Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2014 Jul 16.
Published in final edited form as: J Orthop Res. 2013 Nov 26;32(3):355–361. doi: 10.1002/jor.22534

Peri-implant Bone Strains and Micro-motion Following In-Vivo Service: A Postmortem Retrieval Study of 22 Tibial Components from Total Knee Replacements

Kenneth A Mann 1, Mark A Miller 1, Jacklyn R Goodheart 1, Timothy H Izant 2, Richard J Cleary 3
PMCID: PMC4100998  NIHMSID: NIHMS591456  PMID: 24277230

Abstract

Biological adaptation following placement of total knee replacements (TKR) is known to affect peri-implant bone mineral density (BMD) and implant fixation. The goals of this project were to quantify the proximal tibial bone strain and implant-bone micro-motion for functioning postmortem retrieved TKRs and to assess the strain/micro-motion relationships with chronological (donor age and time in service) and patient (body weight and BMD) factors. Twenty-two tibial constructs were functionally loaded to one body weight (60% medial/40% lateral) and the proximal tibial bone strains and tray/bone micro-motion were measured using a digital image correlation system. Donors with more time in service had higher bone strains (p=0.044), but there was not a significant (p=0.333) contribution from donor age. Donors with lower peri-implant BMD (p=0.0039) and higher body mass (p=0.0286) had higher bone strains. Long term implants (> 11 years) had proximal bone strains (900 με) that were almost twice as high as short term (< 5 years) implants (570 με). Micro-motion was greater for younger donors (p=0.0161) and longer time in service (p=0.0008). Increased bone strain with long term in-vivo service could contribute to loosening of TKRs by failure of the tibial peri-implant bone.

Keywords: knee replacement, loosening, bone strain, postmortem, micro-motion

Introduction

Based on recent estimates of TKR burden in the United States (1), the lifetime probability of receiving a total knee replacement (TKR) is 7% for males and 9.5% for females. In addition, for those already with a primary TKR, the relative risk of revision at some point during the patient’s life has been estimated to be 15% for males and 17.5% for females. Clearly, efforts to reduce the risk of revision in this very large number of TKR recipients is warranted. Substantial efforts have been made to improve implant design, materials, surgical techniques/alignment to improve the outcomes for TKR patients and to minimize the risk of revision. Information from retrospective and prospective clinical studies, implant registries, and retrieval studies provide vital ‘feedback’ to improve total joint function and longevity.

Clinical densitometry studies indicate that following TKR implantation there is a reduction in bone mineral density (BMD) in regions around the implant (2) in the short term. Over the longer term (eight years), bone density in the proximal tibia has been reported to decrease 36% (3) following TKR implantation. The authors report that this bone loss rate (~5%/year) is substantially greater than for osteoporosis patients in the same age group (~1–2%/year) and posit that the loss of mechanical support provided by the peri-implant bone could contribute to high bone strain and loosening of components over the long term. The general loss of bone around implants is thought to be due to a stress shielding phenomenon (4) where articular joint forces bypass bone and instead pass through the stiffer implant, thereby shielding the bone from physiologic stress. However, there is not a directly validated relationship between reduced BMD and high peri-implant strain in TKRs.

Based on these clinical observations, one might anticipate that bone strain would be higher for TKR constructs with more time in service (due to loss in BMD), and from older patients because of the natural loss of bone mass with age (5). To date, a link between chronological factors (time in service or age) with high peri-implant bone strain has not been made. It is also not possible to measure bone strain in-vivo for TKR patients using available tools. High patient mass and body mass index (BMI) have also been cited as factors that increase tibial component loosening rates (6; 7). However, other studies have found no direct linkage between high BMI and loosening (8; 9). High patient mass, coupled with low peri-implant BMD could result in higher bone strains that would put the supporting bone at risk of failure. The relationship between these patient factors (patient mass and peri-implant BMD) and bone strain has not been explored.

In contrast to the loss of peri-implant bone, loss of mechanical connection between implant and bone can occur and is often documented clinically by progressive radiographic lucencies at the bone-implant interface (10). Large inducible micro-motions and migration at the bone-implant interface have been related to increased risk of clinical loosening of tibial components (11; 12). However, to date, most laboratory studies have focused on quantifying the initial stability of components which represent the immediate postoperative situation, and few have explored the state of fixation after a period of in-vivo service (13; 14). Recent work (15) has shown that there is substantial loss of interlock between bone cement and trabecular bone with in-vivo service for cemented tibial components. The loss of interlock could result in increased micro-motion for TKR constructs. As with the peri-implant bone strains described above, it is not known whether TKR micro-motion is influenced by the chronological (age and time in service) and patient factors (peri-implant BMD and patient mass).

