A semi-automated quantitative CT method for measuring rotator cuff muscle degeneration in shoulders with primary osteoarthritis

Impact of Deltoid Computer Tomography Image Data on the Accuracy of Machine Learning Predictions of Clinical Outcomes after Anatomic and Reverse Total Shoulder Arthroplasty

Background: Despite the importance of the deltoid to shoulder biomechanics, very few studies have quantified the three-dimensional shape, size, or quality of the deltoid muscle, and no studies have correlated these measurements to clinical outcomes after anatomic (aTSA) and/or reverse (rTSA) total shoulder arthroplasty in any statistically/scientifically relevant manner. Methods: Preoperative computer tomography (CT) images from 1057 patients (585 female, 469 male; 799 primary rTSA and 258 primary aTSA) of a single platform shoulder arthroplasty prosthesis (Equinoxe; Exactech, Inc., Gainesville, FL) were analyzed in this study. A machine learning (ML) framework was used to segment the deltoid muscle for 1057 patients and quantify 15 different muscle characteristics, including volumetric (size, shape, etc.) and intensity-based Hounsfield (HU) measurements. These deltoid measurements were correlated to postoperative clinical outcomes and utilized as inputs to train/test ML algorithms used to predict postoperative outcomes at multiple postoperative timepoints (1 year, 2-3 years, and 3-5 years) for aTSA and rTSA. Results: Numerous deltoid muscle measurements were demonstrated to significantly vary with age, gender, prosthesis type, and CT image kernel; notably, normalized deltoid volume and deltoid fatty infiltration were demonstrated to be relevant to preoperative and postoperative clinical outcomes after aTSA and rTSA. Incorporating deltoid image data into the ML models improved clinical outcome prediction accuracy relative to ML algorithms without image data, particularly for the prediction of abduction and forward elevation after aTSA and rTSA. Analyzing ML feature importance facilitated rank-ordering of the deltoid image measurements relevant to aTSA and rTSA clinical outcomes. Specifically, we identified that deltoid shape flatness, normalized deltoid volume, deltoid voxel skewness, and deltoid shape sphericity were the most predictive image-based features used to predict clinical outcomes after aTSA and rTSA. Many of these deltoid measurements were found to be more predictive of aTSA and rTSA postoperative outcomes than patient demographic data, comorbidity data, and diagnosis data. Conclusions: While future work is required to further refine the ML models, which include additional shoulder muscles, like the rotator cuff, our results show promise that the developed ML framework can be used to evolve traditional CT-based preoperative planning software into an evidence-based ML clinical decision support tool.

Management of irreparable subscapularis tears: Current concepts

Primary repair of acute subscapularis (SSC) tears provides excellent results, but tendon retraction, muscle atrophy, fatty infiltration, and humeral head migration may render a more chronic tear irreparable. These irreparable SSC tears present a diagnostic and treatment challenge for orthopaedic surgeons. Careful physical examination and imaging evaluation can help to distinguish those with reparable versus irreparable tears, but they are still not very reliable due to the methodological limitations of current evidence. Therefore, future research using 3D and quantitative measurement techniques is necessary to better predict the irreparability of the SSC. When conservative treatment of an irreparable SSC tear fails, reversed shoulder arthroplasty has been established as the preferred treatment option for older, low-demand patients with arthropathy, providing reliable improvements in pain and function. In younger patients without significant arthropathy, musculotendinous transfers are the treatment of choice. The pectoralis major transfer is historically the most frequently performed procedure and provides improved range of motion and pain relief, but fails to adequately restore strength and shoulder function. The latissimus dorsi transfer has gained increased interest over the last few years due to its biomechanical superiority, and early clinical studies suggest improved outcomes as well. More recently, anterior capsular reconstruction has been proposed as an alternative to musculotendinous transfers, but clinical data are completely lacking. Future high-quality randomised controlled trials are necessary to reliably compare the different musculotendinous transfers and anterior capsular reconstruction.

