Computed tomography analysis of the relationship between the coronoid and the radial head.

BACKGROUND: The coronoid process is an important stabilizer of the elbow, and its anatomy has been extensively studied. However, data documenting the relationship of the coronoid relative to the radial head (RH) are limited. The latter is a good landmark for the surgeon when débriding or reconstructing the coronoid. This imaging-based study quantified the anatomic relationship between the coronoid and the proximal radius and ulna. METHODS: We investigated 80 cadaveric upper extremities (18 paired elbows) by 3-dimensional digital analysis of computed tomography data. After construction of a standardized coordinate system, the relationships between the coronoid, the anterior-most point of the RH, the deepest point of the articular surface of the RH, the top of the lesser sigmoid notch, and the deepest point of the guiding ridge of the trochlear notch were analyzed. RESULTS: The mean height of the tip of the coronoid was 36 ± 4 mm (range, 26-43 mm). The mean height of the anterior-most point of the RH was 40 ± 4 mm (range, 28-47 mm). The mean distance between the tip of the coronoid and the anterior-most point of the RH was 4.5 ± 1 mm (range, 2-10 mm). For paired elbows, the heights of the tip of the coronoid and the anterior-most point of the RH were similar between sides. CONCLUSION: This study described the relationship between the coronoid and RH. This information should prove useful when reconstructing a coronoid from a medial approach in the case of an intact RH. The difference in radiographic height between the tip of the coronoid and anterior RH in the normal elbow averages 5 mm. However, when we account for the normal cartilage thickness of the RH and coronoid, a 3- to 6-mm difference in height would be seen at surgery depending on whether the cartilage of the coronoid process is intact or removed. The distance between the tip of the coronoid and the anterior-most point of the RH is similar to the size of shavers used when débriding osteophytes during arthroscopy.

Coronoid process reconstruction with a distal clavicle autograft: an in silico analysis of fitting accuracy.

BACKGROUND: The coronoid process plays a vital role in preserving elbow stability. In cases of acute or chronic deficiency of the coronoid process, reconstruction is warranted to restore stability and to avoid early joint degeneration. The distal clavicle might be a useful osteochondral autograft for coronoid reconstruction with low donor-site morbidity. This study evaluated the fitting accuracy of the distal clavicle as an autograft for coronoid process reconstruction. METHODS: One hundred upper-extremity computed tomography scans of 85 body donors were available for this study (mean age, 69 ± 17 years; 46 male and 39 female donors; 15 bilateral specimens). Standardized 40% transverse defects of the coronoid process were digitally created; the distal clavicles were digitally harvested and placed onto the defects by a best-fit technique in 2 different orientations using commercially available software: (1) with the superior aspect of the articular surface of the graft oriented toward the coronoid tip and (2) with the inferior aspect of the articular surface of the graft oriented toward the coronoid tip. The fitting accuracy of the grafts to the native coronoid process was evaluated from lateral to medial using custom code. RESULTS: Regardless of the orientation of the graft, the distal clavicle provided a good fit in the central portion of the coronoid process. In the lateral and medial aspects of the defect, however, the fitting accuracy of the graft declined significantly (P ≤ .044). No significant differences were observed between ipsilateral and contralateral grafts (P ≥ .199). The intrarater reliability was excellent. CONCLUSION: The results of this study suggest that a distal clavicle autograft may be suitable to replace a transverse defect of the coronoid process; however, it may not fully reconstruct the anteromedial and anterolateral aspects of the coronoid.

Revision shoulder arthroplasty: a systematic review and comparison of North American vs. European outcomes and complications.

