Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Apr 1.
Published in final edited form as: J Orthop Res. 2021 Jun 6;40(4):862–870. doi: 10.1002/jor.25107

The Relative Contribution of Bone Microarchitecture and Matrix Composition to Implant Fixation Strength in Rats

Kyle D Anderson 1, Frank C Ko 1,2, Spencer Fullam 2, Amarjit S Virdi 1,2, Markus A Wimmer 1,2, DR Sumner 1,2, Ryan D Ross 1,2
PMCID: PMC8633073  NIHMSID: NIHMS1711888  PMID: 34061392

Abstract

Bone microarchitectural parameters significantly contribute to implant fixation strength but the role of bone matrix composition is not well understood. To determine the relative contribution of microarchitecture and bone matrix composition to implant fixation strength, we placed titanium implants in 12-week-old intact Sprague-Dawley rats, Ovariectomized-Sprague-Dawley rats, and Zucker Diabetic Fatty rats. We assessed bone microarchitecture by microcomputed tomography, bone matrix composition by Raman spectroscopy, and implant fixation strength at 2, 6, and 10 weeks post-implantation. A stepwise linear regression model accounted for 83.3% of the variance in implant fixation strength with osteointegration volume/total volume (50.4%), peri-implant trabecular bone volume fraction (14.2%), cortical thickness (9.3%), peri-implant trabecular crystallinity (6.7%), and cortical area (2.8%) as the independent variables. Group comparisons indicated that osseointegration volume/total volume was significantly reduced in the ovariectomy group at week 2 (~28%) and week 10 (~21%) as well as in the diabetic group at week 10 (~34%) as compared to the age matched Sprague-Dawley group. The crystallinity of the trabecular bone was significantly elevated in the ovariectomy group at week 2 (~4%) but decreased in the diabetic group at week 10 (~3%) with respect to the Sprague-Dawley group. Our study is the first to show that bone microarchitecture explains most of the variance in implant fixation strength, but that matrix composition is also a contributing factor. Therefore, treatment strategies aimed at improving bone-implant contact and peri-implant bone volume without compromising matrix quality should be prioritized.

Keywords: Implant Fixation Strength, Bone Composition, Bone-Implant Contact, MicroCT, Raman Spectroscopy

Introduction

Orthopedic implant failure represents a significant challenge to patient quality of life. Compared to patients undergoing primary total joint replacement (TJR) surgery, patients undergoing revision surgery for a failed implant are at a higher risk for in-hospital mortality, postoperative hemorrhage, pulmonary embolism, and infection1. Revision surgeries are also more costly procedures and can result in longer hospital stays1. Therefore, increasing primary implant survivorship would both maximize patient quality of life and reduce the economic burden of revision TJR surgery. The most common cause of implant failure is aseptic loosening, leading to loss of bone implant fixation2. Identifying the factors that contribute to implant fixation strength can potentially guide in the development of strategies to prevent implant loosening and eventual failure.

Clinically, implant fixation strength is assessed using planar X-ray imaging or dual energy X-ray absorptiometry (DXA)-based measures of bone mineral density (BMD)3. Both strategies primarily assess the quantity of bone in proximity to the implant as surrogate markers for bone-implant contact. In animal models, bone-implant contact is a significant driver of implant fixation strength, but additional factors independently contribute to implant fixation. For example, we recently found that osseointegration (OV/TV), peri-implant trabecular bone volume fraction (BV/TV), and cortical bone thickness (Ct.Th) independently contribute to implant fixation in rats4. Although these three factors were able to explain a significant portion of the variance in the implant fixation strength, we are not aware of any study that accounts for the contribution of the bone matrix composition.

Bone matrix composition is defined as the chemical composition of the bone at the tissue-level. Bone matrix composition is altered in prevalent diseases relevant to orthopedic surgery such as osteoporosis and type 2 diabetes mellitus (T2DM). While osteoporosis is mainly thought of as a disease of low bone mineral density, both preclinical5 and clinical observation studies6–9 have demonstrated that bone matrix composition is also altered by osteoporosis and these alterations contribute to fragility fracture risk10. T2DM is another condition known to change bone matrix composition in pre-clinical11 and clinical studies12,13, which likely contribute to the elevated rates of fragility fracture despite normal bone mineral density14,15.

Despite the clear contribution of bone matrix composition to fracture risk in these metabolic bone disorders, the importance of bone matrix composition to implant fixation has been relatively unexplored. Therefore, the goal of the current study was to determine the relative contribution of bone microarchitecture and matrix composition to implant fixation strength. To vary both microarchitecture and matrix composition simultaneously, we used rat models of osteoporosis (ovariectomy surgery) and T2DM (Zucker Diabetic Fatty Rat) to examine implant fixation strength, bone-implant contact (osseointegration), peri-implant microarchitecture, and bone matrix composition at 2, 6, and 10 weeks following implant surgery. We then performed step-wise linear regression modeling to determine the relative contributions of the microarchitectural and composition bone properties to implant fixation strength. Based on our previous research4, we hypothesized that bone-implant contact would provide the largest contribution to implant fixation strength and that both microarchitecture and matrix composition independently contribute to fixation strength.

