The accuracy of plain radiographs in diagnosing degenerate rotator cuff disease

Deep learning-based screening tool for rotator cuff tears on shoulder radiography

BACKGROUND

Early diagnosis of rotator cuff tears is essential for appropriate and timely treatment. Although radiography is the most used technique in clinical practice, it is difficult to accurately rule out rotator cuff tears as an initial imaging diagnostic modality. Deep learning-based artificial intelligence has recently been applied in medicine, especially diagnostic imaging. This study aimed to develop a deep learning algorithm as a screening tool for rotator cuff tears based on radiography.

METHODS

We used 2803 shoulder radiographs of the true anteroposterior view to develop the deep learning algorithm. Radiographs were labeled 0 and 1 as intact or low-grade partial-thickness rotator cuff tears and high-grade partial or full-thickness rotator cuff tears, respectively. The diagnosis of rotator cuff tears was determined based on arthroscopic findings. The diagnostic performance of the deep learning algorithm was assessed by calculating the area under the curve (AUC), sensitivity, negative predictive value (NPV), and negative likelihood ratio (LR-) of test datasets with a cutoff value of expected high sensitivity determination based on validation datasets. Furthermore, the diagnostic performance for each rotator cuff tear size was evaluated.

RESULTS

The AUC, sensitivity, NPV, and LR- with expected high sensitivity determination were 0.82, 84/92 (91.3%), 102/110 (92.7%), and 0.16, respectively. The sensitivity, NPV, and LR- for full-thickness rotator cuff tears were 69/73 (94.5%), 102/106 (96.2%), and 0.10, respectively, while the diagnostic performance for partial-thickness rotator cuff tears was low at 15/19 (78.9%), NPV of 102/106 (96.2%) and LR- of 0.39.

CONCLUSIONS

Our algorithm had a high diagnostic performance for full-thickness rotator cuff tears. The deep learning algorithm based on shoulder radiography helps screen rotator cuff tears by setting an appropriate cutoff value.

LEVEL OF EVIDENCE

Level III: Diagnostic Study.

Automated detection and classification of the rotator cuff tear on plain shoulder radiograph using deep learning

BACKGROUND

The diagnosis of rotator cuff tears (RCTs) using radiographs alone is clinically challenging; thus, the utility of deep learning algorithms based on convolutional neural networks has been remarkable in the field of medical imaging recognition. We aimed to evaluate the diagnostic performance of artificial intelligence (a deep learning algorithm; a convolutional neural network) to detect and classify RCTs using shoulder radiographs, and compare its diagnostic performance with that of orthopedic surgeons.

METHODS

A total of 1169 plain shoulder anteroposterior radiographs (1 image per shoulder) were included in the total dataset and divided into four groups: intact, small, medium, and large to massive tear groups. The sensitivity, specificity, positive predictive value, negative predictive value, accuracy, and area under the receiver operating curve were measured for the detection of RCTs through binary classification. The average accuracy, recall, precision, and F1-score were divided into four groups by cuff tear size for multiclass classification.

RESULTS

The convolutional neural network demonstrated a high performance, with 92% sensitivity, 69% specificity, 86% accuracy, and an area under the receiver operating curve of 0.88 for the detection of RCTs. The average accuracy, recall, precision, and F1-score of the convolutional neural network for classification were 60%, 0.42, 0.49, and 0.45, respectively. The accuracy of the convolutional neural network for the detection and classification of RCTs was significantly better than that of orthopedic surgeons.

CONCLUSION

The convolutional neural network demonstrated the diagnostic ability to detect and classify RCTs using plain shoulder radiographs, and the diagnostic performance exhibited equal to superior accuracy when compared with those of shoulder experts.