The goals of this project were to 1) quantify the proximal tibial bone strain and implant/bone micro-motion due to biomechanical loading using en bloc retrieved tibial components from cemented TKRs and 2) to assess the strain/micro-motion relationships with chronological (donor age and time in service) and patient (body mass and peri-implant BMD) factors. Note that the implants were not obtained from revision surgery for a loose implant, but rather after death; thus the implants could be considered to be successful for the lifetime of the patient. We asked three research questions, guided by the clinical observations above: (1) are peri-implant tibial bone strains greater for implants with more time in service and from older donors?, (2) is tibial bone strain greater for constructs from donors with high body mass and lower peri-implant BMD?, and (3) are inducible micro-motions at the implant/bone interface also related to these chronological and patient factors?

METHODS

En bloc retrieved Total Knee Replacements

Twenty-one human knees with cemented total knee replacements were obtained from the SUNY Upstate Medical University Anatomical Gift Program. These implants were obtained en bloc, were not embalmed, and were frozen at −20 deg C until time of preparation and testing. Donor sex, age, body mass, BMI, and time in service were obtained from the Anatomical Gift Program (Table S1-supplemental). Anterior-posterior and lateral x-rays of the tibial component were used to score fixation status with three levels: well fixed, possibly loose, or loose by our arthroplasty surgeon (THI). One additional total knee construct was prepared in the laboratory using methods that simulated the operative procedure; this was included in the study as a ‘time zero’ construct where there would be no bony remodeling in response to the implantation procedure. All tibial components were metal-backed, the majority had a tibial stem or keel (n=18, one had an 80mm long stem), while four components did not have a stem. Wear of the articulating surface of the polyethylene insert was made using the scoring method described by Hood et al.(16).

Bone Densitometry

Clinical bone density scans (Lunar DPX-IQ, GE Healthcare, Waukesha, Wisconsin) were obtained of the proximal tibia with TKRs in the anterior-posterior direction. Bone mineral density (BMD) regions of interest were identified medial and lateral to the implant stem (peri-implant BMD), and for a region distal to the stem (distal to implant BMD) using the methods described by Abu-Rajab et al (17). An area-weighted total BMD was calculated for the composite of peri-implant and distal to implant BMD. When there was no implant stem, the peri-implant bone was defined as the region from the tibal tray surface to 35mm distal from the tray.

Mechanical Loading Protocol

The proximal tibiae were stripped of soft tissue and fixed in a rectangular pot 70 mm distal to the tibial component. The constructs were mounted in a mechanical test frame (MTS Renew, Eden Prarie MN) such that the tibial tray was perpendicular to the loading axis and axial compressive loads were applied to the tibia through the wear contact patches identified on each polyethylene insert. Non-destructive, donor specific, one body weight loads were applied with 60/40% medial to lateral distribution using a ramp (5 mm/min) loading pattern. Axial loads were chosen for this test in order to replicate the primary loading direction and magnitude found in instrumented knee replacements (18) and also to allow the tibia to remain in a relatively fixed position required for the strain mapping measurement. Because repeated loading cycles were needed to perform all of the strain measurements, one body weight loading was used ensure a non-destructive testing regime. All tests were performed in laboratory air at room temperature.

3-D Digital Image Correlation (DIC) Strain Mapping and Micro-motion Measurement

Strain and micro-motion measurements were made using a three-dimensional non-contacting digital image correlation (DIC) system. Black and white acrylic spray paint was applied to the bone surface to create texture for the DIC analysis. A pair of digital cameras (Diagnositc Instruments, Sterling Heights, Michigan) with 2–55mm telecentric lenses (Edmund Optics, Barrington, NJ) was mounted to a rigid aluminum stand. The cameras were spaced 37 cm apart and were focused on the bone surface (30 μm/pixel resolution) with an intersection angle of 60 deg between cameras (Fig. 1A). Field of view size was a maximum of 36 × 48 mm. MatchID (Catholic University College Ghent, Ghent, Belgium) 3-D DIC software was used to measure bone displacements and calculate strains. A normalized sum of squares difference (NSSD) approach was used for the DIC analysis with sub-regions of 20 pixels and step size of 10 pixels. Strain calculations used a bilinear quadrilateral interpolation (Q4) and the error in minimum compressive strain measurements were determined in a pilot calibration study (RMSE = 32.8 με, 95% confidence intervals 19 to 42 με).

Figure 1.