Methods

Animals

Female outbred Sprague-Dawley (n=24) and Zucker Diabetic Fatty rats (n=12) were divided into three experimental groups: intact Sprague-Dawley (SPD), ovariectomized-Sprague-Dawley (OVX), and Zucker Diabetic Fatty (ZDF). The OVX surgery occurred at 10 weeks of age. SPD and OVX rats were kept on a standard rodent diet (Purina LabDiet 5008, 6.5% kcal from fat) while ZDF rats were maintained on a high-fat (Research Diet Inc #D12468, 28.8% kcal from fat) chow to induce overt diabetes. Diabetic status was confirmed by circulating glucose measures of >250 mg/dL for the entirety of the study. All groups were allowed to access to food and water ad libitum.

Implants

15 mm long by 1.5 mm diameter titanium rods (99.6% in purity, Goodfellow, Oakdale, PA) were roughened by sonicating in hexane, methanol, acetone, and water for 15 min each, followed by dual acid etching according to previous established methods16. Implants were kept sterilized in 70% ethanol until the time of surgery.

Rodent Implant Placement Surgery

All animals had bilateral titanium implants placed into their distal femoral canals at 12 weeks of age. Animals were randomly divided into three groups for post-surgical sacrifice: 2 weeks, 6 weeks, and 10 weeks (n=4 per group). After sacrifice, the right femurs were excised, cleaned of the surrounding soft tissue, and stored in saline-soaked gauze at −20°C. Two animals were excluded due to improper implant placement resulting in the following number of animals per group: SPD (n=11), OVX (n=12), ZDF (n=11). Animal procedures were approved by the Rush University Animal Care and Use Committee.

Slab Preparation & Selection

Following scanning, right femora were encased in a non-infiltrating epoxy resin (EpoThin, Buehler) and cut into ~3mm thick transverse slabs (Exakt). Slabs were then hand-ground using a silicon carbide polishing pad (grit size 60, ~260 micron size) with ethylene glycol lubrication to provide a smoothened surface (Buehler). The distal most slab, ~500 μm proximal to distal growth plate, was selected for imaging via microcomputed tomography (μCT) and pushout testing while an adjacent slab was used for Raman spectroscopy.

Microcomputed Tomography (μCT)

Imaging of Peri-Implant Bone

Slabs were submerged in a phosphate buffed saline solution to prevent dehydration and imaged with μCT (Scanco μCT50 14.8 μm isotropic voxels, 90 kVp, 88 μA, 750 ms integration, 1600 projections/180°, 0.5 mm Al filter). Measurements were made transverse to the long axis of the implant and bone along the entire 3 mm thickness of the slab. The primary μCT outcome variables were abbreviated according to ASBMR guidelines and include: trabecular bone volume fraction (BV/TV), trabecular number (Tb.N), trabecular thickness (Tb.Th), trabecular separation (Tb.Sp), cortical bone area (Ct.Ar), total cortical area (Tt.Ar), cortical porosity (Ct.Po), and cortical thickness (Ct.Th).

High Resolution Imaging of Osseintegration

High resolution images were collected from the same slab section to assess bone-implant contact using validated scan settings17 (μCT50 1.5 μm isotropic voxels, 90 kVP, 88 μA, 750 ms integration, 1600 projections/180°, 0.5 mm Al filter). Measurements were made at 50% of the slab thickness for a total of 0.37 mm. The primary outcome variable was the osseointegration volume/total volume (OV/TV).

Raman Spectroscopy

Region selection

Bone matrix composition was assessed in three compartments; peri-implant trabecular bone, peri-implant cortical bone, and osseointegrated bone (bone within ~50 μm of the implant surface). Three spectra were collected and averaged to obtain a single matrix compositional measurement within each of the three compartments per animal. Spectra from cortical bone was collected approximately halfway between the endosteal and periosteal surface. Spectra from trabecular bone was collected at approximately half the thickness of an individual trabeculae.

Spectral Acquisition & Analysis

Raman spectra were collected using a 785 nm laser with a grading of 600 gr/mm, a 50X objective with a 12 sec acquisition time, and averaging 3 spectral accumulations per ROI (Horiba LabRAM HR Evolution). The resulting spectra were evaluated according to previously defined matrix composition spectral peaks18, and integrated areas calculated by PLS Toolbox (Version 8.9; Eigenvector Research Inc., Manson, WA). The mineral-to-matrix ratio was defined by the ratio of the integrated areas of ν1PO4 at ~960 cm−1 per proline at ~855 cm−1. Type B carbonate substitution was defined by the ratio of the integrated areas of the CO3 peak at ~1072 cm−1 per ν1PO4 at ~960 cm−1. Post-translational modifications were defined as the hydroxyproline/proline ratio (Hyp at ~877 cm−1 per Pro at ~922 cm−1). Pentosidine content was defined as the peak at ~1362 cm−1 per CH2-wag at ~1453 cm−1. Carboxy-methyl-lysine content was defined as the peak at ~1150 cm−1 per CH2-wag. The fullwidth at half maximum of the ν1PO4 peak (at ~960 cm−1) was calculated by Origin (Version 2019; OriginLab, Northampton, MA) and defined as a measure of crystallinity. The outcome variables include: mineral-to-matrix ratio (mineral-to-matrix), type B carbonate substitution (carb. sub.), post translational modifications (post trans. mod.), pentosidine content (pen content), carboxy-methyl-lysine (CML content), and crystallinity.

