Re-tear after arthroscopic rotator cuff tear surgery: risk analysis using machine learning

The influence of age on cellular senescence in injured versus healthy muscle and its implications on rotator cuff injuries

BACKGROUND

Advanced age increases the prevalence of rotator cuff tears and affects the success of repair surgeries. Cellular senescence is proposed as a key mechanism behind these age-related differences, likely due to contribution of the senescence-associated secretory phenotype. This state is linked to various age-related diseases, including rotator cuff injuries.

MATERIALS AND METHODS

Rotator cuff muscle samples were obtained from young and aged patients who underwent surgery. Samples were processed for single-cell RNA sequencing to analyze cellular differences. Cells were isolated and sequenced to identify different cell populations and their gene expression profiles.

RESULTS

Six major cell populations were identified in rotator cuff muscle tissue, including fibroadipogenic progenitor cells (FAPs), satellite cells, endothelial cells, pericytes, macrophages, and T cells. Aged FAPs showed higher expression of senescence markers and genes associated with fibrosis and inflammation. Younger FAPs had higher levels of extracellular matrix remodeling genes. Specifically, ATF3-a senescence marker-was found to be elevated in aged FAPs. In silico analysis highlighted a potential role of ATF3 in regulating FAP differentiation.

CONCLUSIONS

Markers of cellular senescence are significantly elevated in older human rotator cuff tissue samples compared with young rotator cuff. Of specific interest is ATF3, a gene that has been previously implicated in regulating adipogenesis, which demonstrates a trend to function in a protective capacity against the formation of fibrosis in computational analysis of our data.

Preoperative Increases in T2-Weighted Fat-Suppressed Magnetic Resonance Imaging Signal Intensities Associated With Advanced Tissue Degeneration and Mitochondrial Dysfunction in Rotator Cuff Tears

PURPOSE

To investigate the relationship between magnetic resonance imaging (MRI) signal intensities and mitochondrial function in patients undergoing arthroscopic rotator cuff repair, assessed through histological and genetic profiling of tendon tissue.

METHODS

This study, conducted between April 2022 and January 2023, included 20 patients undergoing rotator cuff repair for atraumatic/degenerative tears. Rotator cuff tendon edge samples were obtained during arthroscopic rotator cuff repair. Patients were classified based on signal intensity from preoperative T2-weighted fat suppressed MRI. Specifically, they were categorized as having either high or low signal intensity at the rotator cuff tendon edge, with the deltoid muscle serving as a reference. Comparative analyses specifically compared the histological features and genetic profiles of the tendon tissue at the rotator cuff tendon edge. Histological evaluation of harvested tendon specimens during arthroscopic rotator cuff repair employed the modified Bonar score. Real-time polymerase chain reaction was used to assess expression of various mitochondrial and apoptosis-related genes. The mitochondrial morphology of the rotator cuff torn site was examined using electron microscopy.

RESULTS

The higher signal intensity group showed significantly higher modified Bonar scores (P = .0068), decreased mitochondrial gene expression, increased TdT-mediated dUTP-biotin nick end labeling-positive cells (P = .032), lower superoxide dismutase activity (P = .011), reduced ATP5A (P = .031), and increased cleaved caspase-9 activity (P = .026) compared with the lower signal intensity group. Electron microscopy revealed fewer mitochondrial cristae in the higher signal intensity group.

CONCLUSIONS

Our results suggest correlations between high MRI signal intensities and the presence of degeneration, mitochondrial dysfunction, and increased apoptosis in rotator cuff tissues. This underscores the utility of MRI signal intensity as an indicator of tissue condition.

CLINICAL RELEVANCE

This work builds on the premise that elevated preoperative MRI signal intensities may indicate higher rates of postoperative rotator cuff re-tears, substantiating these findings from a mitochondrial function perspective.