Figure 1

Open in a new tab

Strain and micro-motion measurements (A) were made using a two camera digital image correlation (DIC) system with an axial load of one body weight (BW) using 60/40% distribution applied to the medial/lateral plateau. Strain measurements were made at five regions (A, L, PL, PM, M) proximally (5–15 mm from tibial tray) and four regions (A, P, M, L) for regions 20 to 50 mm from the tibial tray (B). Micro-motion measurements were made using a 2mm gage length between the metal tray and bone at five locations (A, L, PL, PM, M). Example minimum principal strain maps are shown for PM (C) and lateral (D) regions. Nomenclature: A-anterior, P-posterior, M-medial, L-lateral, PM-postero-medial, and PL-postero-lateral.

Strain was measured over six regions of the bone surface (anterior (A), posterior (P), medial (M), lateral (L), postero-medial (PM), postero-lateral (PL)) (Fig 1B). For each region, the camera system was focused on the region of interest and the minimum principal strain (the most compressive strain) was determined. Strain data was collected moving from proximal to distal from the tibial tray surface (A, M, and L: 5 to 50mm; PM and PL: 5 to 15mm; P: 20–50mm)) using 5×10mm sampling regions in 5 mm increments. Examples of 5×10mm sampling regions are shown in Figure 1C&D. Loading was cycled between 1% BW and 100% BW with image collection after three preconditioning cycles. Proximal bone strain was calculated as the average of the minimum principal strains from the 5–15mm A-PM-PL-M-L regions. Peri-implant bone strain was calculated as the average of minimum principal strain measures from the 20 to 50mm A-P-M-L regions. A division between proximal and peri-implant strains was made based on bone morphology and strain distributions. Proximally there was large change in geometry and bone strain moving from proximal to distal, particularly over PM and PL regions. The PM and PL regions have been identified as strain ‘hot spots’ (19). The regional maximum strain was calculated as the maximum compressive strain magnitude from the five (A-PM-PL-M-L) proximal measurement regions.

Relative motion (micro-motion) between the metal tray and adjacent bone was also measured using the DIC system (error analysis from pilot accuracy study: RMSE = 1.1 μm, 95% CI: 0.54 to 1.16 μm) using a 2mm gage length between tray and bone sampling points (Fig 1B). The change in distance between the two sampling points for the 1% and 100% body weight load was calculated for the A, PM, PL, M, and L regions. Peak and average tray-bone micro-motion were calculated from the five sampling regions.

Statistical Methods

Descriptive statistics were calculated for all donor variables and for the biomechanical response (strain and micro-motion) to 1 body weight loading. Peri-implant bone strain and peak tray-bone micro-motion were chosen as the primary outcome variables to relate to the chronological and donor variables. Linear multiple regression was used to model the relationship between dependent variables (strain or micro-motion) and independent chronological variables (time in service and age) or donor variables (peri-implant BMD and mass).

RESULTS

Twenty-one biomechanical tests were completed successfully with collection of all bone surface strain and implant/bone micro-motion results. One biomechanical test (donor C) resulted in local failure of the bone directly under the tibial tray on the medial and lateral edges of the TKR due to an experimental error that caused 1 BW loading of just the medial and lateral plateau. For this donor, the micro-motion results and strain measures near the interface were excluded.

Peri-implant bone mineral density (BMD) spanned a wide range (0.07–1.16 gr/cm2, mean: 0.61 gr/cm2) and was greater for the BMD distal to the implant (0.46–2.3 gr/cm2, mean: 1.05 gr/cm2) (Table 1). The proximal bone strain (5 to 15mm from the tibial tray) was greater than the peri-implant bone strain (20 – 50mm from tibial tray) (Table 1). Proximally, bone surface strain was highest in the postero-medial (1120±803 (sd) μm) and postero-lateral (1270± 644 μm) regions, followed by medial (508±433 μm), lateral (426±644 μm), and anterior (362±267 μm) regions. In the peri-implant regions, bone strain was greatest in the posterior region (567±654 μm), followed by medial (269±165 μm), lateral (263±243 μm), and anterior (240±143 μm) regions. Details of the distribution of bone strain and tray-bone micro-motion for all of the measured regions is provided in Figure S1 (supplemental).

Table 1.

Descriptive statistics for TKR donor information, bone strain, and tray/bone micro-motion outcome measures (n=22).