Implant Pushout Testing

Testing Apparatus

Implant fixation strength was assessed within the same slab samples used for the μCT analysis. Pushout testing was performed using a materials testing instrument (Criterion 43, MTS Systems Eden Prarie, MN) according to a previously established protocol19. The plunger was a 304 stainless steel dowel pin (Uxcell, Hong Kong) with dimensions of 15.8 mm x 1 mm. The base plate was made from stainless steel with a 2.5 mm hole. The sample was centered over the hole of the base plate and preloaded with 1N. Tests were conducted with a 0.1 mm/s displacement rate and 100 Hz data acquisition rate. The resultant force displacement curves were used to calculate the force at failure. Implant fixation strength was calculated by dividing this force by the implant surface area, which was calculated according to the surface area of a 1.5 mm by 3 mm cylinder. Two test samples, both from SPD rats at two weeks post implantation, had irregularities in test performance, and were excluded from further analysis.

Statistical Analysis

Prism (Version 8; GraphPad, San Diego, CA) and SPSS (Version 19.0); SPSS Inc, Chicago, IL) software packages were used for plotting and data analysis, respectively. A two-way analysis of variance was also used to determine the effects of the rat model (SPD, OVX and ZDF), time post-implantation (2, 6, and 10 weeks), and their interaction. When group effects were significant, a student’s t-test was used to determine post-hoc significance without p-value correction. A value of p<0.05 was used as the threshold for statistical significance. The coefficient of variation (CV) was determined to assess the variability of measured variables collapsed across both model and time post-implantation. Stepwise linear regression was used to determine the factors that contribute to implant fixation strength. Microarchitectural variables that were entered into the stepwise linear regression included BV/TV, Tb.N, Tb.Th, Tb.Sp, Ct.Ar, Tt.Ar, Ct.Po, Ct.Th, and OV/TV and matrix compositional variables included mineral-to-matrix ratio, type b carbonate substitution, post-translational modifications, pentosidine content, carboxy-methyl-lysine content, and crystallinity. Three separate model were constructed, one limited to the microarchitectural parameters, one limited to the matrix composition parameters, and one inclusive of all measured parameters. The measured fixation strength was then plotted as a function of the predicted fixation strength from the statistical model.

Results

Implant fixation strength

Implant fixation strength was significantly altered by model, time, and their interaction (Figure 1). When compared to SPD animals, implant fixation strength was significantly reduced in the OVX (~96% at week 2, ~62% at week 6, and ~69% at week 10) and ZDF animals (~86% at week 2, ~69% at week 6, and ~84% at week 10). The OVX animals also had decreased implant fixation strength compared to the ZDF animals at all experimental timepoints (~71% at week 2, ~26% at week 6, ~97% at week 10). There was also a general increase in implant fixation strength overtime as denoted by the significant time effect. The increase in the SPD implant fixation strength over time was much greater than the other two groups as indicated by the significant model by time interaction term.

Figure 1:

Figure 1:

Open in a new tab

Fixation strength was determined by dividing the force at failure by the implant surface area. Data is reported as the means and standard deviations (SD) with each data point representing an individual animal. Week 2, week 6, and week 10 represent the elapsed time post-implant surgery. Results from the two-way ANOVA are reported in the figure legend. When appropriate, significant post-hoc group effects are indicated by a letter. A. Significantly different compared to the Sprague Dawley (SPD) group. B. Significantly different compared to the Ovariectomized-Sprague-Dawley (OVX) group. C. significantly different compared to the Zucker Diabetic Fatty (ZDF) group. The coefficient of variation was calculated by dividing the standard deviation by the mean of data of a given variable from all groups.

Predictors of implant fixation strength

Stepwise linear regression analysis was performed to determine the relative contribution of microarchitectural and matrix compositional variables to implant fixation strength (Table 1). The most significant variable was osseointegration, OV/TV, explaining 50.4% of the variance, followed by BV/TV and Ct.Th, which explained an additional 14.2% and 9.3%, respectively. Trabecular crystallinity explained an additional 6.7% and Ct.Ar an addition 2.8% of to the variance in fixation strength. In total, the structural and compositional explained 83.3% of the variance in implant fixation strength (Figure 2). When the stepwise linear regression analysis was applied to only the microarchitectural variables the most significant variables were OV/TV, BV/TV, and Ct.Th with identical contributions to model and the additional variable of Tb.N contributing 3.3% to the variance explained. When the stepwise linear regression analysis was applied to only the matrix compositional variables, none of the variables were included in the overall model.

Table 1:

Linear regression model outcomes.

Stepwise Model Summary r2 Δ r2 Adjusted r2 Δ Adjusted r2 Sig.
1. OV/TV 0.520 0.520 0.504 0.504 p<0.0001
2. OV/TV, BV/TV 0.668 0.149 0.645 0.142 p<0.0001
3. OV/TV, BV/TV, Ct.Th 0.764 0.096 0.738 0.093 p<0.0001
4. OV/TV, BV/TV, Ct.Th, Tb.Crystallinity 0.830 0.067 0.805 0.067 p<0.0001
5. OV/TV, BV/TV, Ct.Th, Tb.Crystallinity, Ct.Ar 0.860 0.030 0.833 0.028 p<0.0001
Open in a new tab

Results of the stepwise linear regression that include osseointegration volume/ total volume (OV/TV), bone volume fraction (BV/TV), cortical thickness (Ct.Th), trabecular crystallinity (Tb.Crystallinity), and cortical area (Ct.Ar). Models ranked in the order of variable importance to the regression.