BACKGROUND: Joint registries provide invaluable data on primary arthroplasties with revision as the endpoint; however, the revision outcomes are often excluded. Therefore, a PROSPERO registered review (CRD42015032531) of all revision studies in North America and Europe was conducted to evaluate demographics, etiologies and indications, implant manufacturer, and complications by geographic region. METHODS: The MEDLINE, EMBASE, and CENTRAL databases were searched for revision arthroplasty clinical studies with a minimum mean 24-month follow-up. There were no language exclusions. Articles published in German, French, and Italian were reviewed by research personnel proficient in each language. RESULTS: The mean age at revision was 66 ± 5 years (male = 759, female = 1123). The male-female ratio in North American and Europeans studies was 43:57 and 34:66, respectively. The most common etiology for primary surgery in both regions was osteoarthritis or glenoid arthrosis (38%). The most common revision indication overall was rotator cuff tear, deficiency, or arthropathy (26%). The most common implant type used in revisions was a reverse shoulder arthroplasty (54%). The complication rate for all revisions was 17%. There were a total of 465 complications, and of those, 74% lead to a reoperation. CONCLUSION: Generally, shoulder arthroplasties are designed to last 10-15 years; however, revisions are being performed at a mean 3.9 years from the primary procedure, based on the published studies included in this systematic review. Additionally, of the complications, a large number (74%) went on to a reoperation. Further insight into the reasons for early revisions and standardized reporting metrics and data collection on revisions is needed.

Type E2 glenoid bone loss orientation and management with augmented implants.

BACKGROUND: The purpose of this study was 2-fold: (1) to quantify type E2 bone loss orientation and its association with rotator cuff fatty infiltration and (2) to examine reverse baseplate designs used to manage type E2 glenoids. METHODS: Computed tomography scans of 40 patients with type E2 glenoids were examined for pathoanatomic features and erosion orientation. The rotator cuff fatty infiltration grade was compared with the erosion orientation angle. To compare reconstructive options in light of the pathoanatomic findings, virtual implantation of 4 glenoid baseplate designs (standard, half wedge, full wedge, and patient-matched) was conducted to determine the volume of bone removal for seating and impingement-free range of motion. RESULTS: The mean type E2 erosion orientation angle was 47° ± 17° from the 0° superoinferior glenoid axis, resulting in the average erosion being located in the posterosuperior quadrant directed toward the 10:30 clock-face position. The type E2 neoglenoid, on average, involved 67% of the total glenoid surface (total surface area, 946 ± 209 mm2; neoglenoid surface area, 636 ± 247 mm2). The patient-matched baseplate design resulted in significantly (P ≤ .01) less bone removal (200 ± 297 mm3) for implantation, followed by the full-wedge design (1228 ± 753 mm3), half-wedge design (1763 ± 969 mm3), and standard (non-augmented) design (4009 ± 1210 mm3). We noted a marked difference in erosion orientation toward a more superior direction as the subscapularis fatty infiltration grade increased from grade 3 to grade 4 (P < .001). CONCLUSION: The average type E2 erosion orientation was directed toward the 10:30 clock-face position in the posterosuperior glenoid quadrant. This orientation resulted in the patient-matched glenoid augmentation requiring the least amount of bone removal for seating, followed by the full-wedge, half-wedge, and standard designs. Implant selection also substantially affected computationally derived range of motion in external rotation, flexion, extension, and adduction.

Density distribution of the type E2 glenoid in cuff tear arthropathy.

BACKGROUND: Little is known about the cortical-like and cancellous bone density variations in superiorly eroded glenoids due to cuff tear arthropathy. The purpose of this study was to analyze regional bone density in type E2 glenoids. METHODS: Clinical shoulder computed tomography scans were obtained from 32 patients with a type E2 superior erosion (10 men and 22 women; mean age, 73 years). Measurement regions were organized into quadrants (superior, inferior, anterior, and posterior) and depth regions. The depth regions were incremented by 2 mm from 0 to 10 mm. A repeated-measures multiple analysis of variance was performed to assess differences and interactions between mean densities (cortical-like and cancellous bone) in each depth, in each quadrant, and between sexes. RESULTS: The lowest cancellous bone density was found in the inferior glenoid quadrant compared with all other quadrants (307 ± 50 Hounsfield units [HU], P < .001). At the glenoid surface, the superior quadrant contained the highest mean density for cortical-like bone (895 ± 97 HU); this differed significantly from the posterior, anterior, and inferior quadrants (P ≤ .033). As for depth of measurement, cortical-like bone was most dense at the glenoid surface (0-2 mm, 892 ± 91 HU), and density decreased significantly at depths greater than 2 mm (P ≤ .019). CONCLUSION: In patients with type E2 glenoids due to cuff tear arthropathy, the densest bone was found in the superior quadrant in the area of erosion. The inferior quadrant, which tends to be unloaded as the humeral head migrates superiorly, had the lowest density bone. In addition, the best-quality bone was located at the glenoid surface as compared with deeper in the vault.

