. Femoral endoprosthesis fixation with a soft, flexible low modulus stem coating. Four to six year clinical results. Orthopedics 1989;12:529–35. [27] Andrew TA, Flanagan JP, Gerundini M, Bombelli R. The isoelastic, noncemented total hip arthroplasty. Preliminary experience with 400 cases. Clin Orthop Relat Res 1986;206:127–38. [28] Thien TM, Thanner J, Karrholm J. Fixation and bone remodeling around a low-modulus stem seven-year follow-up of a randomized study with use of radiostereometry and dual-energy x-ray absorptiometer. J Arthroplasty 2012;27:134–42, e1. [29] Hartzband MA, Glassman AH, Goldberg VM, Jordan LR, Crowninshield RD, Fricka KB, et al. Survivorship of a low-stiffness extensively porous-coated femoral stem at 10 years. Clin Orthop Relat Res 2010;468:433–40. [30] Karrholm J, Anderberg C, Snorrason F, Thanner J, Langeland N, Malchau H, et al. Evaluation of a femoral stem with reduced stiffness. A randomized study with use of radiostereometry and bone densitometry. J Bone Joint Surg Am 2002;84A:1651–8.
1671
[31] Glassman AH, Crowninshield RD, Schenck R, Herberts P. A low stiffness composite biologically fixed prosthesis. Clin Orthop Relat Res 2001;393:128–36. [32] Akhavan S, Matthiesen MM, Schulte L, Penoyar T, Kraay MJ, Rimnac CM, et al. Clinical and histologic results related to a low-modulus composite total hip replacement stem. J Bone Joint Surg Am 2006;88:1308–14. [33] Wang KK, Gustavson LJ, Dumbleton JH. Microstructure and properties of a new beta titanium alloy, Ti-12MO-6ZR-2FE, developed for surgical implants. ASTM STP 1996;1272:76–86. [34] Kuroda D, Niinomi M, Morinaga M, Kato Y, Yashiro T. Design and mechanical properties of new  type titanium alloys for implant materials. Mater Sci Eng A 1998;243:244–9. [35] Hao YL, Li SJ, Sun SY, Zheng CY, Yang R. Elastic deformation behaviour of Ti24Nb-4Zr-7.9Sn for biomedical applications. Acta Biomater 2007;3:277–86. [36] Jung TK, Semboshi S, Masahashi N, Hanada S. Mechanical properties and microstructures of  Ti-25Nb-11Sn ternary alloy for biomedical applications. Mater Sci Eng C 2013;33:1629–35. [37] Matsumoto H, Watanabe S, Hanada S. Beta TiNbSn alloys with low Young’s modulus and high strength. Mater Trans 2005;46:1070–8. [38] Niinomi M, Nakai M, Hieda J. Development of new metallic alloys for biomedical applications. Acta Biomater 2012;8:3888–903. [39] Hanada S, Masahashi N, Jung TK, Yamada N, Yamako G, Itoi E. Fabrication of a high-performance hip prosthetic stem using beta Ti-33.6Nb-4Sn. J Mech Behav Biomed Mater 2013;30C:140–9. [40] Hanada S, Masahashi N, Jung TK. Effect of stress-induced ␣ martensite on Young’s modulus of  Ti-33.6Nb-4Sn alloy. Mater Sci Eng A 2013;588: 403–10. [41] Jung T-K, Lee H-S, Semboshi S, Masahashi N, Abumiya T, Hanada S. A new concept of hip joint stem and its fabrication using metastable TiNbSn alloy. J Alloys Compd 2012;536:S582–5. [42] Miura K, Yamada N, Hanada S, Jung TK, Itoi E. The bone tissue compatibility of a new Ti–Nb–Sn alloy with a low Young’s modulus. Acta Biomater 2011;7:2320–6. [43] Niinomi M, Nakai M, Akahori T. Frictional wear characteristics of biomedical Ti-29Nb-13Ta-4.6Zr alloy with various microstructures in air and simulated body fluid. Biomed Mater 2007;2:S167–74. [44] Bourne RB, Rorabeck CH, Patterson JJ, Guerin J. Tapered titanium cementless total hip replacements: a 10- to 13-year followup study. Clin Orthop Relat Res 2001;393:112–20. [45] Park MS, Choi BW, Kim SJ, Park JH. Plasma spray-coated Ti femoral component for cementless total hip arthroplasty. J Arthroplasty 2003;18:626–30. [46] Pettersen SH, Wik TS, Skallerud B. Subject specific finite element analysis of stress shielding around a cementless femoral stem. Clin Biomech (Bristol, Avon) 2009;24:196–202. [47] Kessler O, Lacatusu E, Sommers MB, Mayr E, Bottlang M. Malrotation in total knee arthroplasty: effect on tibial cortex strain captured by laser-based strain acquisition. Clin Biomech (Bristol, Avon) 2006;21:603–9. [48] Gotze C, Steens W, Vieth V, Poremba C, Claes L, Steinbeck J. Primary stability in cementless femoral stems: custom-made versus conventional femoral prosthesis. Clin Biomech (Bristol, Avon) 2002;17:267–73. [49] Decking R, Puhl W, Simon U, Claes LE. Changes in strain distribution of loaded proximal femora caused by different types of cementless femoral stems. Clin Biomech (Bristol, Avon) 2006;21:495–501. [50] Cristofolini L, Viceconti M. In vitro stress shielding measurements can be affected by large errors. J Arthroplasty 1999;14:215–9. [51] Brown TE, Larson B, Shen F, Moskal JT. Thigh pain after cementless total hip arthroplasty: evaluation and management. J Am Acad Orthop Surg 2002;10:385–92. [52] Ostbyhaug PO, Klaksvik J, Romundstad P, Aamodt A. An in vitro study of the strain distribution in human femora with anatomical and customised femoral stems. Bone Joint Surg Br 2009;91:676–82. [53] Aamodt A, Lund-Larsen J, Eine J, Andersen E, Benum P, Husby OS. Changes in proximal femoral strain after insertion of uncemented standard and customised femoral stems. An experimental study in human femora. Bone Joint Surg Br 2001;83:921–9. [54] Bieger R, Ignatius A, Reichel H, Durselen L. Biomechanics of a short stem: in vitro primary stability and stress shielding of a conservative cementless hip stem. J Orthop Res 2013;31:1180–6. [55] Kim YH, Kim JS, Cho SH. Strain distribution in the proximal human femur. An in vitro comparison in the intact femur and after insertion of reference and experimental femoral stems. Bone Joint Surg Br 2001;83:295–301. [56] Kuiper JH, Huiskes R. Friction and stem stiffness affect dynamic interface motion in total hip replacement. J Orthop Res 1996;14:36–43. [57] Kuiper JH, Huiskes R. Mathematical optimization of elastic properties: application to cementless hip stem design. J Biomech Eng 1997;119:166–74. [58] Schneider E, Kinast C, Eulenberger J, Wyder D, Eskilsson G, Perren SM. A comparative study of the initial stability of cementless hip prostheses. Clin Orthop Relat Res 1989;248:200–9. [59] Buhler DW, Berlemann U, Lippuner K, Jaeger P, Nolte LP. Three-dimensional primary stability of cementless femoral stems. Clin Biomech (Bristol, Avon) 1997;12:75–86. [60] Pilliar RM, Lee JM, Maniatopoulos C. Observations on the effect of movement on bone ingrowth into porous-surfaced implants. Clin Orthop Relat Res 1986;208:108–13.