Figure 2:

Figure 2:

Open in a new tab

The predicted implant fixation strength was plotted against the measured fixation strength. The results of the linear regression are denoted by the r2 value in the top right of the graph. Each dot represents an individual animal and the color-coding denotes the group and time post-implantation from each group. Groups are abbreviated as follows: Sprague-Dawley (SPD), Ovariectomized-Sprague-Dawley (OVX), and Zucker Diabetic Fatty (ZDF).

Microarchitecture and Bone Matrix Composition.

Although our primary goal was focused on determining the relative contribution of bone microarchitecture and matrix composition to implant fixation strength, we also analyzed the effect of model and time on the independent variables. The largest microarchitectural and compositional variables to contribute to the stepwise linear regression (osseointegrated microarchitecture and trabecular bone matrix composition) are presented in the main text (Figure 3 and Table 2 respectfully) while trabecular microarchitecture, cortical microarchitecture, cortical matrix composition, and osseointegrated matrix composition are presented in the supplemental materials (see Supplemental Materials). Group comparisons indicated that OV/TV was significantly reduced in the ovariectomy group at week 2 (~28%) and week 10 (~21%) as well as in the diabetic group at week 10 (~34%) as compared to the age matched Sprague-Dawley group. There was also a significant time effect for OV/TV, as OV/TV generally increased with time. Trabecular bone crystallinity was the only compositional measure to produce a significant group effect. The crystallinity was significantly elevated in the ovariectomy group at week 2 (~4%) but decreased in the diabetic group at week 10 (~3%) with respect to their age match controls. There was also a significant time effect in the mineral-to-matrix, carbonate substitution and post translational modification parameters measured within the trabecular compartment. In general, the mineral-to-matrix tended to increase with time, while carbonate substitution and the post-translational modifications decreased.

Figure 3:

Figure 3:

Open in a new tab

The effect of Ovariectomized Sprague-Dawley (OVX) and Zucker Diabetic Fatty (ZDF) group on osseointegrated bone (OV/TV) as determined by μCT and results from the two-way ANOVA are reported in the figure legend. Data is reported as the means and standard deviations (SD). When appropriate, significant post-hoc group effects are indicated by a letter. Significant differences (p<0.05) compared to the Sprague-Dawley (SPD) group are denoted by the letter A. Significant differences (p<0.05) compared to the OVX group are denoted by the letter B. Significant differences (p<0.05) compared to ZDF group are denoted by the letter C. Coefficient of variation was calculated by dividing the standard deviation by the mean of data of a given variable from all groups.

Table 2.

Trabecular Matrix Composition

Trabecular Composition Week 2 Week 6 Week 10 Two-Way ANOVA Results Coefficient of Variation
Rat Model SPD OVX ZDF SPD OVX ZDF SPD OVX ZDF Group Time Interaction
Mineral to Matrix 10.34 (1.45) 11.51 (3.49) 11.46 (2.68) 16.12 (5.87) 16.52 (4.90) 13.79 (4.31) 11.04 (3.17) 17.93A (2.56) 15.31 (4.59) p=0.2134 p=0.0231 p=0.3807 0.31
Carbonate Sub 0.21 (0.01) 0.23 (0.01) 0.22 (0.01) 0.21 (0.01) 0.21 (0.01) 0.20 (0.01) 0.22 (0.01) 0.21 (0.01) 0.19A (0.01) p=0.0850 p=0.0046 p=0.0498 0.06
Post Trans Mod. 0.94 (0.14) 0.96 (0.20) 0.85 (0.13) 0.71 (0.24) 0.65 (0.14) 0.77 (0.28) 0.87 (0.17) 0.57A (0.03) 0.67 (0.16) p=0.3034 p=0.0075 p=0.3092 0.26
PEN Content 0.25 (0.05) 0.33 (0.09) 0.32 (0.05) 0.31 (0.09) 0.38 (0.14) 0.25 (0.08) 0.35 (0.05) 0.45 (0.07) 0.40 (0.20) p=0.1531 p=0.0624 p=0.8173 0.33
CML Content 0.13 (0.03) 0.14 (0.03) 0.15 (0.03) 0.18 (0.03) 0.15 (0.01) 0.16 (0.01) 0.13 (0.03) 0.16 (0.03) 0.16 (0.04) p=0.7193 p=0.1597 p=0.6239 0.20
Crystallinity 17.66 (0.17) 18.35A,C (0.37) 17.49B (0.50) 17.64 (0.03) 17.80 (0.52) 17.25 (0.32) 17.81 (0.25) 18.01C (0.44) 17.33A,B (0.16) p=0.0004 p=0.2238 p=0.5842 0.03
Open in a new tab

The effect of Ovariectomized Sprague-Dawley (OVX) and Zucker Diabetic Fatty (ZDF) group on trabecular mineral to matrix ratio (Mineral to matrix), carbonate substitution (Carb. Sub.), post translational modifications (Post Trans. Mod.), pentosidine content (PEN Content), carboxy-methyl-lysine content (CML content), and crystallinity as determined by Raman spectroscopy. Data is reported as the means and standard deviations (SD). Week 2, week 6, and week 10 represent the elapsed time after implant placement before sacrifice. Results from the two-way ANOVA are reported in the figure legend. When appropriate, significant post-hoc group effects are indicated by a letter. Significant differences (p<0.05) compared to the Sprague-Dawley (SPD) group are denoted by the letter A. Significant differences (p<0.05) compared to the OVX group are denoted by the letter B. Significant differences (p<0.05) compared to ZDF group are denoted by the letter C. Coefficient of variation was calculated by dividing the standard deviation by the mean of data of a given variable from all groups.

