Michael D. Newton

Neer Award 2017: wear rates of 32-mm and 40-mm glenospheres in a reverse total shoulder arthroplasty wear simulation model

BACKGROUND Larger glenosphere diameters have been used recently to increase prosthesis stability and impingement-free range of motion in reverse total shoulder arthroplasty. The goal of this study was to evaluate the rate of polyethylene wear for 32-mm and 40-mm glenospheres. METHODS Glenospheres (32 mm and 40 mm, n = 6/group) and conventional polyethylene humeral liners underwent a 5-million cycle (MC) wear simulation protocol. Abduction-adduction and flexion-extension motion profiles were alternated every 250,000 cycles. At each interval, mass loss was determined and converted to volume loss and wear rate. At 0, 2.5 MC, and 5 MC, liners were imaged using micro-computed tomography to determine surface deviation. White light interferometry was performed on liners and glenospheres at 0 and 5 MC to quantify surface roughness. Wear particle morphology was characterized by environmental scanning electron microscopy. RESULTS Total volume loss was significantly higher in 40-mm liners from 1.5 MC onward (P < .05). Overall, volumetric wear rate was significantly higher in 40-mm liners compared with 32-mm glenospheres (81.7 ± 23.9 mm3/MC vs. 68.0 ± 18.9 mm3/MC; P < .001). However, micro-computed tomography surface deviation results demonstrated increased linear penetration on 32-mm glenospheres compared with 40-mm glenospheres (0.36 ± 0.03 µm vs. 0.28 ± 0.01 µm; P = .002). Surface roughness measurements showed no difference for liners; however, increased roughness was noted for 40-mm glenospheres at 5 MC compared with 32 mm (P < .05). CONCLUSION Larger glenospheres underwent significantly greater polyethylene volume loss and volumetric wear rates, whereas smaller glenospheres underwent greater polyethylene surface deviations. The enhanced stability provided by larger glenospheres must be weighed against the potential for increased polyethylene wear.

Error introduced by common reorientation algorithms in the assessment of rodent trabecular morphometry using micro-computed tomography: ERROR INTRODUCED BY IMAGE REORIENTATION

Quantitative analyses of bone using micro‐computed tomography (μCT) are routinely employed in preclinical research, and virtual image reorientation to a consistent reference frame is a common processing step. The purpose of this study was to quantify error introduced by common reorientation algorithms in μCT‐based characterization of bone. Mouse and rat tibial metaphyses underwent μCT scanning at a range of resolutions (6–30 μm). A trabecular volume‐of‐interest (VOI) was manually selected. Image stacks were analyzed without rotation, following 45° In‐Plane axial rotation, and following 45° Triplanar rotation. Interpolation was performed using Nearest‐Neighbor, Linear, and Cubic interpolations. Densitometric (bone volume fraction, tissue mineral density, bone mineral density) and morphometric variables (trabecular thickness, trabecular spacing, trabecular number, structural model index) were computed for each combination of voxel size, rotation, and interpolation. Significant reorientation error was measured in all parameters, and was exacerbated at higher voxel sizes, with relatively low error at 6 and 12 μm (max. reorientation error in BV/TV was 2.9% at 6 μm, 7.7% at 12 μm and 36.5% at 30 μm). Considering densitometric parameters, Linear and Cubic interpolations introduced significant error while Nearest‐Neighbor interpolation caused minimal error, and In‐Plane rotation caused greater error than Triplanar. Morphometric error was strongly and intricately dependent on the combination of rotation and interpolation employed. Reorientation error can be eliminated by avoiding reorientation altogether or by “de‐rotating” VOIs from reoriented images back to the original reference frame prior to analysis. When these are infeasible, reorientation error can be minimized through sufficiently high resolution scanning, careful selection of interpolation type, and consistent processing of all images.

