Shoulder Imaging: Systematic Interpretation

A full-thickness rotator cuff tear on MRI shows fluid signal intensity (bright on T2/STIR) extending completely through the tendon from the bursal surface to the articular surface. This creates a gap in the tendon that fills with joint fluid. I assess severity by measuring the tear in two dimensions: anteroposterior extent (along the tendon) and medial-lateral retraction. I classify tears as small (less than 1cm), medium (1-3cm), large (3-5cm), or massive (greater than 5cm or involving 2+ tendons). I also assess for retraction - the distance the torn tendon edge has pulled back from the footprint. Additionally, I evaluate the muscle quality using the Goutallier classification for fatty infiltration (stages 0-4) and assess for atrophy using the tangent sign. Goutallier stage 3-4 or significant atrophy predicts poor repair outcomes and may influence surgical decision-making. I note which tendons are involved - supraspinatus is most common, but assess infraspinatus, subscapularis, and long head of biceps as well.

For recurrent anterior instability, I order plain radiographs first including true AP, axillary, and Stryker notch or West Point views to assess for bony Bankart and Hill-Sachs lesions. MR arthrography is my next choice as it has superior sensitivity for labral tears compared to non-contrast MRI - I look for the Bankart lesion (anteroinferior labral tear at 3-6 o'clock position with contrast undercutting), and assess for variants like ALPSA or HAGL lesions. For bone loss quantification, I order CT with 3D reconstruction. I use the best-fit circle method on an en face view of the glenoid - drawing a circle on the inferior 2/3 and calculating the percentage deficiency. Glenoid bone loss greater than 20-25% typically requires bone augmentation such as Latarjet rather than soft tissue repair alone. For the Hill-Sachs lesion, I assess depth, width, and engagement risk - the glenoid track concept helps determine if it's engaging. I would also note any off-track Hill-Sachs which combined with glenoid bone loss significantly increases recurrence risk with soft tissue repair alone.

A shoulder trauma series comprises three views: the true AP (Grashey), axillary, and scapular Y view. The true AP or Grashey view is taken with the patient rotated 40 degrees so the X-ray beam is perpendicular to the scapular plane - this provides an unobstructed view of the glenohumeral joint space without overlap, allowing assessment of arthritis, joint congruity, and identification of Hill-Sachs lesions on the humeral head. The axillary view is absolutely essential in trauma and is obtained with the beam directed through the axilla - this is the only reliable view to confirm glenohumeral joint reduction and to detect posterior dislocations, which are easily missed on AP views alone. If the patient cannot abduct, a Velpeau axillary view can be obtained. The scapular Y view provides a true lateral of the scapula showing the relationship of the humeral head to the glenoid fossa - it helps identify dislocation direction (anterior or posterior to the Y) and shows acromion morphology. For instability workup, I would add AP internal and external rotation views, and potentially a West Point or Stryker notch view to better visualize Hill-Sachs lesions.