The Walch type B humerus: glenoid retroversion is associated with torsional differences in the humerus.

BACKGROUND: The Walch type B glenoid has the hallmark features of retroversion, joint subluxation, and bony erosion. Although the type B glenoid has been well described, the morphology of the corresponding type B humerus is poorly understood. As such, the aim of this imaging-based anthropometric study was to investigate humeral torsion in Walch type B shoulders. METHODS: Three-dimensional models of the full-length humerus were generated from computed tomography data for the Walch type B group (n = 59) and for a control group of normal nonarthritic shoulders (n = 59). An anatomic humeral head-neck plane was created and used to determine humeral torsion relative to the epicondylar axis. Measurements were repeated, and intraclass correlation coefficients were calculated. RESULTS: The type B humeri had significantly (P < .001) less retrotorsion (14° ± 9°) than the control group (36° ± 12°) relative to the epicondylar axis. Male and female individuals within the control group showed statistically significant differences in humeral torsion (P = .043), which were not found in the type B group. Inter-rater reliability showed excellent agreement for humeral torsion (intraclass correlation coefficient, 0.962). A subgroup analysis between Walch type B2 and B3 shoulders showed no significant differences in any of the humeral or glenoid parameters. CONCLUSION: The Walch type B humerus has significantly less retrotorsion than non-osteoarthritic shoulders. At present, it is unknown whether the altered humeral retrotorsion is a cause or effect of the type B glenoid. In addition, it is unknown whether surgeons should be reconstructing type B2 humeral component version to pathologic torsion or to nonpathologic population means to optimize arthroplasty survivorship.

The shape match of the olecranon tip for reconstruction of the coronoid process: influence of side and osteotomy angle.

BACKGROUND: The integrity of the coronoid process is critical to maintaining elbow stability. Unreconstructible fractures and chronic coronoid deficiency are challenging clinical problems with no clear solution. The purposes of this study were to investigate the shape match of the ipsilateral and contralateral olecranon tips as graft options and to determine the influence of the osteotomy angle on fitment. METHODS: Nineteen paired cadaveric elbow joints were investigated by 3-dimensional digital analysis of computed tomography DICOM (Digital Imaging and Communications in Medicine) data. After construction of an ulnar coordinate system, the ipsilateral and contralateral olecranon tips were digitally harvested at 10°, 20°, 30°, 40°, 50°, and 60° osteotomy angles. In an overlay analysis, we compared the shape match of the ipsilateral and contralateral grafts and the different angles. RESULTS: The ipsilateral grafts showed an average mismatch of 1.8 mm (standard deviation, 1.38 mm), whereas the contralateral grafts had a significantly lower (P < .001) mean mismatch of 1.3 mm (standard deviation, 0.95 mm). The 50° osteotomy plane showed the best shape match in comparison with the native coronoid-in both the ipsilateral and contralateral grafts. Evaluation of the intraclass correlation coefficient was calculated at r = 0.944, showing high repeatability of the measurements. CONCLUSIONS: The contralateral olecranon tip graft showed significantly better shape matching to the native coronoid than the ipsilateral olecranon graft. Specifically, the contralateral graft more closely matched the biomechanically critical anteromedial coronoid facet. Finally, both the contralateral and ipsilateral olecranon grafts had better shape matching with the native coronoid when osteotomy was performed at higher angles, specifically 50°.

Methods for Post Hoc Quantitative Computed Tomography Bone Density Calibration: Phantom-Only and Regression.