Discussion

Orthopedic implant failure is a significant clinical challenge that may benefit from a better understanding of the parameters that contribute to implant fixation strength. In the present study we sought to determine the relative contribution of bone microarchitectural and matrix composition on implant fixation strength in rats using osteoporotic (OVX) and type 2 diabetes mellitus (ZDF) rat models to increase the variance in the measured parameters. Consistent with our hypothesis, our results demonstrate that OV/TV remained the strongest predictor and that both microarchitectural and matrix compositional parameters independently contribute to implant fixation strength, explaining 83.3% of the variance. Had we not included matrix composition, the structural parameters explained 77.1% of the variance.

Our previous work determined that OV/TV, BV/TV, and Ct.Th are significant predictors of implant fixation strength in rats, explaining 83% of the variance in fixation strength4. We also demonstrate in the present study that again the three most important variables are OV/TV, BV/TV, and Ct.Th. Taken together, these studies support the robustness of these measures in determining implant fixation strength. The current study also demonstrates that bone matrix composition, in particular, peri-implant trabecular crystallinity, contributes to the explanation of variance in fixation strength.

Crystallinity is a measure of the size and shape of the hydroxyapatite crystalline structure as a result of a highly regulated mineralization process of type I collagen in bone tissue20. Our finding that trabecular crystallinity significantly contributed to implant fixation strength could indicate that slight disruptions to the mineralization process could have a deleterious effect on bone-implant mechanical properties. Previous work has shown that crystallinity, also determined by Raman spectroscopy, was significantly associated with tissue-level strength and stiffness in human cortical bone21. While the crystallinity of the trabecular bone contributed more to implant fixation strength than that of the osseointegrated bone, previous studies have demonstrated that loss of implant fixation strength is associated with failure of trabeculae rather than failure of the bone-implant interface22.

The OVX rat is a well-characterized model of osteoporosis that recapitulates the loss of bone mass and deterioration of bone microarchitecture noted in osteoporosis23. Consistent with previous studies of implant fixation in OVX rats24, we noted a significant reduction in peri-implant bone microarchitecture, osseointegration, and implant fixation strength when compared to SPD rats. In addition to effects on bone structure, ovariectomy has been found to increase mineral-to-matrix ratio and crystallinity in the bone matrix of rodents25,26 as well as in samples from humans with osteoporosis compared to healthy controls9,10,27,28. In the current study, crystallinity was increased in OVX rats but this effect was specific to the trabecular bone compartment and dependent on the experimental duration. This finding could be due to OVX-induced changes in mineralization kinetics, which have been reported in humans with idiopathic osteoporosis29. It is also possible this effect was greatest at 2 weeks post-implantation because this time point is closer to the initial period of bone formation after implant placement, while BV/TV in the OXV group decreases at 6 and 10 weeks, which likely reflects a period of greater bone resorption.

Type 2 Diabetes Mellitus (T2DM) is another metabolic disease that has been shown to alter skeletal properties. Unlike osteoporosis, diabetic patients are often at either normal or increased BMD16,30–32. Pre-clinical animal models of T2DM, however, typically do not recapitulate the clinical state and generally have decreased bone mass, particularly in the trabecular compartment30. Similarly, we found a significant reduction in peri-implant trabecular bone mass. We also found a significant reduction in the amount of osseointegrated bone in the ZDF rats, which has similarly been reported in other rodent models of T2DM33–36. Clinically, patients with T2DM are at increased fracture risk, despite the relatively normal BMD measures37 likely due to changes in the bone matrix composition that cause embrittlement of the bone, such as an accumulation of Advanced Glycation End-products (AGEs) within the collagen matrix38,39. While AGEs accumulation, which include pentosidine and carboxy-methyl-lysine, are commonly reported in bone biopsies from T2DM patients, rodent models do not usually present with increased AGEs40,41, which is consistent with the current study.

Both osteoporosis and T2DM are often clinically managed with drugs known to effect bone microarchitecture and bone matrix composition. For example, bisphosphonates are known to have positive effects on early implant fixation clinically3, as well as osseointegration42. Bisphosphonates have also been shown to alter bone matrix composition including crystallinity6–8. Based on our results, it is likely that both changes in the bone structure and composition likely contribute to the increased fixation strength reported in preclinical animal models43. Commonly prescribed T2DM medications, including metformin44 and insulin45, have also been reported to increase osseointegration in diabetic rats. However, whether these same treatments increase implant fixation strength or bone matrix composition is currently unknown.

The strengths of the present study include the well-validated measure of peri-implant bone microarchitecture17, the variance in the independent microarchitectural and matrix compositional parameters, and the detailed measurements of compartment-specific bone matrix compositional parameters. Further, while other studies have investigated bone composition and mechanical properties of bone surrounding an implant46–48, this study is the first to report that bone matrix composition significantly contributes to implant fixation strength. The limitations include the relatively limited variance in the bone matrix composition, which is likely due to the well-controlled process of bone matrix mineralization. Future work could attempt to increase the variance in matrix composition by deliberately altering the mineralization kinetics using a vitamin D or calcium deficient diet. While our strategy was to utilize the OXV and ZDF rat models to introduce variance in bone composition, it is worth noting that these models do not fully recapitulate the human diseases. Thus, while there is clear evidence that osteoporosis is deleterious to orthopedic implant fixation3,24, the effects of T2DM on orthopedic implant fixation are less clear and require further study. Nevertheless, the model systems are useful for understanding how bone microarchitecture and composition influence implant fixation strength.