Glenohumeral osteoarthritis: what the surgeon needs from the radiologist

Glenohumeral osteoarthritis (GHOA) is a widely prevalent disease with increasing frequency due to population aging. Both clinical manifestations and radiography play key roles in the initial diagnosis, staging, and management decisions. Radiographic disease progression evaluation is performed using validated staging systems, such as Kellgren and Lawrence, Samilson, and Hamada. For young patients with mild to moderate GHOA and failed conservative treatment, arthroscopic preservation surgery (APS) is usually considered. Older patients and those with severe GHOA benefit from different types of arthroplasties. Preoperative magnetic resonance imaging (MRI) is essential for APS surgical planning, as it maps repairable labral, cartilage, and rotator cuff lesions. For arthroplasty planning, the status of glenoid cartilage and intactness of rotator cuff as well as glenoid morphology represent key factors guiding the decision regarding the most suitable hardware design, whether resurfacing, partial, total, or reverse joint replacement. Pre-surgical MRI or alternatively computed tomography arthrogram is employed to evaluate the cartilage and rotator cuff. Finally, three-dimensional computed tomography (3D CT) is indicated to optimally assess the glenoid morphology (to determine Walch classification, version, inclination, and bone loss) and analyze the necessity for glenoid osteotomy or graft augmentation to correct the glenoid structural abnormalities for future success and longevity of the shoulder implants or chosen constructs. Understanding the purpose of each imaging and treatment modality allows more efficient image interpretation. This article reviews the above concepts and details what a surgeon needs from a radiologist and could benefit from accurate reporting of preoperative imaging studies.

Development and Validation of a Web-Based Dynamic Nomogram to Improve the Diagnostic Performance of Subscapularis Tendon Tear

Background

There are still some challenges in diagnosing subscapularis (SSC) tendon tears as accurately as posterosuperior rotator cuff tears on magnetic resonance imaging (MRI). The omission of SSC tendon tears can lead to muscle atrophy, fatty infiltration and increased tear accompanied by aggravated shoulder pain and loss of function. An effective noninvasive evaluation tool will be beneficial to early identification and intervention. The study aims to identify sensitive predictors associated with SSC tendon tears and develop a dynamic nomogram to improve diagnostic performance.

Methods

From July 2016 to October 2021, 528 consecutive cases of patients who underwent shoulder arthroscopic surgery with preoperative shoulder MRI were retrospectively analyzed. The least absolute shrinkage and selection operator (LASSO) method was used to identify the sensitive factors associated with SSC tendon tears, which were then incorporated into the nomogram. The prediction performance of the nomogram was evaluated by concordance index (C index) and calibrated with 1,000 bootstrap samples combined with external validation of another cohort.

Results

The LASSO method showed that six items including coracohumeral distance (oblique sagittal plane), effusion (Y-face), effusion (subcoracoid), malposition of the long head tendon of the biceps, multiple posterosuperior rotator cuff tears, and considering SSC tendon tears on MRI (based on direct signs) were determined as sensitive predictors. The nomogram achieved a good C index of 0.878 (95% CI, 0.839–0.918) with a good agreement on the risk estimation of calibration plots. The areas under the receiver operator characteristic (ROC) curves of the two methods showed that dynamic nomograms had better prediction performance than MRI diagnosis based on direct signs (training set 0.878 vs. 0.707, validation set 0.890 vs. 0.704).

Conclusion

The study identified sensitive predictors associated with SSC tendon tears and first developed a web-based dynamic nomogram as a good supplementary evaluation tool for imaging diagnosis that could provide an individualized risk estimate with superior prediction performance, even in patients with small or partial tears.

Identifying key factors associated with subscapularis tendon tears and developing a risk prediction model to assist diagnosis

Background

There are still some challenges in diagnosing subscapularis (SSC) tendon tears as accurately as posterosuperior rotator cuff tears on MRI. The omission of SSC tendon tear can lead to muscle atrophy, fatty infiltration, and increased tearing accompanied by aggravated shoulder pain and loss of function. An effective non-invasive evaluation tool will be beneficial to early identification and intervention. The study aims to identify sensitive predictors associated with SSC tendon tears and develop a risk prediction model to assist in diagnosis.