Quantitative computed tomography (qCT) relies on calibrated bone mineral density data. If a calibration phantom is absent from the CT scan, post hoc calibration becomes necessary. Scanning a calibration phantom after-the-fact and applying that calibration to uncalibrated scans has been used previously. Alternatively, the estimated density is known to vary with CT settings, suggesting that it may be possible to predict the calibration terms using CT settings. This study compares a novel CT setting regression method for post hoc calibration to standard and post hoc phantom-only calibrations. Five cadaveric upper limbs were scanned at 11 combinations of peak tube voltage and current (80-140 kV and 100-300 mA) with two calibration phantoms. Density calibrations were performed for the cadaver scans, and scans of the phantoms alone. Stepwise linear regression determined if the calibration equation terms were predictable using peak tube voltage and current. Peak tube voltage, but not current, was significantly correlated with regression calibration terms. Calibration equation slope was significantly related to the type of phantom (p < 0.001), calibration method (p = 0.026), and peak tube voltage (p < 0.001), but not current (p = 1.000). The calibration equation vertical intercept was significantly related to the type of phantom (p < 0.001), and peak tube voltage (p = 0.006), but not calibration method (p = 0.682), or current (p = 0.822). Accordingly, regression can correlate peak tube voltage with density calibration terms. Suggesting that, while standard qCT calibration is preferable, regression calibration may be an acceptable post hoc method when necessary.

Characterization of the Walch B3 glenoid in primary osteoarthritis.

BACKGROUND: The type B3 glenoid is an addition to the Walch classification. A potential etiologic theory is that it is a progression of the B2. It is characterized by uniconcavity, absent paleoglenoid, medialization, retroversion, and subluxation. The purpose of this study was to describe the morphology of B3 glenoids. METHODS: Fifty-two patients with B3 glenoids underwent 3-dimensional analysis of computed tomography data. Glenoid measurements (retroversion, inclination, medialization) and humeral head subluxation according to the scapular and glenoid planes were determined. The measured variables were compared between male and female patients. RESULTS: The mean B3 retroversion, inclination, and medialization were 24° ± 7°, 8° ± 6° superior, and 14 ± 4 mm, respectively. The mean posterior subluxation was 80% ± 8% and 54% ± 6% according to the scapular and glenoid planes, respectively. There were no differences in B3 characteristics between sexes (P > .05). A significant correlation existed between glenoid retroversion and humeral head subluxation relative to the scapular plane, with every 1° increase in retroversion translating to a 1% increase in subluxation (P < .001). In contrast, when referencing the glenoid plane, the humeral head remained concentric to the erosion. CONCLUSIONS: The B3 is uniconcave and retroverted. As glenoid retroversion increases, posterior humeral head subluxation significantly increases as referenced to the scapular plane; however, when referenced to the glenoid plane, the head remains concentric to the erosion. This appearance of "concentricity" is acquired secondary to the wear pattern, creating a uniconcave glenoid. Therefore, surgeons should be aware that the visualized concentricity is a product of the erosion pattern and thus may conceal a greater amount of subluxation potential.

Premorbid retroversion is significantly greater in type B2 glenoids.

BACKGROUND: Posterior glenoid erosion is thought to be initiated by humeral head subluxation. However, it is unknown whether subluxation is entirely caused by soft-tissue instability and unbalanced muscle activity or whether osseous morphology is a contributing factor. We hypothesized that patients with posterior erosion may exhibit premorbid glenoid morphology that is inherently retroverted and inferiorly inclined compared with age-matched normal glenoids. METHODS: This study examined 80 scapulae, evenly distributed between 2 groups: osteoarthritic with type B2 glenoids and age-matched normal glenoids. From 3-dimensional computed tomography reconstructions, version and inclination were measured from the anterior paleoglenoid region of the B2 glenoids, which is representative of the premorbid glenoid, and compared with measurements obtained from similar regions in the normal cohort. RESULTS: The anterior paleoglenoid region of B2 glenoids was significantly (P < .001) more retroverted (-14° ± 6°) compared with similar regions in nonarthritic normal glenoids (-5° ± 5°). There were no significant differences (P = .166) in the glenoid inclination angle between type B2 glenoids (0° ± 6°) and nonarthritic normal glenoids (2° ± 5°). Negative values represent retroverted and inferiorly inclined glenoids. DISCUSSION: Understanding premorbid glenoid morphologic variations may provide insight into the pathoanatomy of humeral head subluxation, osteoarthritis, and posterior glenoid erosion. The results of this study indicate that patients with type B2 osteoarthritic glenoids have significantly greater premorbid glenoid retroversion compared with nonarthritic normal glenoids, suggesting that this premorbid morphologic variation may be one contributing factor to posterior erosion.

A comparison of normal and osteoarthritic humeral head size and morphology.