Quantitative evaluation of retrieved reverse total shoulder arthroplasty liner surface deviation and volumetric wear: QUANTITATIVE EVALUATION OF HUMERAL LINER WEAR

Polyethylene wear is a known complication in total joint arthroplasty, however, in vivo wear rates in reverse total shoulder arthroplasty (RTSA) remain largely unknown. This study aimed to quantify volumetric and surface deviation changes in retrieved RTSA humeral liners using a novel micro‐computed tomography (μCT)‐based technique. After IRB‐approval, 32 humeral liners (single manufacturer and model) with term‐of‐service greater than 90 days were analyzed. Clinical demographics and surgical data were collected via chart review. Unworn liners were used as geometric controls. Retrieved and unworn liners underwent μCT scanning. Retrieved liner volumes were isolated, co‐registered to controls of matching geometry, and surface deviations of the articulation surface and rim were computed. Differences in total volume loss (TVL), volumetric wear rate (VWR), and surface deviation were reported. Semi‐quantitative grading evaluated rim damage presence and severity. Mean term‐of‐service for all liners was 2.07 ± 1.33 years (range: 0.30–4.73). Mean TVL and VWR were 181.3 ± 208.2 mm3 and 114.5 ± 160.3 mm3/year, respectively. Mean articulation and rim surface deviations were 0.084 ± 0.065 and 0.177 ± 0.159 mm, respectively. Articulation surface deviation was positively correlated to term‐of‐service. Rim damage was present on 63% of liners and correlated significantly to rim surface deviation. This study reports in vivo wear rates of retrieved RTSA implants. Our results demonstrate volumetric and articulation surface wear in select RTSA liners that is correlated to term‐of‐service. Calculation of in vivo wear rates can help bridge the gap between clinical outcomes and experimental models such as wear simulations and computational models.

Contrast-enhanced μCT of the intervertebral disc: A comparison of anionic and cationic contrast agents for biochemical and morphological characterization.

The objective of this study was to quantify and compare the contrast‐enhancing properties of the anionic contrast agent ioxaglate/Hexabrix, and cationic contrast agent CA4+ for biochemical and morphological characterization of the intervertebral disc (IVD) via μCT. Optimal contrast agent concentrations were determined by incubating rat lumbar IVDs in dilutions of Hexabrix‐320 (20%, 30%, 40%, and 50%) and CA4+ (10, 20, 30, and 40 mg I/ml). μCT imaging was performed at 70 kVp, 114 μA, and 250 ms integration time, 12 μm voxel size. The kinetics of contrast enhancement were quantified with cumulative incubations for 0.5, 1, 2, 12, 16, 20, and 24 h using both agents. Agreement in morphological quantification was assessed via serial scans of the same IVDs. Correlation of attenuation to glycosaminoglycan (GAG) content was determined by enzymatic digestion of IVDs, subsequent μCT imaging, and GAG quantification via dimethylmethylene blue assay. Forty percent Hexabrix and 30 mg I/ml CA4+ were chosen as optimal concentrations. Hexabrix enabled greater delineation of the IVD from surrounding tissues, and CA4+ had the lowest uptake in surrounding soft tissue. Twenty‐four hour incubation was sufficient for >99% equilibration of both agents. A high level of agreement was observed in the quantification of IVD volume (ICC = 0.951, r = 0.997) and height (ICC = 0.947, r = 0.991). Both agents exhibited strong linear correlations between μCT attenuation and GAG content (Hexabrix: r = −0.940; CA4+: r = 0.887). Both agents enable biochemical and morphological quantification of the IVD via contrast‐enhanced μCT and are effective tools for preclinical characterization.

Articular cartilage surface roughness as an imaging-based morphological indicator of osteoarthritis: A preliminary investigation of osteoarthritis initiative subjects

Current imaging‐based morphometric indicators of osteoarthritis (OA) using whole‐compartment mean cartilage thickness (MCT) and volume changes can be insensitive to mild degenerative changes of articular cartilage (AC) due to areas of adjacent thickening and thinning. The purpose of this preliminary study was to evaluate cartilage thickness‐based surface roughness as a morphometric indicator of OA. 3D magnetic resonance imaging (MRI) datasets were collected from osteoarthritis initiative (OAI) subjects with Kellgren–Lawrence (KL) OA grades of 0, 2, and 4 (n = 10/group). Femoral and tibial AC volumes were converted to two‐dimensional thickness maps, and MCT, arithmetic surface roughness (Sa), and anatomically normalized Sa (normSa) were calculated. Thickness maps enabled visualization of degenerative changes with increasing KL grade, including adjacent thinning and thickening on the femoral condyles. No significant differences were observed in MCT between KL grades. Sa was significantly higher in KL4 compared to KL0 and KL2 in the whole femur (KL0: 0.55 ± 0.10 mm, KL2: 0.53 ± 0.09 mm, KL4: 0.79 ± 0.18 mm), medial femoral condyle (KL0: 0.42 ± 0.07 mm, KL2: 0.48 ± 0.07 mm, KL4: 0.76 ± 0.22 mm), and medial tibial plateau (KL0: 0.42 ± 0.07 mm, KL2: 0.43 ± 0.09 mm, KL4: 0.68 ± 0.27 mm). normSa was significantly higher in KL4 compared to KL0 and KL2 in the whole femur (KL0: 0.22 ± 0.02, KL2: 0.22 ± 0.02, KL4: 0.30 ± 0.03), medial condyle (KL0: 0.17 ± 0.02, KL2: 0.20 ± 0.03, KL4: 0.29 ± 0.06), whole tibia (KL0: 0.34 ± 0.04, KL2: 0.33 ± 0.05, KL4: 0.48 ± 0.11) and medial plateau (KL0: 0.23 ± 0.03, KL2: 0.24 ± 0.04, KL4: 0.40 ± 0.10), and significantly higher in KL2 compared to KL0 in the medial femoral condyle. Surface roughness metrics were sensitive to degenerative morphologic changes, and may be useful in OA characterization and early diagnosis.