The Modified Assessment Tool Based on Scapular Y-View for Global Fatty Infiltration in the Supraspinatus Muscle: Correlation, Predictive Performance, and Reliability Analyses

BACKGROUND

The accurate evaluation of rotator cuff (RC) fatty degeneration after tears is critical for appropriate surgical decision making and prognosis. However, there is currently no reliable and practical tool to reflect the global fatty infiltration (Global-FI) throughout the 3-dimensional (3D) volumetric RC muscles.

PURPOSE

(1) To determine the correlations between 2 modified assessment tools and the Global-FI and their predictive performances and reliabilities for Global-FI prediction, and (2) to compare these predictive parameters with those of the conventional tool using a single scapular Y-view slice.

STUDY DESIGN

Cohort study (diagnosis); Level of evidence, 3.

METHODS

A total of 49 patients with full-thickness RC tears scheduled to undergo arthroscopic repairs were included, and their surgical shoulders underwent 6-point Dixon magnetic resonance imaging preoperatively. The Global-FI was measured by calculating the 3D-volumetric fat fraction (FF) of the whole supraspinatus muscle through all acquired oblique sagittal slices. As a commonly used radiological landmark, the scapular Y-view was used to evaluate single-plane fatty infiltration (Y-FI) by calculating the FF in 1 slice, defined as the conventional assessment tool. Two modified assessment tools expand the analytic imaging by integrating the FFs from the scapular Y-view slice and its neighboring slices, which were calculated by averaging the FFs of these 3 slices (^meanY₃-FI) and accumulating local 3D-volumetric FFs from 3 slices (^volY₃-FI), respectively. The correlations between 3 assessment tools and the Global-FI were analyzed, and the predictive performance for Global-FI prediction using these tools was determined by goodness of fit and agreement. Moreover, the inter- and intraobserver reliabilities of these assessment tools were evaluated. Similar analyses were performed in the small-medium, large, or massive tear subgroups.

RESULTS

The Y-FI was significantly higher than the Global-FI in all cases and tear size subgroups, while the 2 modified assessment tools (^meanY₃-FI and ^volY₃-FI) did not significantly differ from the Global-FI. All assessment tools were significantly correlated with the Global-FI, but the ^meanY₃-FI and ^volY₃-FI showed stronger correlations than the Y-FI, which was also determined in different tear sizes. Moreover, the regression models of the ^meanY₃-FI and ^volY₃-FI showed superior goodness of fit to Y-FI in Global-FI prediction in all cases and subgroups, with larger coefficients of determination (R²) and smaller root mean square errors. The predicted Global-FI using the regression model of ^volY₃-FI had the best agreement with the measured Global-FI, followed by the ^meanY₃-FI, both showing smaller biases and standard deviation of the percentage difference between predicted- and measured Global-FI than the conventional Y-FI. In addition, the 2 modified assessment tools achieved better interobserver and intraobserver reliabilities than the conventional tool in all cases and subgroups.

CONCLUSION

Two modified assessment tools (^meanY₃-FI and ^volY₃-FI) were comparable with the Global-FI of the whole supraspinatus muscle, showing stronger correlations with the Global-FI and better predictive performances and reliabilities than the conventional tool (Y-FI). Moreover, the ^volY₃-FI was slightly superior to ^meanY₃-FI in the predictive performance and reliability.

Classification of rotator cuff tears in ultrasound images using deep learning models