BACKGROUND: The purpose of this study was to evaluate and to compare the size and morphologic patterns among normal and osteoarthritic (OA) humeral heads. METHODS: This comparative anatomic imaging study evaluated 150 humeral heads that were separated into 3 cohorts: normal, OA with symmetric glenoid erosion, and OA with asymmetric (type B2) glenoid erosion. Three-dimensional models were created of the humeral head from computed tomography data, and point coordinates were extracted for evaluation. Parameters measured were diameter (sphere fit and circle fit), chord distance (superoinferior and anteroposterior), and humeral head height. RESULTS: The sphere-fit diameter of the humeral head for the entire OA cohort (100 patients; mean diameter, 59 ± 9 mm) was significantly greater (P < .001) than that of the normal cohort (50 patients; mean diameter, 49 ± 5 mm). Similarly, the humeral head circle-fit diameters in the superoinferior and anteroposterior planes were significantly greater (P < .001) in the combined OA cohorts (59 ± 9 mm and 56 ± 10 mm, respectively) compared with the normal cohort (51 ± 5 mm and 47 ± 5 mm, respectively). However, there were no significant differences (P ≥ .099) between the symmetric and asymmetric OA cohorts in sphere-fit or circle-fit diameters. The mean values of humeral head heights were not significantly different (P = .382) between cohorts, 19 ± 2 mm, 18 ± 2 mm, and 18 ± 2 mm for the normal, symmetric, and asymmetric cohorts, respectively. DISCUSSION: Although OA humeral head morphology varies significantly from normal, it does not vary as a function of the Walch classification between symmetric and asymmetric glenoids. Understanding of the morphologic variability of the pathologic humeral head may provide insight into the pathoanatomy of osteoarthritis and the development of various erosion patterns.

Augmented glenoid component designs for type B2 erosions: a computational comparison by volume of bone removal and quality of remaining bone.

BACKGROUND: The purpose of this computational modeling study was to compare the volume of glenoid bone removal required to implant 3 augmented component designs for management of B2 erosions. In addition, we assessed bone quality of the supporting bone directly beneath the implants by measuring bone density and porosity. METHODS: Three augmented component designs­full-wedge, posterior-wedge, and posterior-step­were studied by virtual implantation in a cohort of 16 patients with B2 glenoids. B2 retroversion was corrected to 0° and 10°. The outcome variables were the volume of glenoid bone removal required for implantation and the density and porosity of the bone immediately beneath the implant. RESULTS: Implant design had a significant effect on the volume of bone removal (P < .001). When correcting to 0°, the posterior-wedge implant removed less bone than the posterior-step (P < .001) and the full-wedge (P = .004). At 10° retroversion, the posterior-wedge removed less bone (P = .029) than the posterior-step but was no different than the full-wedge (P = .143). The residual glenoid bone density with the posterior-wedge was significantly greater than with the posterior-step (P = .048), with no other significant differences (P > .05). Residual glenoid bone porosity was not significantly different between implants (P > .262). CONCLUSIONS: Augmented components can provide a bone-preserving option for B2 glenoid management. Substantial variations in the volume of bone removal and the quality of the remaining glenoid bone were found between 3 different designs of augmented implants. Simulations with the posterior-wedge implant resulted in substantially less glenoid bone removal, with the remaining supporting bone being of better quality.

Quantification of the position, orientation, and surface area of bone loss in type B2 glenoids.

BACKGROUND: The purpose of this computed tomography-based study was to quantify erosions in B2 glenoids. We hypothesized that erosions do not occur symmetrically and that they have a predictable orientation. METHODS: This study evaluated 55 type B2 glenoid cases. Computed tomography data were used to create three-dimensional reconstructions, and point coordinates were extracted from each reconstruction for morphologic analysis of the osteoarthritic glenoid and humerus. RESULTS: There was a significant difference between the mean orientation angle (28° ± 11°) of the posterior glenoid line of erosion and the superoinferior axis (P < .001), which resulted in the average erosion being directed toward the 8-o'clock position. On average, the erosion started 1.6 ± 3.4 mm posterior to the glenoid center point. In 35% of B2 cases, the line of erosion was curved. The mean surface area was 763 ± 296 mm(2) for the neoglenoid and 957 ± 276 mm(2) for the paleoglenoid, indicating that the neoglenoid occupied 44% ± 12% of the total glenoid area. In this cohort, the mean radius of the humeral head neoarticulation was 32 ± 6 mm, the neoglenoid radius was 37 ± 8 mm, and the paleoglenoid radius was 34 ± 7 mm. The radius of the humeral head was significantly less than that of the neoglenoid (P < .001) and the paleoglenoid (P = .009). In addition, the radius of the neoglenoid was significantly greater than the radius of the paleoglenoid (P = .012). DISCUSSION: Type B2 glenoids have a predictable wear pattern, which is not axisymmetric to the glenoid superoinferior axis. In addition, the identified anatomic characteristics of B2 erosions will aid surgeons in the operative management of bone loss and may assist manufacturers in the design of augmented components.