Mean Median Std Dev Min Max
Age (years) 78.5 84.5 11.0 54 90
Age at Implantation (years) 69.5 72.5 10.4 42 82
Time in Service (years) 9.0 10 5.6 0 22
Mass (kg) 84.8 86 13.5 61 103
BMI (kg/m2) 29.4 29.1 4.17 22.7 36.5
Peri-implant bone mineral density (gr/cm2) 0.61 0.51 0.29 0.07 1.16
Distal to Implant bone mineral density (gr/cm2) 1.05 0.91 0.47 0.46 2.30
Total bone mineral density (gr/cm2) 0.77 0.65 0.34 0.28 1.57
Proximal bone strain (5–15mm) (με) 743 733 369 221 1400
Peri-implant bone strain (20–50mm) (με) 298 254 148 119 791
Regional max bone strain (με) 1530 1390 828 342 3000
Tray-bone peak micro-motion (μm) 32 17 31 1.5 105
Tray-bone mean micro-motion (μm) 16 8 17 0.7 69
Open in a new tab

Based on radiographic scoring (Table S1), there was substantial overlap of the peak tray-bone micro-motion between the ‘fixed’ group (range: 1.5 to 89 μm, median: 16.5 μm, n=16), ‘possibly loose’ group (range: 11.5 to 105 μm, median: 17.9 μm, n=5), and ‘loose’ group (87 μm, n=1). This indicates that radiographic scoring from x-rays alone is not very specific or sensitive to actual implant stability for this sample population of retrievals. Median tray/bone micro-motion for the five sampling regions were similar (anterior: 4.4 μm, postero-medial: 5.8 μm, postero-lateral: 4.1 μm, medial: 3.9 μm, lateral: 6.7 μm). In several retrievals, even though some regions of the implant/bone interface had relatively high micro-motions (>50 μm), there were always regions on the same retrieval with small micro-motions (<= 10 μm). There were few indications of vertical ‘liftoff’ where there was an increase in distance between the tray and bone; 4 cases of 22 had a region with liftoff greater than 3 μm, with the largest of 8.5 μm.

The results of the primary outcomes (peri-implant bone strain and peak micro-motion) are shown as a function of the chronological and patient factors (Figure 2). Using the linear regression models (Table 2), donors with more time in service had higher peri-implant bone strains (p=0.044), but there was not a statistically significant contribution (p=0.333) from donor age. The axial distribution of bone strains (Figure 3) for time in service groups of 0 to 6, 7 to 11, and 12–22 years shows that the longest time in service group had higher bone strains both proximally and distally, compared to the shorter time in service groups. For the patient variables of BMD and body mass, donors with lower peri-implant BMD (p=0.0039) and higher mass (p=0.0286) had higher bone strains (Figure 2). Younger donors (p=0.0161) and longer time in service (p=0.0008) were correlated with greater peak interface micro-motion. However, tray-bone micro-motion was not significantly correlated with either peri-implant BMD (p=0.455) or body mass (p=0.081). Although not a primary research question, it is interesting to note that PE wear score was strongly correlated with time in service (r2=0.7, p<0.0001).

Figure 2.

Figure 2

Open in a new tab

Contour plots of distal bone strain as a function of chronological factors (age and time in service) (A) and patient factors (body mass and peri-implant bone mineral density) (B) for 22 tibial components. Peak implant/bone micro-motion is shown as a function of chronological (C) and patient factors (D). Open circles indicate female TKR donors, filled circle indicate male TKR donors.

Table 2.

Linear regression models for peri-implant bone strain and implant/bone micro-motion (dependent variables) due to 1 BW loading with chronological factors (donor age and time in service) and patient factors (peri-implant bone mineral density (BMD) and body mass) as independent variables.

Response Term Parameter Term Estimate Standard Error P value
Peri-implant bone strain (με), r2 = 0.30, p=0.033, RMSE = 130.1 με Intercept −22.0 206 0.916
Age (years) 2.74 2.75 0.333
Time in service (years) 11.6 5.4 0.044
Peri-implant bone strain (με), r2 = 0.36, p = 0.0139, RMSE = 124 ue Intercept −2.87 191 0.99
Peri-implant BMD (gr/cm2) −415 126 0.0039
Mass (kg) 6.52 2.75 0.0286
Tray-bone peak micro-motion (μm), r2 = 0.50, p = 0.0019, RMSE = 23 μm Intercept 100.6 37.2 0.0146
Age (years) −1.32 0.50 0.0161
Time in service (years) 3.92 0.98 0.0008
Tray-bone peak micro-motion (μm), r2 = 0.19, p = 0.14, RMSE = 29 μm Intercept −66.9 48.8 0.187
Peri-implant BMD (gr/cm2) −26.1 34.2 0.455
Mass (kg) 1.38 0.74 0.081
Open in a new tab

Figure 3.

Figure 3

Open in a new tab

Spatial distribution of bone strain as a function of years in service was divided into short term (0 – 6 years), medium term (7 – 11 years), and long term (12 – 22 year) groups. Mean and standard error of the mean shown.