Editorial Commentary: Deteriorated Quality and Medial Retraction of Tendon Following Acute Traumatic Rotator Cuff Tear are Predictors of Retear After Arthroscopic Repair

Rotator cuff tears are a common cause of shoulder pain and dysfunction, affecting up to 33% of the population, and approximately 250,000 arthroscopic rotator cuff repairs are performed each year in the United States. Arthroscopic rotator cuff repair (RCR) is the gold standard for surgical management of full thickness rotator cuff tears and is associated with postoperative improvements in pain and function. However, reported retear rates based on postoperative magnetic resonance imaging (MRI) vary from 7%-90% following arthroscopic rotator cuff repair. Despite variations in repair techniques, implant technology, biologic and patch augmentation, and postoperative rehabilitation algorithms, retear rates following rotator cuff repair have remained high over the past several decades. As a result, there continues to be a significant interest in identifying predictive factors of retear after rotator cuff repair. That said, numerous well-designed studies have demonstrated a poor correlation between postoperative MRI findings and functional outcomes. Regardless, intraoperative evaluation of tendon quality, tear pattern, and tissue mobility is essential to predicting the likelihood of successful rotator cuff repair, although at the current time, this evaluation is largely subjective, and few validated assessment tools exist. As such, future, objective research is required to improve our assessment and documentation of these intraoperative factors, with artificial intelligence and machine learning models possibly serving as useful tools for identifying meaningful intraoperative patterns predictive of postoperative outcomes in the future.

Artificial intelligence applications in the management of musculoskeletal disorders of the shoulder: A systematic review

Purpose

The aim of the present review is to evaluate and report on the available literature discussing artificial intelligence (AI) applications to the diagnosis of shoulder conditions, outcome prediction of shoulder interventions, and the possible application of such algorithms directly to surgical procedures.

Methods

In February 2024, a search of PubMed, Cochrane and Scopus databases was performed. Studies had to evaluate AI model effectiveness for inclusion. Research on healthcare cost predictions, deterministic algorithms, patient satisfaction, protocol studies and upper-extremity fractures not involving the shoulder were excluded. The Joanna Briggs Institute Critical Appraisal tool and the Risk of Bias in Non-randomised Studies of Interventions tools were used to assess bias.

Results

Thirty-three studies were included in the analysis. Seven studies analysed the detection of rotator cuff tears (RCTs) in magnetic resonance imaging and found area under the curve (AUC) values ranged from 0.812 to 0.94 for the detection of RCTs. One study reported Area Under the Receiver Operating Characteristics values ranging from 0.79 to 0.97 for the prediction of clinical outcomes following reverse total shoulder arthroplasty. In terms of outcomes of rotator cuff repair, an AUC value ranging from 0.58 to 0.68 was reported for prediction of patient-reported outcome measures, and an AUC range of 0.87-0.92 was found for prediction of retear rate. Five studies evaluated the identification of shoulder implant models following TSA from radiographs, with reported accuracy ranging from 89.90% to 97.20%.

Conclusion

AI application enables forecasting of clinical outcomes, permits refined diagnostic evaluation and increases surgical accuracy. While promising, the translation of these technologies into routine clinical practice requires careful consideration.

Level of Evidence

Level IV.

Posture Estimation Model Combined With Machine Learning Estimates the Radial Abduction Angle of the Thumb With High Accuracy

The thumb function is complex, and accurate evaluation through images or videos is difficult. Pose estimation, a technology that uses artificial intelligence (AI) to estimate skeletal detection of the body, is gaining popularity. In this study, we combined the pose estimation library MediaPipe-Hands and five machine learning (ML) models to predict the radial abduction angle of the thumb. Radial abduction movements of 20 hands from 10 healthy volunteers were captured on video and processed into 5,000 images. Angle measurements by goniometer were used as true values to evaluate the angle reliability of the MediaPipe-Hands and the angle reliability of the MediaPipe-Hands combined with ML. The correlation coefficient (CC) between the angle measured by goniometry and the angle calculated by MediaPipe-Hands was 0.84. In contrast, applying ML to MediaPipe-Hands resulted in models with improved accuracy, and all models showed high CCs (0.94-099) with angle measurements taken by goniometry. The ML model also predicted the abduction angles when the camera was taken from three different angles. In visualizing the features that the AI deemed important, the ML model predicted the abduction angle by focusing on the tip distance between the thumb and index finger along with the angle of the metacarpophalangeal joint between the thumb and middle finger. These results enable angle estimation even without frontal imaging with a camera, and expansion of this system may lead to real-time functional assessment in telemedicine and rehabilitation without the need for physical contact.