Regional bone density variations in osteoarthritic glenoids: a comparison of symmetric to asymmetric (type B2) erosion patterns.

BACKGROUND: Accurate characterization of regional variations in bone density in symmetric and asymmetric (B2) glenoid erosion patterns can assist with surgical planning, reaming, and component implantation. The purpose of this study was to characterize regional bone density and porosity in symmetric and asymmetric (B2) osteoarthritic glenoids. METHODS: Symmetric (n = 25) and asymmetric (B2) (n = 25) erosion patterns were compared by computed tomography-based imaging software. An orthogonal coordinate system separated each glenoid into quadrants. In addition, a linear best-fit line defined the line-of-erosion between the paleoglenoid and neoglenoid in the asymmetric cohort. All glenoids were further divided into volumes at depths of 0 to 2.5 mm and 2.5 to 5 mm. Average bone density was measured in Hounsfield units. Bone voids or cysts were included to quantify regional porosity as the fraction of void volume to total glenoid volume. RESULTS: For the symmetric cohort, there were no significant differences in bone density between quadrants at either depth (P ≥ .089). For the asymmetric cohort, bone density was significantly higher in the posterior quadrants compared with the anterior quadrants (P < .001), especially posteroinferiorly (P ≤ .007) at both depths. In addition, the neoglenoid had significantly higher density and lower void fraction compared with the paleoglenoid (P < .001). There were also significant differences in void fraction between quadrants for both cohorts, at both depths (P ≤ .004). CONCLUSIONS: This study demonstrates that osteoarthritic glenoids with symmetric erosion have uniform subarticular bone density. However, asymmetric (B2) erosion patterns have potentially important regional variations in bone density and porosity, with the densest bone with the least porosity found posteroinferiorly or in the neoglenoid region.

Comparison of clinical-CT segmentation techniques for measuring subchondral bone cyst volume in glenohumeral osteoarthritis.

PURPOSE: This study aimed to assess the accuracy and reproducibility of four common segmentation techniques measuring subchondral bone cyst volume in clinical-CT scans of glenohumeral OA patients. METHODS: Ten humeral head osteotomies collected from cystic OA patients, having undergone total shoulder arthroplasty, were scanned within a micro-CT scanner, and corresponding preoperative clinical-CT scans were gathered. Cyst volumes were measured manually in micro-CT and served as a reference standard (n = 13). Respective cyst volumes were measured on the clinical-CT scans by two independent graders using four segmentation techniques: Qualitative, Edge Detection, Region Growing, and Thresholding. Cyst volume measured in micro-CT was compared to the different clinical-CT techniques using linear regression and Bland-Altman analysis. Reproducibility of each technique was assessed using intraclass correlation coefficient (ICC). RESULTS: Each technique outputted lower volumes on average than the reference standard (-0.24 to -3.99 mm3). All linear regression slopes and intercepts were not significantly different than 1 and 0, respectively (p < 0.05). Cyst volumes measured using Qualitative and Edge Detection techniques had the highest overall agreement with reference micro-CT volumes (mean discrepancy: 0.24, 0.92 mm3). These techniques showed good to excellent reproducibility between graders. CONCLUSIONS: Qualitative and Edge Detection techniques were found to accurately and reproducibly measure subchondral cyst volume in clinical-CT. These findings provide evidence that clinical-CT may accurately gauge glenohumeral cystic presence, which may be useful for disease monitoring and preoperative planning. LEVEL OF EVIDENCE: Retrospective cohort Level 3 study.