Rotator cuff tears (RCTs) are one of the most common shoulder injuries, which are typically diagnosed using relatively expensive and time-consuming diagnostic imaging tests such as magnetic resonance imaging or computed tomography. Deep learning algorithms are increasingly used to analyze medical images, but they have not been used to identify RCTs with ultrasound images. The aim of this study is to develop an approach to automatically classify RCTs and provide visualization of tear location using ultrasound images and convolutional neural networks (CNNs). The proposed method was developed using transfer learning and fine-tuning with five pre-trained deep models (VGG19, InceptionV3, Xception, ResNet50, and DenseNet121). The Bayesian optimization method was also used to optimize hyperparameters of the CNN models. A total of 194 ultrasound images from Kosin University Gospel Hospital were used to train and test the CNN models by five-fold cross-validation. Among the five models, DenseNet121 demonstrated the best classification performance with 88.2% accuracy, 93.8% sensitivity, 83.6% specificity, and AUC score of 0.832. A gradient-weighted class activation mapping (Grad-CAM) highlighted the sensitive features in the learning process on ultrasound images. The proposed approach demonstrates the feasibility of using deep learning and ultrasound images to assist RCTs' diagnosis.

Three-dimensional muscle loss assessment: a novel computed tomography-based quantitative method to evaluate rotator cuff muscle fatty infiltration

BACKGROUND

Rotator cuff fatty infiltration (FI) is one of the most important parameters to predict the outcome of certain shoulder conditions. The primary objective of this study was to define a new computed tomography (CT)-based quantitative 3-dimensional (3D) measure of muscle loss (3DML) based on the rationale of the 2-dimensional (2D) qualitative Goutallier score. The secondary objective of this study was to compare this new measurement method to traditional 2D qualitative assessment of FI according to Goutallier et al and to a 3D quantitative measurement of fatty infiltration (3DFI).

MATERIALS AND METHODS

102 CT scans from healthy shoulders (46) and shoulders with cuff tear arthropathy (21), irreparable rotator cuff tears (18), and primary osteoarthritis (17) were analyzed by 3 experienced shoulder surgeons for subjective grading of fatty infiltration according to Goutallier, and their rotator cuff muscles were manually segmented. Quantitative 3D measurements of fatty infiltration (3DFI) were completed. The volume of muscle fibers without intramuscular fat was then calculated for each rotator cuff muscle and normalized to the patient's scapular volume to account for the effect of body size (NV_fibers). 3D muscle mass (3DMM) was calculated by dividing the NV_fibers value of a given muscle by the mean expected volume in healthy shoulders. 3D muscle loss (3DML) was defined as 1 - (3DMM). The correlation between Goutallier grading, 3DFI, and 3DML was compared using a Spearman rank correlation.

RESULTS

Interobserver reliability for the traditional 2D Goutallier grading was moderate for the infraspinatus (ISP, 0.42) and fair for the supraspinatus (SSP, 0.38), subscapularis (SSC, 0.27) and teres minor (TM, 0.27). 2D Goutallier grading was found to be significantly and highly correlated with 3DFI (SSP, 0.79; ISP, 0.83; SSC, 0.69; TM, 0.45) and 3DML (SSP, 0.87; ISP, 0.85; SSC, 0.69; TM, 0.46) for all 4 rotator cuff muscles (P < .0001). This correlation was significantly higher for 3DML than for the 3DFI for SSP only (P = .01). The mean values of 3DFI and 3DML were 0.9% and 5.3% for Goutallier 0, 2.9% and 25.6% for Goutallier 1, 11.4% and 49.5% for Goutallier 2, 20.7% and 59.7% for Goutallier 3, and 29.3% and 70.2% for Goutallier 4, respectively.

CONCLUSION

The Goutallier score has been helping surgeons by using 2D CT scan slices. However, this grading is associated with suboptimal interobserver agreement. The new measures we propose provide a more consistent assessment that correlates well with Goutallier's principles. As 3DML measurements incorporate atrophy and fatty infiltration, they could become a very reliable index for assessing shoulder muscle function. Future algorithms capable of automatically calculating the 3DML of the cuff could help in the decision process for cuff repair and the choice of anatomic or reverse shoulder arthroplasty.

