The high failure rate after surgical repair of massive rotator cuff tears is a consistent problem. To evaluate the clinical and radiological outcomes of arthroscopic rotator cuff repair with bone marrow stimulation and patch augmentation in patients with massive rotator cuff tears. Cohort study; Level of evidence, 3. This study included 21 patients who underwent bone marrow stimulation and patch augmentation (group 1) and 54 patients who underwent conventional repair (group 2) for massive rotator cuff tears. Postoperative clinical outcomes were evaluated based on visual analog scale (VAS) for pain, simple shoulder test (SST), University of California, Los Angeles (UCLA), Constant, and American Shoulder and Elbow Surgeons (ASES) scores at baseline, 1 year postoperatively, and final follow-up. Anatomic outcomes were evaluated by using postoperative magnetic resonance imaging at 1 year after surgery. No significant differences in demographic characteristics and baseline data were observed between groups 1 and 2. Clinical symptoms were significantly improved at the final follow-up in both groups (P < .001). At the final follow-up, no significant differences were found in VAS pain (P = .676), SST (P = .598), UCLA (P = .100), Constant (P = .469), or ASES (P = .880) scores. However, the retear rate was lower in group 1 (4/21, 19.0%) than in group 2 (25/54, 46.3%) (P = .036), and the medial-row failure rate (type 2 retears) was much lower in group 1 (0/4, 0%) than in group 2 (18/25, 72.0%) (P = .014). Concomitant bone marrow stimulation and patch augmentation significantly reduced retear and medial-row failure rates in the arthroscopic repair of massive rotator cuff tears.

No study to date has directly evaluated rotator cuff repair results among smokers. To evaluate whether smoking affects healing after arthroscopic rotator cuff repair through propensity score matching (PSM). Cohort study; Level of evidence, 3. Among 249 patients who underwent arthroscopic repair of full-thickness rotator cuff tears, 34 current heavy smokers were selected with a smoking history >20 pack-years (mean ± SD pack-years, 33.91 ± 12.13). Characteristics between current heavy smokers and nonsmokers were compared. According to the PSM technique, 34 nonsmokers were selected after 1:1 matching for age, fatty infiltration, and tear size—the main prognostic factors of outcomes after rotator cuff repair. Each patient’s outcome evaluation was completed anatomically at a minimum of 6 months (magnetic resonance imaging or ultrasonography) and functionally at a minimum of 1 year (pain visual analog scale, range of motion, American Shoulder and Elbow Surgeons, Constant, University of California, Los Angeles, and Simple Shoulder Test scores), and every outcome was analyzed in the matched smoker and nonsmoker groups. Current heavy smokers had a higher incidence of male sex (P < .001), heavy manual work (P = .025), high bone density (P = .036), and poor tendinosis grade (P = .028). After adjustment for the confounding variables by PSM, the matched smoker group showed a significantly higher healing failure rate than the matched nonsmoker group (29.4% vs 5.9%, P = .023). However, we failed to detect significant differences in the functional outcomes between the matched groups (P > .05). Smoking affected healing failure after arthroscopic rotator cuff repair. Attention should be paid to smokers, especially current heavy smokers, in cases of rotator cuff repair surgery.

Rotator cuff tear (RCT) is one of the most common shoulder injuries. When diagnosing RCT, skilled orthopedists visually interpret magnetic resonance imaging (MRI) scan data. For automated and accurate diagnosis of RCT, we propose a full 3D convolutional neural network (CNN) based method using deep learning. This 3D CNN automatically diagnoses the presence or absence of an RCT, classifies the tear size, and provides 3D visualization of the tear location. To train the 3D CNN, the Voxception-ResNet (VRN) structure was used. This architecture uses 3D convolution filters, so it is advantageous in extracting information from 3D data compared with 2D-based CNNs or traditional diagnosis methods. MRI data from 2,124 patients were used to train and test the VRN-based 3D CNN. The network is trained to classify RCT into five classes (None, Partial, Small, Medium, Large-to-Massive). A 3D class activation map (CAM) was visualized by volume rendering to show the localization and size information of RCT in 3D. A comparative experiment was performed for the proposed method and clinical experts by using randomly selected 200 test set data, which had been separated from training set. The VRN-based 3D CNN outperformed orthopedists specialized in shoulder and general orthopedists in binary accuracy (92.5% vs. 76.4% and 68.2%), top-1 accuracy (69.0% vs. 45.8% and 30.5%), top-1±1 accuracy (87.5% vs. 79.8% and 71.0%), sensitivity (0.92 vs. 0.89 and 0.93), and specificity (0.86 vs. 0.61 and 0.26). The generated 3D CAM provided effective information regarding the 3D location and size of the tear. Given these results, the proposed method demonstrates the feasibility of artificial intelligence that can assist in clinical RCT diagnosis.

We aimed to estimate the interrelation between preoperative deltoid muscle status by measuring the 3-dimensional deltoid muscle volume and postoperative functional outcomes after reverse total shoulder arthroplasty(RTSA). Thirty-five patients who underwent RTSA participated in this study. All patients underwent preoperative magnetic resonance imaging(MRI) as well as pre- and postoperative radiography and various functional outcome evaluations at least 1 year. The primary outcome parameter was set as age- and sex-matched Constant scores. The 3-dimensional deltoid muscle model was generated using a medical image processing software and in-house code, and the deltoid muscle volume was calculated automatically. Various clinical and radiographic factors comprising the deltoid muscle volume adjusted for body mass index(BMI) were analyzed, and their interrelation with the outcome parameters was appraised using a multivariate analysis. As a result, all practical consequences considerably improved following surgery(all p<0.01). Overall, 20 and 15 indicated a higher and a lower practical consequence than the average, respectively, which was assessed by the matched Constant scores. The deltoid muscle volume adjusted for BMI(p = 0.009), absence of a subscapularis complete tear (p = 0.040), and greater change in acromion-deltoid tuberosity distance(p = 0.013) were associated with higher matched Constant scores. Multivariate analysis indicated that the deltoid muscle volume was the single independent prognostic factor for practical consequences(p = 0.011). In conclusion, the preoperative deltoid muscle volume significantly affected the functional outcome following RTSA in patients with cuff tear arthropathy or irreparable cuff tears. Therefore, more attention should be paid to patients with severe atrophied deltoid muscle who are at a high risk for poor practical consequences subsequent to RTSA.