Conclusion

This study demonstrates that implant fixation strength is largely determined by bone microarchitectural parameters such as bone implant contact, trabecular bone volume fraction, and cortical thickness. Bone matrix composition, in particular trabecular crystallinity, also independently contributed to implant fixation. Treatment strategies aimed at improving bone implant contact and peri-implant bone volume without compromising matrix composition should be prioritized.

Supplementary Material

tS1-S5

Acknowledgments

We thank Meghan Moran Ph.D., Rylan Martin B.S., and Rush MicroCT/Histology Core for experimental support. Research reported in this publication was supported by the Orthopedic Research and Education Foundation - Smith & Nephew Grant and by the National Institute of Arthritis and Musculoskeletal and Skin Diseases (NIAMS) of the National Institutes of Health (NIH) under award numbers T32AR073157, R01AR066562, R21AR075130, K01AR073923. Approximately 10% of the project costs were financed with federal money and the remainder with non-governmental sources. The content is solely the responsibility of the authors and does not necessarily represent the official views of the National Institutes of Health.

Footnotes

Disclosures: Authors state that they have no conflicts of interest.

References

  • 1.Zhan C, Kaczmarek R, Loyo-Berrios N, Sangl J, Bright RA. Incidence and Short-Term Outcomes of Primary and Revision Hip Replacement in the United States. J Bone Jt Surg. 2007;89(3):526–533. doi: 10.2106/JBJS.F.00952 [DOI] [PubMed] [Google Scholar]
  • 2.Bozic KJ, Kurtz SM, Lau E, et al. The epidemiology of revision total knee arthroplasty in the united states. In: Clinical Orthopaedics and Related Research. Vol 468. Springer New York LLC; 2010:45–51. doi: 10.1007/s11999-009-0945-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Anderson KD, Ko FC, Virdi AS, Sumner DR, Ross RD. Biomechanics of Implant Fixation in Osteoporotic Bone. Curr Osteoporos Rep. 2020. doi: 10.1007/s11914-020-00614-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 4.Ko FC, Meagher MJ, Mashiatulla M, Ross RD, Virdi AS, Sumner DR. Implant surface alters compartmental-specific contributions to fixation strength in rats. J Orthop Res. 2020;38(6):1208–1215. doi: 10.1002/jor.24561 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Taylor EA, Donnelly E, Yao X, et al. Sequential Treatment of Estrogen Deficient, Osteopenic Rats with Alendronate, Parathyroid Hormone (1–34), or Raloxifene Alters Cortical Bone Mineral and Matrix Composition. Calcif Tissue Int. 2020;106(3):303–314. doi: 10.1007/s00223-019-00634-w [DOI] [PubMed] [Google Scholar]
  • 6.Hofstetter B, Gamsjaeger S, Phipps RJ, et al. Effects of alendronate and risedronate on bone material properties in actively forming trabecular bone surfaces. J Bone Miner Res. 2012;27(5):995–1003. doi: 10.1002/jbmr.1572 [DOI] [PubMed] [Google Scholar]
  • 7.Gamsjaeger S, Hofstetter B, Zwettler E, et al. Effects of 3 years treatment with once-yearly zoledronic acid on the kinetics of bone matrix maturation in osteoporotic patients. Osteoporos Int. 2013;24(1):339–347. doi: 10.1007/s00198-012-2202-8 [DOI] [PubMed] [Google Scholar]
  • 8.Ross RD, Anderson K, Davison R, et al. Osteoporosis Treatments Affect Bone Matrix Maturation in a Rat Model of Induced Cortical Remodeling. JBMR Plus. 2020;4(4). doi: 10.1002/jbm4.10344 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 9.Kim G, Cole JH, Boskey AL, Baker SP, Van Der Meulen MCH. Reduced tissue-level stiffness and mineralization in osteoporotic cancellous bone. Calcif Tissue Int. 2014;95(2):125–131. doi: 10.1007/s00223-014-9873-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 10.Rokidi S, Paschalis EP, Klaushofer K, et al. Organic matrix quality discriminates between age- and BMD-matched fracturing versus non-fracturing post-menopausal women: A pilot study. Bone. 2019;127:207–214. doi: 10.1016/j.bone.2019.06.017 [DOI] [PubMed] [Google Scholar]
  • 11.Hammond MA, Gallant MA, Burr DB, Wallace JM. Nanoscale changes in collagen are reflected in physical and mechanical properties of bone at the microscale in diabetic rats. Bone. 2013;60:26–32. doi: 10.1016/j.bone.2013.11.015 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 12.Karim L, Moulton J, Van Vliet M, et al. Bone microarchitecture, biomechanical properties, and advanced glycation end-products in the proximal femur of adults with type 2 diabetes. Bone. 2018;114:32–39. doi: 10.1016/j.bone.2018.05.030 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Rokidi S, Andrade VFC, Borba V, et al. Bone tissue material composition is compromised in premenopausal women with Type 2 diabetes. Bone. 2020. doi: 10.1016/j.bone.2020.115634 [DOI] [PubMed] [Google Scholar]