Methods

Data on 660 patients who received shoulder arthroscopic surgery with preoperative shoulder MRI were collected retrospectively. Of these, patients with SSC tendon tears were defined as the SSC tear group, and patients with intact SSC tendon were enrolled in the non-SSC tear group. Logistic regression analysis was used to identify the key predictors of SSC tendon tears which were then incorporated into the nomogram.

Results

Among 22 candidate factors, five independent factors including coracohumeral distance (CHD, oblique sagittal plane) (OR, 0.75; 95%CI, [0.67–0.84]), fluid accumulation (Y-face) (OR, 2.29; 95%CI, [1.20–4.38]), long head of biceps tendon (LHB) dislocation/subluxation (OR, 3.62; 95%CI, [1.96–6.68]), number of posterosuperior (PS) rotator cuff tears (OR, 5.36; 95%CI, [3.12–9.22]), and MRI diagnosis (based on direct signs) (OR, 1.88; 95%CI, [1.06–3.32]) were identified as key predictors associated with SSC tendon tears. Incorporating these predictors, the nomogram achieved a good C index with a good agreement on the risk estimation of calibration plots. Higher total points of the nomogram were associated with a greater risk of SSC tendon tears.

Conclusion

When evaluating the severity of SSC tendon injury, the combination of reliable predictors can improve the sensitivity and diagnostic performance of MRI. This model provides an individualized probability of risk prediction, which is convenient for clinicians to identify patients at high risk for SSC tendon tears to avoid missed diagnosis.

The acromiohumeral distance in the MRI should not be used as a decision criterion to assess subacromial space width in shoulders with an intact rotator cuff

PURPOSE

The acromiohumeral distance is in practice often evaluated on MRI by radiologists and a reduction diagnosed as subacromial impingement. However, the acromiohumeral distance as indicator for a decentered glenohumeral joint is defined on a true AP radiograph with the patient standing or sitting. The present study therefore evaluated the influence of the patient position by comparing the acromiohumeral distance in both modalities in shoulders with an intact rotator cuff.

METHODS

On MRI images and true AP radiographs of patients > 20 and < 80 years with an intact rotator cuff the acromiohumeral distance was measured. The maximum cranio-caudal size of the glenoid was measured as a reference to allow a direct comparison of both modalities.

RESULTS

Two-hundred and thirty-four shoulders (mean patients age 45.8 ± 14.3 years) were included. The mean acromiohumeral distance/glenoid size ratio of all shoulders was significantly larger (P < 0.0001) on the MRI with 4.6 ± 1.0 in comparison to 4.1 ± 0.9 in the radiographs indicating a smaller acromiohumeral distance on the MRI. In absolute values, a mean acromiohumeral distance of 9.2 mm ± 1.8 on MRI in comparison to 10.4 mm ± 2.4 on the radiographs was calculated. Herewith, the acromiohumeral distance in the MRI was in the average 1.2 mm ± 2.1 (13%) smaller than the in corresponding radiographs (P < 0.0001).

CONCLUSION

The acromiohumeral distance is significantly smaller in the MRI in comparison to AP radiographs in shoulders with an intact rotator cuff and should not be used as a decision criterion on MRI to assess glenohumeral centering or subacromial space width.

LEVEL OF EVIDENCE

IV.

Increased acromiohumeral distance in a double-row arthroscopic rotator cuff surgery compared to a single-row surgery after 12 months

Background

Arthroscopic rotator cuff surgery is an effective treatment for rotator cuff tears with the considered use of double-row repair techniques becoming popular in the last decade. We aim to compare the effects of double- and single-row arthroscopic rotator cuff repairs (ARCR) on repair integrity (RI) and acromiohumeral distance (AHD).