A Matlab toolbox for scaled-generic modeling of shoulder and elbow

There still remains a barrier ahead of widespread clinical applications of upper extremity musculoskeletal models. This study is a step toward lifting this barrier for a shoulder musculoskeletal model by enhancing its realism and facilitating its applications. To this end, two main improvements are considered. First, the elbow and the muscle groups spanning the elbow are included in the model. Second, scaling routines are developed that scale model’s bone segment inertial properties, skeletal morphologies, and muscles architectures according to a specific subject. The model is also presented as a Matlab toolbox with a graphical user interface to exempt its users from further programming. We evaluated effects of anthropometric parameters, including subject’s gender, height, weight, glenoid inclination, and degenerations of rotator cuff muscles on the glenohumeral joint reaction force (JRF) predictions. An arm abduction motion in the scapula plane is simulated while each of the parameters is independently varied. The results indeed illustrate the effect of anthropometric parameters and provide JRF predictions with less than 13% difference compared to in vivo studies. The developed Matlab toolbox could be populated with pre/post operative patients of total shoulder arthroplasty to answer clinical questions regarding treatments of glenohumeral joint osteoarthritis.

Volumetric MicroCT Intensity Histograms of Fatty Infiltration Correlate with the Mechanical Strength of Rotator Cuff Repairs: An Ex Vivo Rabbit Model

BACKGROUND

Fatty infiltration of the rotator cuff occurs after injury to the tendon and results in a buildup of adipose in the muscle. Fatty infiltration may be a biomarker for predicting future injuries and mechanical properties after tendon repair. As such, quantifying fatty infiltration accurately could be a relevant metric for determining the success of tendon repairs. Currently, fatty infiltration is quantified by an experienced observer using the Goutallier or Fuchs staging system, but because such score-based quantification systems rely on subjective assessments, newer techniques using semiautomated analyses in CT and MRI were developed and have met with varying degrees of success. However, semiautomated analyses of CT and MRI results remain limited in cases where only a few two-dimensional slices of tissue are examined and applied to the three-dimensional (3-D) tissue structure. We propose that it is feasible to assess fatty infiltration within the 3-D volume of muscle and tendon in a semiautomated fashion by selecting anatomic features and examining descriptive metrics of intensity histograms collected from a cylinder placed within the central volume of the muscle and tendon of interest.

QUESTIONS/PURPOSES

(1) Do descriptive metrics (mean and SD) of intensity histograms from microCT images correlate with the percentage of fat present in muscle after rotator cuff repair? (2) Do descriptive metrics of intensity histograms correlate with the maximum load during mechanical testing of rotator cuff repairs?

METHODS

We developed a custom semiautomated program to generate intensity histograms based on user-selected anatomic features. MicroCT images were obtained from 12 adult female New Zealand White rabbits (age 8 to 12 months, weight 3.7 kg ± 5 kg) that were randomized to surgical repair or sham repair of an induced infraspinatus defect. Intensity histograms were generated from images of the operative and contralateral intact shoulder in these rabbits which were presented to the user in a random order without identifying information to minimize sources of bias. The mean and SD of the intensity histograms were calculated and compared with the total percentage of the volume threshold as fat. Patterns of fat identified were qualitatively compared with histologic samples to confirm that thresholding was detecting fat. We conducted monotonic tensile strength-to-failure tests of the humeral-infraspinatus bone-tendon-muscle complex, and evaluated associations between histogram mean and SDs and maximum load.

RESULTS

The total percentage of fat was negatively correlated with the intensity histogram mean (Pearson correlation coefficient -0.92; p < 0.001) and positively with intensity histogram SD (Pearson correlation coefficient 0.88; p < 0.001), suggesting that the increase in fat leads to a reduction and wider variability in volumetric tissue density. The percentage of fat content was also negatively correlated with the maximum load during mechanical testing (Pearson correlation coefficient -78; p = 0.001), indicating that as the percentage of fat in the volume increases, the mechanical strength of the repair decreases. Furthermore, the intensity histogram mean was positively correlated with maximum load (Pearson correlation coefficient 0.77; p = 0.001) and histogram SD was negatively correlated with maximum load (Pearson correlation coefficient -0.72; p = 0.004). These correlations were strengthened by normalizing maximum load to account for animal size (Pearson correlation coefficient 0.86 and -0.9, respectively), indicating that as histogram mean decreases, the maximum load of the repair decreases and as histogram spread increases, the maximum load decreases.

