Coracoid Process

All participants were seated during the examination, which was conducted by a physiatrist with 12 years of musculoskeletal US experience using a linear probe of 5–18 MHz (HI VISION Ascendus, Hitachi). The transducer was placed at the level of the coracoid process to evaluate the long head of the biceps tendon inside the bicipital groove. A pathological biceps peritendinous effusion was defined when the thickness of the anechoic fluid exceeded 1 mm. Biceps tendon tear was confirmed if the tendon was invisible or split. The shoulder was then externally rotated to expose the subscapularis tendon. The supraspinatus tendon was investigated in the Middleton position (Lee et al., 2016 (link)) with the hand placed over the ipsilateral iliac crest. The transducer was finally moved to the posterior shoulder in the horizontal plane slightly distal to the scapular spine for visualization of the infraspinatus tendon. The presence of visible gaps or a total absence of tendon tissue in the subacromial space served as the criteria for diagnosing rotator cuff tendon tears. Because the supraspinatus tendon has a large size, its lesions were classified as either full-thickness or partial-thickness tears. A full-thickness tear was indicated by an intra-tendinous gap extending through the entire thickness of the supraspinatus tendon. A partial-thickness tear was indicated by noticeable intra-tendinous cleavage(s) without the involvement of the entire thickness of the tendon. All scanning procedures were performed in accordance with the EURO-MUSCULUS/USPRM basic scanning protocol for the shoulder (Ozcakar et al., 2015 (link)).

Before topographic changes of the breast can be attributed to the insertion of the implant, the two scans of each breast must be perfectly aligned and overlaid. The “native” scan serves as a reference value (base). The Artec Studio 12 software (Artec, Luxembourg) automatically recognizes the corresponding areas and performs an alignment of both scans (“native” and “implanted”). Pre-operative markings (marked by permanent marker) on the jugulum, xyphoid and processus coracoideus served as an aid for alignment as they are fix anatomical landmarks. To guarantee that exactly the same area of the breast is measured within a patient, the scans were cut out en bloc after being perfectly aligned (Figs. 1, 3).Fig. 1

Alignment process using the anatomical landmarks. pt[1]: jugulum, pt[2]: xyphoid, pt[3]: right processus coracoideus, pt[4]: left processus coracoideus

The software’s “measure” tool is able to automatically calculate the mean distance in millimeters (mm) between two overlaid scans. The mean distance (“topographic shift”) between the “native” scan and the “implanted” scan indicated the influence of the implants on the topography of the breast in mm (Fig. 2).Fig. 2

Example of aligned Scans. The Topographic Shift is the mean distance, a between the “native” and “implanted” scans

To be able to quantify the effects of the implants on the breast surface more specifically, the breast was divided into the four commonly used quadrants (quadrant I: lateral, cranial; quadrant II: medial, cranial; quadrant III: medial, caudal; quadrant IV: lateral, caudal). Quadrants were defined on the “native” scan and fixed, so that all further measurements referred to this zoning. This allowed for a more detailed analysis of the influence of the implant on the different areas of the breast. The Topographic Shift was calculated for each quadrant of each breast (Fig. 3).Fig. 3

En bloc cut out of a right breast after alignment and cut out of quadrant I (red area A). The Topographic Shift of this quadrant is the mean distance, b between “native” and “implanted” scans

SD rats with a tooth in the differentiation stage at postnatal day 1.5 (P1.5) and secretory stage at postnatal day 3.5 (P3.5) were treated with inhalation anesthesia. We chose a 30% volume displacement rate per minute for CO₂ euthanasia of rodents and assisted by cervical dislocation. After euthanasia, we separate the head along the upper neck and then dissected the mandible along the corner of the mouth. The location of the mandibular molar tooth embryo was found at the mandibular ascending ramus below the coracoid process. Finally, tooth germs were dissected with ophthalmic tweezers under a stereo microscope (Olympus SZ61, Tokyo, Japan) and photographed with a microscope camera (MSHOT, Guangzhou, China) (Figures 1(a) and 1(b)). Samples were put in solidified carbon dioxide and immediately sent out for high-throughput sequencing (Genesky Biotechnologies Inc, Shanghai, China). We compared gene expression from the secretory stage to the differentiation stage, and each group was prepared as three individual samples. In addition, we prepared three additional rats per group for follow-up validation experiments. To ensure their development stage, pups outside the expected size range were excluded.

A T2-weighted MRI study's most lateral scan, in which the spine is in touch with the scapular body, is used to analyze the tangent sign [27 (link)]. The tangent sign is determined by drawing a line connecting the superior edge of the scapula's spine to the superior edge of the coracoid process. The muscle content should cross superior to the tangent line in healthy tissues. The superior border of the muscle must fell below the tangent line when the tangent sign is positive (showing significant atrophy). When the tangent line and the muscular belly of the supraspinatus intersect, the tangent sign is negative [27 (link)].
Oblique sagittal PD-weighted MRI scans were used to calculate the occupation ratio. The boundary of the supraspinatus fossa was closely followed as possible and the supraspinatus muscle outer rim was traced at the most lateral portion of the scapula's Y-view appearance in the oblique sagittal plane. The superior limit of the supraspinatus fossa was the distal clavicle. We used a formula to determine the Occupation Ratio as a measure of muscle volume using the technique provided by Thomazeau et al. [28 (link)]. Then, using the formula S1 / S2, the occupation ratio was determined. S1 represents the surface area of the supraspinatus muscle, and S2 represents the surface area of the entire supraspinatus fossa. If the ratio (stage 1) is between 1.00 and 0.60, the muscle is either normal or mildly atrophying. Significant atrophy is indicated by values between 0.60 and 0.40 at the stage 2 level. Stage 3 values below 0.40 indicate severe or moderate atrophy [28 (link)].
Using the Goutallier classification as modified by Fuchs et al. [29 (link)] on T1-weighted turbo spin-echo sequences, all muscles were evaluated at the most lateral scan on the sagittal view, where the spine was in contact with the scapular body [29 (link)] Two additional scans were taken into account for the infraspinatus and subscapularis tendons: an inferior scan at the lowest point of the glenohumeral joint and a superior scan at the level of the lateral attachment of the spine. Five stages are used to categorize changes: Stages 0 and 1 indicate no fat, stage 2 indicates there is more muscle than fat, stage 3 indicates there is an equal amount of fat and muscle, and stage 4 indicates there is more fat than muscle [30 (link)].
The MRI images were reassessed by the same author (an Orthopaedic and Trauma Surgery PhD Student), and the two results were compared. The author who performed the evaluation was blinded to clinical findings and to patient’s identities. Intraobserver and interobserver agreement was calculated. When there was only a 1-grade difference between the two assessments, the highest result was chosen. If an inconsistency (> 1 grade on the scoring system described below) existed between the 2 results, the slides were reassessed with the help of a board certified orthopaedist (Senior Author).