Methods

In this observational study, we retrospectively identified 98 patients with degenerative rotator cuff tear treated with arthroscopic rotator cuff repair between 2016 and 2019. We excluded 22 patients with partial-thickness tears, 15 with associated subscapularis or SLAP tears, 13 with massive tears, and 5 patients lost to follow-up; we included 43 patients who had ARCR for full-thickness cuff tear and clinical, radiologic follow-up. Of these 43 patients, 23 are grouped as double-row repair group (DRG) and 20 as single-row repair group (SRG). A minimum of 12 months after the surgery, bilateral shoulder MRIs were obtained. Contralateral shoulders without asymptomatic rotator cuff tears served as a control group (CG). The operating surgeon and two other surgeons experienced in arthroscopy blindly measured the AHD and determined the RI at the control MRIs in all groups. Functional assessments relied on UCLA and qDASH Scores.

Results

The mean age was 57.89 (45–78) years, and the mean follow-up time was 28,65 (21–43) months. The mean AHD of the CG was 9.7 ± 0.96 mm, the preoperative AHD of DRG was 8.62 ± 1.45 mm, and SRG was 9.71 ± 0.95 mm. The postoperative mean AHD of DRG 9.61 ± 1.83 mm and SRG was 10.21 ± 1.97 mm. AHD differences between the preoperative and postoperative groups were significant (P=0.009). The increase of the AHD in the double-row group was significantly higher than the single-row group (P=0.004). There was a high correlation between the RI and DASH scores (P=0.005). RI did not correlate with the repair method (P=0.580).

Conclusion

Although double-row repairs can maintain greater AHD than single-row repairs in the clinical setting, this difference did not affect functional results. Regardless of the surgical intervention, functional results are favourable if RI is achieved.

Level of evidence

Level III, Retrospective Cohort Study

Radiographic greater tuberosity spurs and narrow acromiohumeral intervals are associated with advanced retraction of the supraspinatus tendon in patients with symptomatic rotator cuff tears

Background

Degenerative signs on shoulder radiographs, including spur formation and narrow acromiohumeral intervals (AHIs), have been recognized as indicative of atrophic and fat-infiltrated rotator cuff muscles. Past studies have demonstrated that patients with poor quality muscles are prone to retraction of the supraspinatus tendon and failure to repair. However, the association between radiographic signs and tendon retraction has never been elucidated in previous literature. The present study aimed to investigate the association between the degenerative signs on shoulder radiographs and the severity of supraspinatus retraction.

Methods

Images of 67 individuals, who had undergone an arthroscopic rotator cuff repair, were retrospectively reviewed. The greater tuberosity (GT) morphology, subacromial spur, AHI, and acromial thickness were evaluated on the radiographs, whereas the retraction of the supraspinatus tendon was assessed via an MRI in accordance with the Patte classification. Simple regression analyses between the radiographic signs and Patte stages were performed, and factors reaching statistical significance were then included in the multiple ordinal logistic regression. Statistically significant predictors from the multiple regression analysis were constructed into combinations, for which the sensitivity and specificity were calculated.

Results

The GT morphology (P = .004), AHI (P = .083), subacromial spur (P = .008), and age (P = .004) were associated with supraspinatus retraction in the simple regression analyses. These four parameters were incorporated into the multiple ordinal logistic regression, where the GT spur (adjusted odds ratio 8.63, 95% confidence interval 2.16-34.53, P = .002) and AHI (AOR 0.79, 95% CI 0.63-0.98, P = .032) were demonstrated to be predictive of the Patte stage of supraspinatus retraction. The acromial spur implied a higher risk of severe retraction although this finding was not statistically significant (AOR 2.89, 95% CI 0.90-9.29, P = .075). The presence of concurrent GT spur and narrow AHI was highly specific (sensitivity 27.3% / specificity 91.1%) for advanced supraspinatus retraction.

Conclusion

The presence of a radiographic GT spur, narrow AHI, and subacromial spur indicated advanced retraction of the supraspinatus tendon. When patients with clinical suspicion of rotator cuff tear present with combinations of these radiographic signs, a prompt MRI examination and a referral to a shoulder specialist are recommended.