DISCUSSION

The goal of the current study was to assess the influence of chronological (age and time in service) and patient (body mass and tibial bone density) factors on the bone strain patterns and micro-motion for functioning total knee replacements. In response to research question #1, bone surface strains were greater for donors with longer time in service, but did not depend on donor age for our sample population. Donors with short time in service (< 5 years) had proximal bone strains (mean: 570 με), but with longer time in service (> 11 years), proximal bone strains were almost twice as high (mean: 900 με) for 1 body weight (BW) loading. The short term proximal bone strains (<5 years) were similar to those reported for laboratory prepared TKR implantations in cadaver bone (520 με/BW) without remodeling (19), after normalizing to 1 BW loading. Medial and lateral bone strains reported for intact proximal tibia (20; 21) at 10 mm (125 με/BW), 25 mm (160 με/BW) and 50 mm (75 με/BW) from the articular surface (see Figure S1) were less than the strains (~200 με/BW) reported here for short term service (< 5 years), and much less than that for long term service (~500 με/BW). These findings suggest that the postmortem retrieved TKRs may not be strain shielded compared to intact tibia, and over the longer term may have higher than normal bone strain. Additional work comparing strain mapping in age/sex matched control cadaver tibia without implants is needed to confirm this observation.

Medial-midshaft tibial bone strain has been measured in vivo in humans performing a variety of activities (22–24) and compressive strains of 308 to 414 με have been reported during walking. These are similar to the bone strains measured at the most distal 50mm level in the current study (223, and 310 με for short and long term implantation) for 1 body weight loading. If this loading is scaled to gait levels (~2.6 BW) (18), and assume that bone strains scale linearly with load, strain magnitudes would increase to 580 to 805 με (short to long term), which approaches twice the strain magnitude measured in vivo for gait. This further supports the concept that the bone strains measured here may exceed normal in vivo levels.

The largest regional bone strain (3000 με) for one body weight loading was less than the yield strain reported for trabecular and cortical bone (~7000 με) (25; 26). If we scale to load levels for level walking (~2.6 BW), then three donors (C, E(R), O(R) from Table S1) would be estimated to exceed the yield strain for normal bone. Note that donor C was the bone that was overloaded on the medial and lateral margins when 1BW was applied to just the medial and lateral plateaus. These results suggest that regional bone failure would be possible if larger loads were applied and is the subject of future investigation in our lab. It is also possible that the strain-loading response is not linear and higher loads do not result in proportional increases in bone strain. Increased peri-implant bone strains have been correlated with increased implant migration and component loosening (27).

In response to research question #2, bone strains were found to be higher for donors with lower BMD and higher body mass. Low BMD is associated with increased risk of bone fracture, but in the context of TKR there also may be less stress applied through the bone below the tibial tray due to stress shielding (28; 29). There does not appear to be clinical data that directly relates lower peri-implant BMD with increased risk of TKR loosening, but the work here supports the concept that low BMD constructs will have increased bone strain. There is also clinical data showing that bisphosphonates can maintain bone mass and reduce revision risk (30); maintaining bone mass may prevent an increase in bone strains thereby reducing risk of bone yielding. High body mass has also been shown to be a clinical factor associated with increased component migration and implant loosening (6). Thus, high body mass, coupled with lower BMD could result in an increased risk of failure due to higher peri-implant bone strains. Note that without BMD in the regression model, body mass (by itself) was not a predictor of bone strain (r2=0.01). This may partially explain why some clinical studies have not reported a clear relationship between high mass or BMI and aseptic loosening. Adding BMD measures to clinical studies may be useful to help explore this issue. Finally, it is important to note that each construct was loaded with the donor’s body mass (one body weight), so that an equitable comparison between donors with high and low body mass could be made.

In response to research question #3, micro-motion was greater for younger donors and for donors with longer time in service. Loss of implant fixation resulting in increased micro-motion is a plausible failure scenario and is consistent with clinical data that shows that the revision rate for younger patients was greater than older patients, and that there is an increased risk of revision with time in service (31). However, it should be noted that the micro-motions measured here (peak median of 17 μm with a maximum of 105 μm) were relatively small compared to magnitudes known to result fibrous tissue formation with porous-surface implants (150 μm) (32). Micro-motion magnitude was similar to a recent study (13) of postmortem retrieved TKRs where 500N loads were applied to the medial and lateral plateaus and micro-motions were generally less than 20 μm with a maximum of 300 μm.

This project has several limitations. The loading was uni-axial and did not include forces experienced by the tibia during the entire gait cycle or other activities. However, axial loading represents, by far, the largest force component based on information obtained from instrumented knee replacements (18). This retrieval study included a variety of metal-backed tibial components from a number of different manufacturers. It is likely that implant design, as well as component alignment and cement distribution all contribute to the stability and stress distribution in the implant, interface, and bone. Of particular note were the four components that did not have a central stem. Presence of a stem was not a significant predictor in the regression models of peri-implant strain for the chronological (p=0.132) or donor (p=0.708) parameters. However, despite the variety of metal-backed implant types, our results showed that chronological and donor parameters could be correlated with micro-motion and bone strains.