Clinical Outcomes and Re-Tear Rates for Partial Arthroscopic Rotator Cuff Repair With or Without Biceps Augmentation for Large-to-Massive Tears: A Systematic Review and Meta-analysis

BACKGROUND

The recent addition of biceps tendon augmentation to partial arthroscopic rotator cuff repair (ARCR) for the treatment of large-to-massive rotator cuff tears is proposed to improve clinical outcomes and reduce re-tears.

MATERIALS AND METHODS

The purpose of this systematic review and meta-analysis (5 studies) was to compare outcomes between partial ARCR with (142 patients) and without (149 patients) biceps augmentation.

RESULTS

Partial ARCR with and without biceps augmentation were comparable in pain, function, and range of motion. However, biceps augmentation vs no augmentation at all during ARCR may lower re-tear rates for irreparable large-to-massive rotator cuff tears (42.9% vs 72.5%, P=.007).

CONCLUSION

More research is needed to investigate this technique and guide surgical decision-making. [Orthopedics. 2024;47(5):e217-e224.].

Study on Shoulder Joint Parameters and Available Supraspinatus Outlet Area Using Three-Dimensional Computed Tomography Reconstruction

Studies addressing the anatomical values of the supraspinatus outlet area (SOA) and the available supraspinatus outlet area (ASOA) are insufficient. This study focused on precisely measuring the SOA and ASOA values in a sample from the Chinese population using 3D CT (computed tomography) reconstruction. We analyzed CT imaging of 96 normal patients (59 males and 37 females) who underwent shoulder examinations in a hospital between 2011 and 2021. The SOA, ASOA, acromiohumeral distance (AHD), coracohumeral distance (CHD), coracoacromial arch radius (CAR), and humeral head radius (HHR) were estimated, and statistical correlation analyses were performed. There were significant sex differences observed in SOA (men: 957.62 ± 158.66 mm2; women: 735.87 ± 95.86 mm2) and ASOA (men: 661.35 ± 104.88 mm2; women: 511.49 ± 69.26 mm2), CHD (men: 11.22 ± 2.24 mm; women: 9.23 ± 1.35 mm), CAR (men: 37.18 ± 2.70 mm; women: 33.04 ± 3.15 mm), and HHR (men: 22.65 ± 1.44 mm; women: 20.53 ± 0.95 mm). Additionally, both SOA and ASOA showed positive and linear correlations with AHD, CHD, CAR, and HHR (R: 0.304-0.494, all p < 0.05). This study provides physiologic reference values of SOA and ASOA in the Chinese population, highlighting the sex differences and the correlations with shoulder anatomical parameters.

Re-tear after arthroscopic rotator cuff repair can be predicted using deep learning algorithm

The application of artificial intelligence technology in the medical field has become increasingly prevalent, yet there remains significant room for exploration in its deep implementation. Within the field of orthopedics, which integrates closely with AI due to its extensive data requirements, rotator cuff injuries are a commonly encountered condition in joint motion. One of the most severe complications following rotator cuff repair surgery is the recurrence of tears, which has a significant impact on both patients and healthcare professionals. To address this issue, we utilized the innovative EV-GCN algorithm to train a predictive model. We collected medical records of 1,631 patients who underwent rotator cuff repair surgery at a single center over a span of 5 years. In the end, our model successfully predicted postoperative re-tear before the surgery using 62 preoperative variables with an accuracy of 96.93%, and achieved an accuracy of 79.55% on an independent external dataset of 518 cases from other centers. This model outperforms human doctors in predicting outcomes with high accuracy. Through this methodology and research, our aim is to utilize preoperative prediction models to assist in making informed medical decisions during and after surgery, leading to improved treatment effectiveness. This research method and strategy can be applied to other medical fields, and the research findings can assist in making healthcare decisions.