  • 14.Jepsen KJ, Schlecht SH. Biomechanical mechanisms: Resolving the apparent conundrum of why individuals with type II diabetes show increased fracture incidence despite having normal BMD. J Bone Miner Res. 2014;29(4):784–786. doi: 10.1002/jbmr.2189 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Piccoli A, Cannata F, Strollo R, et al. Sclerostin Regulation, Microarchitecture, And Advanced Glycation End-Products In The Bone Of Elderly Women With Type 2 Diabetes. J Bone Miner Res. August 2020. doi: 10.1002/jbmr.4153 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Liu S, Virdi AS, Sena K, Hughes WF, Sumner DR. Bone turnover markers correlate with implant fixation in a rat model using LPS-doped particles to induced implant loosening. J Biomed Mater Res Part A. 2012;100A(4):918–928. doi: 10.1002/jbm.a.34029 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 17.Meagher MJ, Parwani RN, Virdi AS, Sumner DR. Optimizing a micro-computed tomography-based surrogate measurement of bone-implant contact. J Orthop Res. 2018;36(3):979–986. doi: 10.1002/jor.23716 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Rubin MR, Paschalis EP, Poundarik A, et al. Advanced glycation endproducts and bone material properties in type 1 diabetic mice. PLoS One. 2016;11(5). doi: 10.1371/journal.pone.0154700 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Berzins A, Sumner D. Implant pushout and pullout tests. In: An Y, Draughn R, eds. Mechanical Testing of Bone and the Bone-Implant Interface. CRC Press; 1999:463–476. [Google Scholar]
  • 20.Boskey AL, Coleman R. Critical reviews in oral biology & medicine: Aging and bone. J Dent Res. 2010;89(12):1333–1348. doi: 10.1177/0022034510377791 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Yerramshetty JS, Akkus O. The associations between mineral crystallinity and the mechanical properties of human cortical bone. Bone. 2008;42(3):476–482. doi: 10.1016/j.bone.2007.12.001 [DOI] [PubMed] [Google Scholar]
  • 22.Joffre T, Isaksson P, Procter P, Persson C. Trabecular deformations during screw pull-out: a micro-CT study of lapine bone. Biomech Model Mechanobiol. 2017;16(4):1349–1359. doi: 10.1007/s10237-017-0891-9 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Yousefzadeh N, Kashfi K, Jeddi S, Ghasemi A. Ovariectomized rat model of osteoporosis: A practical guide. EXCLI J. 2020;19:89–107. doi: 10.17179/excli2019-1990 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Ross RD, Hamilton JL, Wilson BM, Sumner DR, Virdi AS. Pharmacologic augmentation of implant fixation in osteopenic bone. Curr Osteoporos Rep. 2014;12(1):55–64. doi: 10.1007/s11914-013-0182-z [DOI] [PubMed] [Google Scholar]
  • 25.Pezzotti G, Rondinella A, Marin E, et al. Raman spectroscopic investigation on the molecular structure of apatite and collagen in osteoporotic cortical bone. J Mech Behav Biomed Mater. 2017;65:264–273. doi: 10.1016/j.jmbbm.2016.08.030 [DOI] [PubMed] [Google Scholar]
  • 26.Shah FA, Stoica A, Cardemil C, Palmquist A. Multiscale characterization of cortical bone composition, microstructure, and nanomechanical properties in experimentally induced osteoporosis. J Biomed Mater Res Part A. 2018;106(4):997–1007. doi: 10.1002/jbm.a.36294 [DOI] [PubMed] [Google Scholar]
  • 27.Paschalis EP, Fratzl P, Gamsjaeger S, et al. Aging Versus Postmenopausal Osteoporosis: Bone Composition and Maturation Kinetics at Actively-Forming Trabecular Surfaces of Female Subjects Aged 1 to 84 Years. J Bone Miner Res. 2016;31(2):347–357. doi: 10.1002/jbmr.2696 [DOI] [PubMed] [Google Scholar]
  • 28.Gourion-Arsiquaud S, Faibish D, Myers E, et al. Use of FTIR spectroscopic imaging to identify parameters associated with fragility fracture. J Bone Miner Res. 2009;24(9):1565–1571. doi: 10.1359/jbmr.090414 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 29.Fratzl-Zelman N, Roschger P, Misof BM, et al. Fragility fractures in men with idiopathic osteoporosis are associated with undermineralization of the bone matrix without evidence of increased bone turnover. Calcif Tissue Int. 2011;88(5):378–387. doi: 10.1007/s00223-011-9466-4 [DOI] [PubMed] [Google Scholar]