High performance multi-platform computing for large-scale image-based finite element modeling of bone.

BACKGROUND: Image-based finite element (FE) modeling of bone is a non-invasive method to estimate bone stiffness and strength. High-resolution imaging data as input allows for inclusion of bone microarchitecture but results in large amounts of data unsuitable for traditional FE solvers. Bone-specific mesh-free solvers have been developed over the past 20 years to improve on memory efficiency in simulated bone loading applications. The objective of this study was to provide linear performance benchmarking for a bone-specific, mesh-free solver (FAIM) using µCT and HR-pQCT image data on Mac, Linux, and Windows operating systems using both single- and multi-thread CPU and GPU processing. METHODS: The focus is on the linear gradient-descent solver using standardized uniaxial loading of bone models from µCT, and first- and second-generation HR-pQCT scans of the radius and tibia. Convergence, speedup, memory, and batch performance tests were completed using CPUs and GPUs on three laboratory-based systems with Windows, Linux, and Mac operating systems. RESULTS: Although varying by system and model size, time-per-iteration was as low as 0.03 s when an HR-pQCT-based radius model (6.45 million DOF) was solved with 3 GPUs. Strong scaling was achieved with GPU and CPU parallel processing, with strong parallel efficiencies when models were solved using 3 GPUs or ≤ 10 CPU threads. Errors in force, strain energy density, and Von Mises stress were as low as 0.1% when a convergence tolerance of 10-3 or smaller was used. CONCLUSION: The results of this study indicate that to maximize computational efficiency and minimize model solution times using FAIM software under the standardized tested conditions using µCT, XCT1 and XCT2 HR-pQCT image data, convergence tolerance set to 10-4, and 10 threads or 2 GPUs are sufficient for efficient solution times. Less strict convergence tolerances will improve solution times but will introduce more error in the outcome measures.

Experimental DVC validation of heterogeneous micro finite element models applied to subchondral trabecular bone of the humeral head.

Subchondral trabecular bone (STB) undergoes adaptive changes during osteoarthritic (OA) disease progression. These changes alter both the mineralization patterns and structure of bone and may contribute to variations in the mechanical properties. Similarly, when images are downsampled - as is often performed in micro finite element model (microFEM) generation - the morphological and mineralization patterns may further alter the mechanical properties due to partial volume effects. MicroFEMs accounting for material heterogeneity can account for these tissue variations, but no studies have validated these with robust full-field testing methods. As such, this study compared homogeneous and heterogeneous microFEMs to experimentally loaded trabecular bone cores from the humeral head combined with digital volume correlation (DVC). These microFEMs were used to compare apparent mechanical properties between normal and OA STB. Morphological and mineralization patterns between groups were also compared. There were no significant differences in tissue or bone mineral density between groups. The only significant differences in morphometric parameters were in trabecular thickness between groups. There were no significant differences in linear regression parameters between normal and OA STB apparent mechanical properties estimated using heterogeneous microFEMs with an element-wise bilinear elastic-plastic constitutive model. Clinical significance: Validated heterogeneous microFEMs applied to STB of the humeral head have the potential to significantly improve our understanding of mechanical variations in the bone that occur during OA progression.

Independent changes in bone mineralized and marrow soft tissues following acute knee injury require dual-energy or high-resolution computed tomography for accurate assessment of bone mineral density and stiffness.