Bone strain and micro-motion were only measured on the periphery of the construct. Internal bone strain, particularly bone strain near the distal tip of stem, is likely important in describing the load transfer from tibial tray to bone. Failure of trabecular bone supporting the tibial trays would also be an important component of the process. Micro-motion is likely lower in regions away from the periphery of the tray, because the periphery of the tray often show loss of interlock between cement and bone. Many of the specimens used in this study were from older donors (80+ years), and the mean age at implantation was (69.5 ± 10.4 years). This is very similar to the registry data from Australia (68.8 ± 9.5 years) (33) suggesting that our retrievals are representative of a typical TKR age distribution.

Taking into account in-vivo changes to fixation and supporting bone, the results from this work suggest two possible scenarios for loosening of the tibial components in TKR. Younger, more active patients may load the implant/bone interface such that there is increased inducible micro-motion, and this could increase with time in service due to loss of interlock between cement and bone. The mechanism for this is not clear, but is likely related to morphology changes at the interface due to polyethylene debris (34) or fluid induced lysis (35). Older patients with long time in service may also have a higher risk of loosening due to failure of bone supporting the implant due to high bone strains. This may also be related to low peri-implant BMD and high body mass. Further work is needed to determine if high magnitude joint loading results in increased inducible micro-motion and clinically relevant component migration for these postmortem retrievals.

Supplementary Material

Figure S1

Figure S1 (Supplemental). Minimum principal bone strain (compression shown as positive) on the proximal tibial surface as a function of anatomical location and distance from the tibial tray. Micro-motion between the metal tray and bone with 2mm gage length is shown. Results are stratified by time in service. Existing literature data for intact tibial bone strain (no implant), normalized to 1 body weight loading (* Reilly et al (21), +Bourne et al (20)) are shown as dashed lines. Mean and standard error of the mean results are shown.

Table S1 (Supplemental): Donor information for 21 postmortem retrieved cemented knee replacements and one laboratory prepared construct (0 years in service). There were 15 individual donors (A-O) with 7 bilateral (14 TKRs) and 8 unilateral TKRs.

Acknowledgments

This work was funded by National Institutes of Health AR42017. We would like to thank Pascal Lava and Demitri Debruyne from the Catholic University College Ghent, Belgium for providing the MatchID software used in this project. Dr. Izant is a paid consultant to Stryker Orthopaedics and receives research funding from Stryker Orthopaedics for total joint arthroplasty projects, but these are unrelated to the content and scope of the present study.