  • 30.Fajardo RJ, Karim L, Calley VI, Bouxsein ML. A review of rodent models of type 2 diabetic skeletal fragility. J Bone Miner Res. 2014;29(5):1025–1040. doi: 10.1002/jbmr.2210 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 31.Thrailkill KM, Lumpkin CK, Bunn RC, Kemp SF, Fowlkes JL. Is insulin an anabolic agent in bone? Dissecting the diabetic bone for clues. Am J Physiol - Endocrinol Metab. 2005;289(5 52–5). doi: 10.1152/ajpendo.00159.2005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Dennison EM, Syddall HE, Sayer AA, Craighead S, Phillips DIW, Cooper C. Type 2 diabetes mellitus is associated with increased axial bone density in men and women from the Hertfordshire Cohort Study: Evidence for an indirect effect of insulin resistance? Diabetologia. 2004;47(11):1963–1968. doi: 10.1007/s00125-004-1560-y [DOI] [PubMed] [Google Scholar]
  • 33.Hasegawa H, Ozawa S, Hashimoto K, Takeichi T, Ogawa T. Type 2 diabetes impairs implant osseointegration capacity in rats. Int J Oral Maxillofac Implants. 2008;23(2):237–246. [PubMed] [Google Scholar]
  • 34.Wang F, liang Song Y, hua Li D, et al. Type 2 diabetes mellitus impairs bone healing of dental implants in GK rats. Diabetes Res Clin Pract. 2010;88(1). doi: 10.1016/j.diabres.2010.01.017 [DOI] [PubMed] [Google Scholar]
  • 35.Wang B, Song Y, Wang F, et al. Effects of local infiltration of insulin around titanium implants in diabetic rats. Br J Oral Maxillofac Surg. 2010;49:225–229. doi: 10.1016/j.bjoms.2010.03.006 [DOI] [PubMed] [Google Scholar]
  • 36.Hashiguchi C, Kawamoto SI, Kasai T, Nishi Y, Nagaoka E. Influence of an antidiabetic drug on biomechanical and histological parameters around implants in type 2 diabetic rats. Implant Dent. 2014;23(3):264–269. doi: 10.1097/ID.0000000000000021 [DOI] [PubMed] [Google Scholar]
  • 37.Romero-Díaz C, Duarte-Montero D, Gutiérrez-Romero SA, Mendivil CO. Diabetes and Bone Fragility. Diabetes Ther. 2020. doi: 10.1007/s13300-020-00964-1 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Garnero P, Borel O, Gineyts E, et al. Extracellular post-translational modifications of collagen are major determinants of biomechanical properties of fetal bovine cortical bone. Bone. 2006;38(3):300–309. doi: 10.1016/j.bone.2005.09.014 [DOI] [PubMed] [Google Scholar]
  • 39.Vashishth D. The role of the collagen matrix in skeletal fragility. Curr Osteoporos Rep. 2007;5(2):62–66. doi: 10.1007/s11914-007-0004-2 [DOI] [PubMed] [Google Scholar]
  • 40.Creecy A, Uppuganti S, Unal M, Clay Bunn R, Voziyan P, Nyman JS. Low bone toughness in the TallyHO model of juvenile type 2 diabetes does not worsen with age. Bone. 2018;110:204–214. doi: 10.1016/j.bone.2018.02.005 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Hunt HB, Pearl JC, Diaz DR, King KB, Donnelly E. Bone Tissue Collagen Maturity and Mineral Content Increase With Sustained Hyperglycemia in the KK-Ay Murine Model of Type 2 Diabetes. J Bone Miner Res. 2018;33(5):921–929. doi: 10.1002/jbmr.3365 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 42.Guimarães MB, Antes TH, Dolacio MB, Pereira DD, Marquezan M. Does local delivery of bisphosphonates influence the osseointegration of titanium implants? A systematic review. Int J Oral Maxillofac Surg. 2017;46(11):1429–1436. doi: 10.1016/j.ijom.2017.04.014 [DOI] [PubMed] [Google Scholar]
  • 43.Ross RD, Hamilton JL, Wilson BM, Sumner DR, Virdi AS. Pharmacologic augmentation of implant fixation in osteopenic bone. Curr Osteoporos Rep. 2014;12(1):55–64. doi: 10.1007/s11914-013-0182-z [DOI] [PubMed] [Google Scholar]
  • 44.Sun R, Liang C, Sun Y, Xu Y, Geng W, Li J. Effects of metformin on the osteogenesis of alveolar BMSCs from diabetic patients and implant osseointegration in rats. Oral Dis. 2021. doi: 10.1111/odi.13808 [DOI] [PubMed] [Google Scholar]
  • 45.Wu YY, Yu T, Yang XY, et al. Vitamin D3 and insulin combined treatment promotes titanium implant osseointegration in diabetes mellitus rats. Bone. 2013;52(1):1–8. doi: 10.1016/j.bone.2012.09.005 [DOI] [PubMed] [Google Scholar]
  • 46.Vayron R, Barthel E, Mathieu V, Soffer E, Anagnostou F, Haiat G. Nanoindentation measurements of biomechanical properties in mature and newly formed bone tissue surrounding an implant. J Biomech Eng. 2012;134(2). doi: 10.1115/1.4005981 [DOI] [PubMed] [Google Scholar]
  • 47.Claffey N, Bashara H, O’Reilly P, Polyzois I. Evaluation of New Bone Formation and Osseointegration Around Subperiosteal Titanium Implants with Histometry and Nanoindentation. Int J Oral Maxillofac Implants. 2015;30(5):1004–1010. doi: 10.11607/jomi.3647 [DOI] [PubMed] [Google Scholar]
  • 48.Fraulob M, Pang S, Le Cann S, et al. Multimodal characterization of the bone-implant interface using Raman spectroscopy and nanoindentation. Med Eng Phys. 2020;84:60–67. doi: 10.1016/j.medengphy.2020.07.013 [DOI] [PubMed] [Google Scholar]

Associated Data

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

Supplementary Materials

tS1-S5
NIHMS1711888-supplement-tS1-S5.docx (33.1KB, docx)

RESOURCES