CONCLUSION

In this ex vivo rabbit model, a semiautomated approach to quantifying fat on microCT images was a noninvasive way of quantifying fatty infiltration associated with the strength of tendon healing.

CLINICAL RELEVANCE

Histogram-derived variables may be useful as surrogate measures of repair strength after rotator cuff repair. The preclinical results presented here provide a foundation for future studies to translate this technique to patient studies and additional imaging modalities. This semiautomated method provides an accessible approach to quantification of fatty infiltration by users of varying experience and can be easily adapted to any intensity-based imaging approach. To translate this approach to clinical practice, this technique should be calibrated for MRI or conventional CT imaging and applied to patient scans. Further investigations are needed to assess the correlation of volumetric intensity histogram descriptive metrics to clinical mechanical outcomes.

Deep learning for the rapid automatic quantification and characterization of rotator cuff muscle degeneration from shoulder CT datasets

Objectives

This study aimed at developing a convolutional neural network (CNN) able to automatically quantify and characterize the level of degeneration of rotator cuff (RC) muscles from shoulder CT images including muscle atrophy and fatty infiltration.

Methods

One hundred three shoulder CT scans from 95 patients with primary glenohumeral osteoarthritis undergoing anatomical total shoulder arthroplasty were retrospectively retrieved. Three independent radiologists manually segmented the premorbid boundaries of all four RC muscles on standardized sagittal-oblique CT sections. This premorbid muscle segmentation was further automatically predicted using a CNN. Automatically predicted premorbid segmentations were then used to quantify the ratio of muscle atrophy, fatty infiltration, secondary bone formation, and overall muscle degeneration. These muscle parameters were compared with measures obtained manually by human raters.

Results

Average Dice similarity coefficients for muscle segmentations obtained automatically with the CNN (88% ± 9%) and manually by human raters (89% ± 6%) were comparable. No significant differences were observed for the subscapularis, supraspinatus, and teres minor muscles (p > 0.120), whereas Dice coefficients of the automatic segmentation were significantly higher for the infraspinatus (p < 0.012). The automatic approach was able to provide good–very good estimates of muscle atrophy (R² = 0.87), fatty infiltration (R² = 0.91), and overall muscle degeneration (R² = 0.91). However, CNN-derived segmentations showed a higher variability in quantifying secondary bone formation (R² = 0.61) than human raters (R² = 0.87).

Conclusions

Deep learning provides a rapid and reliable automatic quantification of RC muscle atrophy, fatty infiltration, and overall muscle degeneration directly from preoperative shoulder CT scans of osteoarthritic patients, with an accuracy comparable with that of human raters.

Key Points

• Deep learning can not only segment RC muscles currently available in CT images but also learn their pre-existing locations and shapes from invariant anatomical structures visible on CT sections.

• Our automatic method is able to provide a rapid and reliable quantification of RC muscle atrophy and fatty infiltration from conventional shoulder CT scans.

• The accuracy of our automatic quantitative technique is comparable with that of human raters.

Correction to: Qualitative and quantitative analysis of glenoid bone stock and glenoid version: inter-reader analysis and correlation with rotator cuff tendinopathy and atrophy in patients with shoulder osteoarthritis