References

  • 1.Losina E, Thornhill TS, Rome BN, Wright J, Katz JN. The dramatic increase in total knee replacement utilization rates in the United States cannot be fully explained by growth in population size and the obesity epidemic. J Bone Joint Surg Am. 2012;94:201–207. doi: 10.2106/JBJS.J.01958. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 2.Li MG, Nilsson KG. Changes in bone mineral density at the proximal tibia after total knee arthroplasty: a 2-year follow-up of 28 knees using dual energy X-ray absorptiometry. J Orthop Res. 2000;18:40–47. doi: 10.1002/jor.1100180107. [DOI] [PubMed] [Google Scholar]
  • 3.Levitz CL, Lotke PA, Karp JS. Long-term changes in bone mineral density following total knee replacement. Clin Orthop Relat Res. 1995:68–72. [PubMed] [Google Scholar]
  • 4.Huiskes R, Weinans H, Grootenboer HJ, Dalstra M, Fudala B, Slooff TJ. Adaptive bone-remodeling theory applied to prosthetic-design analysis. J Biomech. 1987;20:1135–1150. doi: 10.1016/0021-9290(87)90030-3. [DOI] [PubMed] [Google Scholar]
  • 5.Burr DB. Muscle strength, bone mass, and age-related bone loss. J Bone Miner Res. 1997;12:1547–1551. doi: 10.1359/jbmr.1997.12.10.1547. [DOI] [PubMed] [Google Scholar]
  • 6.Berend ME, Ritter MA, Hyldahl HC, Meding JB, Redelman R. Implant migration and failure in total knee arthroplasty is related to body mass index and tibial component size. J Arthroplasty. 2008;23:104–109. doi: 10.1016/j.arth.2008.05.020. [DOI] [PubMed] [Google Scholar]
  • 7.Foran JR, Mont MA, Etienne G, Jones LC, Hungerford DS. The outcome of total knee arthroplasty in obese patients. J Bone Joint Surg Am. 2004;86-A:1609–1615. doi: 10.2106/00004623-200408000-00002. [DOI] [PubMed] [Google Scholar]
  • 8.Bordini B, Stea S, Cremonini S, Viceconti M, De Palma R, Toni A. Relationship between obesity and early failure of total knee prostheses. BMC Musculoskelet Disord. 2009;10:29. doi: 10.1186/1471-2474-10-29. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Jackson MP, Sexton SA, Walter WL, Walter WK, Zicat BA. The impact of obesity on the mid-term outcome of cementless total knee replacement. J Bone Joint Surg Br. 2009;91:1044–1048. doi: 10.1302/0301-620X.91B8.22129. [DOI] [PubMed] [Google Scholar]
  • 10.Berend ME, Ritter MA, Meding JB, Faris PM, Keating EM, Redelman R, Faris GW, Davis KE. Tibial component failure mechanisms in total knee arthroplasty. Clin Orthop Relat Res. 2004:26–34. doi: 10.1097/01.blo.0000148578.22729.0e. [DOI] [PubMed] [Google Scholar]
  • 11.Uvehammer J, Karrholm J. Inducible displacements of cemented tibial components during weight-bearing and knee extension observations during dynamic radiostereometry related to joint positions and 2 years history of migration in 16 TKR. J Orthop Res. 2001;19:1168–1177. doi: 10.1016/S0736-0266(01)00046-8. [DOI] [PubMed] [Google Scholar]
  • 12.Pijls BG, Valstar ER, Nouta KA, Plevier JW, Fiocco M, Middeldorp S, Nelissen RG. Early migration of tibial components is associated with late revision: a systematic review and meta-analysis of 21,000 knee arthroplasties. Acta Orthop. 2012;83:614–624. doi: 10.3109/17453674.2012.747052. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Rao AS, Engh JA, Engh GA, Parks NL. Mechanical stability of well-functioning tibial baseplates from postmortem-retrieved total knee arthroplasties. J Arthroplasty. 2010;25:481–485. doi: 10.1016/j.arth.2009.01.006. [DOI] [PubMed] [Google Scholar]
  • 14.Gebert de Uhlenbrock A, Puschel V, Puschel K, Morlock MM, Bishop NE. Influence of time in-situ and implant type on fixation strength of cemented tibial trays - a post mortem retrieval analysis. Clin Biomech (Bristol, Avon) 2012;27:929–935. doi: 10.1016/j.clinbiomech.2012.06.008. [DOI] [PubMed] [Google Scholar]
  • 15.Mann KA, Miller MA, Pray CL, Verdonschot N, Janssen D. A new approach to quantify trabecular resorption adjacent to cemented knee arthroplasty. J Biomech. 2012;45:711–715. doi: 10.1016/j.jbiomech.2011.12.008. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Hood RW, Wright TM, Burstein AH. Retrieval analysis of total knee prostheses: a method and its application to 48 total condylar prostheses. J Biomed Mater Res. 1983;17:829–842. doi: 10.1002/jbm.820170510. [DOI] [PubMed] [Google Scholar]
  • 17.Abu-Rajab RB, Watson WS, Walker B, Roberts J, Gallacher SJ, Meek RM. Peri-prosthetic bone mineral density after total knee arthroplasty. Cemented versus cementless fixation. J Bone Joint Surg Br. 2006;88:606–613. doi: 10.1302/0301-620X.88B5.16893. [DOI] [PubMed] [Google Scholar]
  • 18.Kutzner I, Heinlein B, Graichen F, Bender A, Rohlmann A, Halder A, Beier A, Bergmann G. Loading of the knee joint during activities of daily living measured in vivo in five subjects. J Biomech. 2010;43:2164–2173. doi: 10.1016/j.jbiomech.2010.03.046. [DOI] [PubMed] [Google Scholar]