Cartilage Thickness and Surface Roughness Patterns in Healthy and Osteoarthritis Knees: Novel 3D Analysis of Subjects from the Osteoarthritis Initiative

Objectives: Three-dimensional (3D) magnetic resonance imaging (MRI) enables characterization of articular cartilage (AC) morphology. AC is traditionally analyzed using mean cartilage thickness (MCT), but OA can occur without drastic changes in MCT due to regions of both thickening and thinning, as shown in recent studies. Our group recently developed a method to assess 3D AC morphology in terms of both MCT and surface roughness (Sa) using mesh parameterization, an image processing technique that projects 3D data onto a 2D domain. The objective of this study was to apply this technique to characterize changes in MCT and Sa in subjects from the Osteoarthritis Initiative (OAI) with varying degrees of OA. Methods: Under institutional approval, image data was obtained from OAI. Inclusion criteria were availability of a baseline 3D double-echo steady state (DESS) MRI of the right knee and Kellgren-Lawrence (KL) score. Exclusion criteria were history of systemic testosterone, estrogen, GNRH, PTH, or bisphosphonate use, prior fracture, knee replacement, hyaluronic acid or steroid injections, and evidence of unreported knee injury or other anomaly on x-ray review. From the resulting pool, 10 subjects (5 men and 5 women) were randomly selected from each KL grade (0 - 4). Using our parameterization method, AC regions of interest were isolated from the MRI stacks and converted to 2D height maps (Figure 1). MCT and normalized surface roughness (Sa) were calculated for the whole femur, whole tibia and individual compartments. Femurs and tibias of KL0, KL2 and KL4 subjects have been analyzed. Analysis of patellae and remaining KL grades is ongoing. Results were compared between groups using t-tests with α = 0.05. Results: Representative KL0 and KL4 AC thickness maps are shown in Fig1A, B. Compared to KL0, KL4 exhibits thinning with adjacent thickening on the medial femoral condyle (MFC). There were no significant differences in MCT between KL grades in any femoral compartment. On the tibia, KL0 exhibits congruent AC with natively-thicker AC at the weight-bearing aspect. KL4 tibiae exhibit global thinning with a zone of severe thinning on the medial plateau (MP). Whole-tibia MCT of KL4 was significantly lower compared to KL2 and KL0 (Fig. 1D). In contrast to MCT, Sa was highly sensitive to compartment-dependent degeneration. In the whole femur, Sa was significantly higher in KL4 compared to both KL2 and KL0. On the MFC, Sa increased steadily with increasing KL grade (Fig1E), and the lateral condyle of KL4 exhibited higher Sa compared to KL0. No differences in Sa were observed on the trochlea. On the tibia, Sa was significantly elevated in KL4 compared to both KL0 and KL2 in all compartments (Fig1F). Conclusion: The presented technique enabled repeatable visualization, compartmental segmentation, and quantification of MCT and Sa of the whole joint. No differences in femoral MCT were found, which can be attributed to adjacent thickening and thinning. Femoral and tibial OA changes were detected more sensitively using Sa, with significant increases observed in the whole femur and both condyles and plateaux. Significant differences in Sa between KL0 and KL2 femurs indicate sensitivity of this technique to subtle changes in early OA. More sensitive characterization of compartmental and sub-compartmental morphologic changes associated with OA can increase our understanding of its progression and facilitate more sensitive early diagnosis.