Purpose Glenoid bone stock and morphology and rotator cuff muscle quality and tendon integrity affect the outcome of total shoulder arthroplasty. We hypothesized that glenoid bone loss correlates with rotator cuff muscle fatty infiltration (FI), tendinopathy, and atrophy.Design Forty-three 3-D CT scans and MRIs of 43 patients (mean age 62 years; SD 13 years; range 22-77 years) referred for primary shoulder pain were evaluated. Measurements of glenoid bone stock, version, and posterior humeral subluxation index (HSI) were assessed on an axial CT image reconstructed in the true scapular plane. Measurements utilized the Friedman line to approximate the pre-pathologic surface. Glenoid morphology was assigned by modified Walch classification. Rotator cuff FI, atrophy, and tendon integrity were assessed on corresponding MRIs.Results There was a very strong negative correlation between increasing glenoid version and HSI (r = - 0.908; p < 0.0001). There was a moderately negative correlation between anterior bone loss and HSI (r = - 0.562; p < 0.0001) and a moderately positive correlation between posterior bone loss and HSI (r = 0.555; p < 0.0001). Subscapularis muscle FI correlated moderately with increased anterior and central bone loss and increased humeral head medialization (r = 0.512, p = 0.0294; r = 0.479, p = 0.033; r = 0.494, p = 0.0294, respectively). Inter-observer reliability (intra-class correlation coefficient [ICC] and kappa) was good to excellent for all measurements and grading.Conclusion Glenoid anteversion and anterior and posterior bone loss are associated with varying HSI. Subscapularis muscle FI, not tendon integrity, correlates to anterior and central glenoid erosion. The study adds evidence that neither rotator cuff tendinopathy nor muscle atrophy exhibits a significant relationship to HSI.

Reduction of scapulohumeral subluxation with posterior augmented glenoid implants in anatomic total shoulder arthroplasty: Short-term 3D comparison between pre- and post-operative CT

BACKGROUND

Failure rates in anatomic total shoulder arthroplasty (aTSA) are higher in case of asymmetric glenoid bone loss secondary to posterior wear, and in persistent static posterior subluxation of the humeral head (PSH).

HYPOTHESIS

This study aimed to test the hypothesis that the combined use of posterior augmented glenoid (PAG) implants with three-dimensional (3D) surgical planning and patient-specific instrumentation (PSI) guides helps reduce short-term PSH after aTSA in patients with type B2-B3 glenoids.

PATIENTS AND METHODS

We included nine consecutive patients with primary glenohumeral osteoarthritis and type B2 or B3 glenoids, who underwent aTSA with cemented keeled PAG implants (posterior augments of 15, 25, or 35 degrees). All patients underwent preoperative shoulder computed tomography (CT) scans, with 3D surgical planning coupled to PSI at the time of surgery. Postoperative shoulder CT scans were performed at an average of 14 weeks (range, 10-21 weeks). Scapulohumeral subluxation and glenoid version and inclination were measured in 3D, on both pre- and post-operative CT scans, using the same reliable quantitative method.

RESULTS

There was a significant decrease in scapulohumeral subluxation from 49±12% preoperatively to 22±17% postoperatively (p=0.0039), with a large effect size (Cohen's d=1.89). Preoperative glenoid version was corrected from -17.3±9.4 degrees to -5.2±7.5 degrees postoperatively. The absolute difference between the postoperative and surgically planned version and inclination was 5.4±3.6 degrees and 3.3±2.0 degrees, respectively.

DISCUSSION

The combined use of PAG implants with 3D preoperative planning and PSI results in a significant decrease in short-term PSH and glenoid version in patients with asymmetric posterior glenoid wear. We suggest that such implants should not be limited to posterior augmentation, because glenoid deficiency can also be observed in other glenoid sectors.

LEVEL OF EVIDENCE

IV, Basic science study.

A Robotic Glenohumeral Simulator for Investigating Prosthetic Implant Subluxation

Total shoulder arthroplasty (TSA) is an effective treatment for glenohumeral (GH) osteoarthritis. However, it still suffers from a substantial rate of mechanical failure, which may be related to cyclic off-center loading of the humeral head on the glenoid. In this work, we present the design and evaluation of a GH joint robotic simulator developed to study GH translations. This five-degree-of-freedom robot was designed to replicate the rotations (±40 deg, accuracy 0.5 deg) and three-dimensional (3D) forces (up to 2 kN, with a 1% error settling time of 0.6 s) that the humeral implant exerts on the glenoid implant. We tested the performances of the simulator using force patterns measured in real patients. Moreover, we evaluated the effect of different orientations of the glenoid implant on joint stability. When simulating realistic dynamic forces and implant orientations, the simulator was able to reproduce stable behavior by measuring the translations of the humeral head of less than 24 mm with respect to the glenoid implant. Simulation with quasi-static forces showed dislocation in extreme ranges of implant orientation. The robotic GH simulator presented here was able to reproduce physiological GH forces and may therefore be used to further evaluate the effects of glenoid implant design and orientation on joint stability.