  • 19.Green GV, Berend KR, Berend ME, Glisson RR, Vail TP. The effects of varus tibial alignment on proximal tibial surface strain in total knee arthroplasty: The posteromedial hot spot. J Arthroplasty. 2002;17:1033–1039. doi: 10.1054/arth.2002.35796. [DOI] [PubMed] [Google Scholar]
  • 20.Bourne RB, Finlay JB. The influence of tibial component intramedullary stems and implant-cortex contact on the strain distribution of the proximal tibia following total knee arthroplasty. An in vitro study. Clin Orthop Relat Res. 1986:95–99. [PubMed] [Google Scholar]
  • 21.Reilly D, Walker PS, Ben-Dov M, Ewald FC. Effects of tibial components on load transfer in the upper tibia. Clin Orthop Relat Res. 1982:273–282. [PubMed] [Google Scholar]
  • 22.Burr DB, Milgrom C, Fyhrie D, Forwood M, Nyska M, Finestone A, Hoshaw S, Saiag E, Simkin A. In vivo measurement of human tibial strains during vigorous activity. Bone. 1996;18:405–410. doi: 10.1016/8756-3282(96)00028-2. [DOI] [PubMed] [Google Scholar]
  • 23.Lanyon LE, Hampson WG, Goodship AE, Shah JS. Bone deformation recorded in vivo from strain gauges attached to the human tibial shaft. Acta Orthop Scand. 1975;46:256–268. doi: 10.3109/17453677508989216. [DOI] [PubMed] [Google Scholar]
  • 24.Al Nazer R, Lanovaz J, Kawalilak C, Johnston JD, Kontulainen S. Direct in vivo strain measurements in human bone-a systematic literature review. J Biomech. 2012;45:27–40. doi: 10.1016/j.jbiomech.2011.08.004. [DOI] [PubMed] [Google Scholar]
  • 25.Morgan EF, Keaveny TM. Dependence of yield strain of human trabecular bone on anatomic site. Journal of Biomechanics. 2001;34:569–577. doi: 10.1016/s0021-9290(01)00011-2. [DOI] [PubMed] [Google Scholar]
  • 26.Biewener AA. Safety factors in bone strength. Calcif Tissue Int. 1993;53(Suppl 1):S68–74. doi: 10.1007/BF01673406. [DOI] [PubMed] [Google Scholar]
  • 27.Taylor M, Tanner KE, Freeman MA. Finite element analysis of the implanted proximal tibia: a relationship between the initial cancellous bone stresses and implant migration. J Biomech. 1998;31:303–310. doi: 10.1016/s0021-9290(98)00022-0. [DOI] [PubMed] [Google Scholar]
  • 28.Bartel DL, Burstein AH, Santavicca EA, Insall JN. Performance of the tibial component in total knee replacement. J Bone Joint Surg Am. 1982;64:1026–1033. [PubMed] [Google Scholar]
  • 29.Chong DY, Hansen UN, van der Venne R, Verdonschot N, Amis AA. The influence of tibial component fixation techniques on resorption of supporting bone stock after total knee replacement. J Biomech. 2011;44:948–954. doi: 10.1016/j.jbiomech.2010.11.026. [DOI] [PubMed] [Google Scholar]
  • 30.Prieto-Alhambra D, Javaid MK, Judge A, Murray D, Carr A, Cooper C, Arden NK. Association between bisphosphonate use and implant survival after primary total arthroplasty of the knee or hip: population based retrospective cohort study. BMJ. 2011;343:d7222. doi: 10.1136/bmj.d7222. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Wainwright C, Theis JC, Garneti N, Melloh M. Age at hip or knee joint replacement surgery predicts likelihood of revision surgery. J Bone Joint Surg Br. 2011;93:1411–1415. doi: 10.1302/0301-620X.93B10.27100. [DOI] [PubMed] [Google Scholar]
  • 32.Jasty M, Bragdon C, Burke D, O’Connor D, Lowenstein J, Harris WH. In vivo skeletal responses to porous-surfaced implants subjected to small induced motions. J Bone Joint Surg Am. 1997;79:707–714. doi: 10.2106/00004623-199705000-00010. [DOI] [PubMed] [Google Scholar]
  • 33.Australian Orthopaedic Association National Joint Replacement Registry. Adelaide, SA: 2009. www.aoa.org.au. [DOI] [PubMed] [Google Scholar]
  • 34.Fisher J, Jennings LM, Galvin AL, Jin ZM, Stone MH, Ingham E. 2009 Knee Society Presidential Guest Lecture: Polyethylene wear in total knees. Clin Orthop Relat Res. 2010;468:12–18. doi: 10.1007/s11999-009-1033-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.Fahlgren A, Bostrom MP, Yang X, Johansson L, Edlund U, Agholme F, Aspenberg P. Fluid pressure and flow as a cause of bone resorption. Acta Orthop. 2010;81:508–516. doi: 10.3109/17453674.2010.504610. [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Figure S1

Figure S1 (Supplemental). Minimum principal bone strain (compression shown as positive) on the proximal tibial surface as a function of anatomical location and distance from the tibial tray. Micro-motion between the metal tray and bone with 2mm gage length is shown. Results are stratified by time in service. Existing literature data for intact tibial bone strain (no implant), normalized to 1 body weight loading (* Reilly et al (21), +Bourne et al (20)) are shown as dashed lines. Mean and standard error of the mean results are shown.

Table S1 (Supplemental): Donor information for 21 postmortem retrieved cemented knee replacements and one laboratory prepared construct (0 years in service). There were 15 individual donors (A-O) with 7 bilateral (14 TKRs) and 8 unilateral TKRs.

NIHMS591456-supplement-Figure_S1.tif (687.2KB, tif)

RESOURCES