Musculoskeletal injuries often induce local accumulation of blood and/or fluid within the bone marrow, which is detected on medical imaging as edema-like marrow signal intensities (EMSI). In addition to its biological effects on post-injury recovery, the displacement of low-attenuating, largely adipocytic marrow by EMSI may introduce errors into quantitative computed tomography (QCT) measurements of bone mineral density (vBMD) and resulting bone stiffness estimates from image-based finite element (FE) analysis. We aimed to investigate the impact of post-injury changes in marrow soft tissue composition on CT-based bone measurements by applying CT imaging at multiple spatial resolutions. To do so, dual energy QCT (DECT) material decomposition was used to detect EMSI in the tibiae of nineteen participants with a recent anterior cruciate ligament tear. We then measured bone density and FE-based apparent modulus within the EMSI region and in a matched volume in the uninjured contralateral knee. Three measurement methods were applied: 1.) standard, QCT density calibration and density-based FEM; 2.) a DECT density calibration that provides density measurements adjusted for marrow soft tissues; and 3.) high-resolution peripheral QCT (HR-pQCT) density and microFE analyses. When measured using standard, single-energy QCT, vBMD and apparent modulus were elevated in the EMSI compared to the contralateral. After adjusting for marrow soft tissue composition using DECT, these measurements were no longer different between the two regions. By allowing for high-resolution, localized density analysis, HR-pQCT indicated that trabecular tissue mineral density was 9 mgHA/cm3 lower, while density of marrow soft tissues was 18 mgHA/cm3 higher, in the EMSI than the contralateral region, suggesting that EMSI have opposite effects on the measured density of trabecular bone and the underlying soft marrow. Thus, after an acute injury, altered composition of marrow soft tissues may artificially inflate overall measurements of bone density and apparent modulus obtained using standard QCT. This can be corrected by accounting for marrow soft tissue attenuation, either by using DECT-based density calibration or HR-pQCT microFE and measurements of local density of trabeculae.

Proximal Tibia Bone Stiffness and Strength in HR-pQCT- and QCT-Based Finite Element Models.

Injury to the ACL significantly increases the risk of developing post-traumatic osteoarthritis. Following injury, considerable focus is placed on visualizing soft tissue changes using MRI, but there is less emphasis on the alterations to the underlying bone. It has recently been shown using high-resolution peripheral quantitative computed tomography (HR-pQCT) that significant reductions in bone quality occur in the knee post ACL-injury. Despite the ability of HR-pQCT to show these changes, the availability of scanners and computational time requirements required to assess bone stiffness and strength with HR-pQCT limit its widespread clinical use. As such, the objective of this study was to determine if clinical quantitative CT (QCT) finite element models (QCT-FEMs) can accurately replicate HR-pQCT FEM proximal tibial stiffness and strength. From FEMs of 30 participants who underwent both QCT and HR-pQCT bilateral imaging, QCT-FEMs were strongly correlated with HR-pQCT FEM stiffness (R2 = 0.79). When QCT-FEM bone strength was estimated using the reaction force at 1% apparent strain, strong correlations were observed (R2 = 0.81), with no bias between HR-pQCT FEMs and non-linear QCT-FEMs. These results indicate that QCT-FEMs can accurately replicate HR-pQCT FEM stiffness and strength in the proximal tibia. Although these models are not able to replicate the trabecular structure or tissue-level strains, they require significantly reduced computational time and use widely available clinical-CT images as input, which make them an attractive choice to monitor bone density, stiffness and strength alterations, such as those that occur post ACL-injury.

Morphological and Apparent-Level Stiffness Variations Between Normal and Osteoarthritic Bone in the Humeral Head.

Osteoarthritis (OA) is characterized by morphological changes that alter bone structure and mechanical properties. This study compared bone morphometric parameters and apparent modulus between humeral heads excised from end-stage OA patients undergoing total shoulder arthroplasty (n = 28) and non-pathologic normal cadavers (n = 28). Morphometric parameters were determined in central cores, with regional variations compared in four medial to lateral regions. Linear regression compared apparent modulus, morphometric parameters, and age. Micro finite element models estimated trabecular apparent modulus and derived density-modulus relationships. Significant differences were found for bone volume fraction (p < 0.001) and trabecular thickness (p < 0.001) in the most medial regions. No significant differences occurred between morphometric parameters and apparent modulus or age, except in slope between groups for apparent modulus versus trabecular number (p = 0.021), and in intercept for trabecular thickness versus age (p = 0.040). Significant differences occurred in both slope and intercept between density-modulus regression fits for each group (p ≤ 0.001). The normal group showed high correlations in the power-fit (r2 = 0.87), with a lower correlation (r2 = 0.61) and a more linear relationship, in the OA group. This study suggests that alterations in structure and apparent modulus persist mainly in subchondral regions of end-stage OA bone. As such, if pathologic regions are removed during joint replacement, computational models that utilize modeling parameters from non-pathologic normal bone may be applied to end-stage OA bone. An improved understanding of humeral trabecular bone variations has potential to improve the surgical management of end-stage OA patients. © 2019 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 38:503-509, 2020.