Biomechanical comparison of glenoid implants with adaptable and fixed backside curvatures in anatomic total shoulder arthroplasty

BACKGROUND

We evaluated the biomechanical effects and potential advantages of glenoid implants with adaptable backside curvature radii and compared them with standard implants having fixed backside curvatures in anatomic total shoulder arthroplasty (aTSA) for primary glenohumeral osteoarthritis with uniconcave glenoids.

METHODS

A glenoid implant with adaptable backside curvatures (Aequalis PerFORM, Tornier SAS, Montbonnot, France) was compared with its previous model having a fixed curvature radius. Virtual aTSAs were performed in 24 patients from preoperative shoulder computed tomography data sets, using both implants in each patient. For all 48 simulated aTSAs, we first measured the glenoid bone reaming depth, subchondral bone quality after reaming, and implant backside surface and then the predicted cement stress, bone-cement interfacial stress, and bone strain at 60° of arm abduction. These biomechanical quantities were tested for differences between adaptable and fixed implants and for correlations between preoperative measurements and postoperative predictions.

RESULTS

Adaptable glenoid implants induced a significant decrease in cement stress (P = .008), bone-cement interfacial stress (P = .045), and bone strain (P = .039), particularly for glenoids with curvature radii larger than 40 mm. However, these biomechanical effects were not significantly correlated with an increase in subchondral glenoid bone quality.

CONCLUSIONS

Our study confirms the presumed biomechanical advantages of adaptable glenoid implants, even though the effects were not directly due to the adaptation of the backside curvature radius. Benefits were more pronounced for glenoids with large curvature radii. Our initial biomechanical findings should now be corroborated with large-scale clinical studies.

Cement stress predictions after anatomic total shoulder arthroplasty are correlated with preoperative glenoid bone quality

HYPOTHESIS

We hypothesized that biomechanical parameters typically associated with glenoid implant failure after anatomic total shoulder arthroplasty (aTSA) would be correlated with preoperative glenoid bone quality.

METHODS

We developed an objective automated method to quantify preoperative glenoid bone quality in different volumes of interest (VOIs): cortical bone, subchondral cortical plate, subchondral bone after reaming, subchondral trabecular bone, and successive layers of trabecular bone. Average computed tomography (CT) numbers (in Hounsfield units [HU]) were measured in each VOI from preoperative CT scans. In parallel, we built patient-specific finite element models of simulated aTSAs to predict cement stress, bone-cement interfacial stress, and bone strain around the glenoid implant. CT measurements and finite element predictions were obtained for 20 patients undergoing aTSA for primary glenohumeral osteoarthritis. We tested all linear correlations between preoperative patient characteristics (age, sex, height, weight, glenoid bone quality) and biomechanical predictions (cement stress, bone-cement interfacial stress, bone strain).

RESULTS

Average CT numbers gradually decreased from cortical (717 HU) to subchondral and trabecular (362 HU) bone. Peak cement stress (4-10 MPa) was located within the keel hole, above the keel, or behind the glenoid implant backside. Cement stress, bone-cement interfacial stress, and bone strain were strongly negatively correlated with preoperative glenoid bone quality, particularly in VOIs behind the implant backside (subchondral trabecular bone) but also in deeper trabecular VOIs.

CONCLUSION

Our numerical study suggests that preoperative glenoid bone quality is an important parameter to consider in aTSA, which may be associated with aseptic loosening of the glenoid implant. These initial results should now be confronted with clinical and radiologic outcomes.