Shoulder MRI

Shoulder Disorders & Spinal Adjustments

Of the lectures that I commonly give on MRI, there is one subject that doctors seem to be most interested in. Invariably, when I am scheduled to give a lecture on this topic, I seem to get the best turnout from them. That subject would be shoulder MRI.

I am not really sure why the topic of shoulder MRI tends to be so popular, but I have a few ideas. Perhaps it is because shoulder problems tend to be a bit more difficult to accurately diagnose. Although that statement may seem misleading, the majority of shoulder pain is due to rotator cuff problems, but the difficulty often lies in trying to distinguish rotator cuff tendinosis from cuff tear. MRI is ideally suited for making the distinction between the two. Another reason may be that plain film shoulder x-rays often appear normal, and thus, MRI provides a window into an area that seems otherwise invisible to imaging. I am sure there are other reasons as well, but in any event, there always appears to be much interest in shoulder MRI Therefore in this column, I will attempt to give some background on normal shoulder MRI, and in the next subsequent column, I will begin to describe shoulder pathology.



Shoulder MRI is sometimes thought of as difficult to interpret, and there are a few reasons for this. The first is that the shoulder has some very small components within it, but the field of view used for shoulder imaging is relatively large. Thus, small structures can be more difficult to visualize. Secondly, there is a fairly wide range of normal in the shoulder. There are normal variants that should not be mistaken for pathology. A third reason may be a result of less than perfect positioning of the patient during imaging, as that can alter the relationship of one structure to another and make interpretation more complicated. Nevertheless, MRI tends to give us a very nice picture of the rotator cuff tendon, and as previously stated, that is the source of a large percentage of pain in the shoulder.

Why is MRI such a useful study for rotator cuff pathology? This modality has the capability of discerning tears in the cuff and allowing us to differentiate cuff tear from tendinosis. In addition, we can determine a partial tear versus a full thickness tear. Even intrasubstance tears, which cannot be seen with arthroscopy, can be seen on MRI. MRI is preferred over CT for shoulder imaging because of its superior evaluation of soft tissue. Also, with MRI we can image a structure directly in any plane we choose, which cannot be done with CT.

Let me begin by describing the typical imaging sequences used in shoulder MRI’s.

  • Angled coronal T1, T2, and fat-suppressed or STIR images are commonly employed (fig. 1). The cuts are angled approximately 30-45 degrees off the true coronal plane in order for them to be oriented parallel to the supraspinatus muscle and rotator cuff tendon. This allows good visualization of the rotator cuff tendon and supraspinatus muscle.
  • Gradient echo is usually the sequence of choice for the axial plane (fig. 2). This sequence allows a superior evaluation of the glenoid labrum. Proton density and T2 sequences are typically employed for the sagittal plane images (fig. 3).
  • Sagittal plane images allow good evaluation on the rotator cuff in a plane oriented approximately 90 degrees to the coronal plane and can confirm tendon pathology or supply additional information about the cuff.

Fat suppression is a useful adjunct in most MRI musculoskeletal examinations. The fat signal is suppressed in this sequence and thus it appears hypointense or dark on the image. Water, in contrast, appears very bright on fat suppression. Any tissues with increased water content will brighten on the fat-suppressed images because most body tissues contain some degree of fat. This can be extremely useful when we suspect subtle injuries such an osseous contusion. These injuries may be difficult to visualize on standard spin echo images but become very obvious on the fat-suppressed image. STIR images are a type of fat suppression and are generally found in studies performed on the lower field strength open MRI scanners.


It is good to become familiar with the supraspinatus muscle/tendon’s normal appearance because it is the component of the rotator cuff most likely to incur tendinosis or tear. Normal tendons are hypointense (dark) on all MRI sequences, and a healthy supraspinatus tendon should appear dark as well. The tendon will appear to begin to merge with the supraspinatus muscle just medial to the apex of the humeral head. This section is termed the muscolotendinous junction. The supraspinatus muscle proper is intermediate to dark gray in appearance depending on the imaging sequence utilized. A well-developed supraspinatus muscle will fill the supraspinous fossa almost completely. If there is no proliferation at the inferior aspect of the AC joint and no spur formation at the insertion of the coracoacromial ligament, there will be a clear suprahumeral space for the tendon to pass through and thus no structural impingement will be encountered. Figures 1 and 3 demonstrate normal low signal intensity supraspinatus tendon as seen on T2 weighted images.

In the June column, we will look at how tendinosis and tendon tear appear on MRI and cover the elements that should be included in a comprehensive report of a shoulder MRI.

About Author

Douglas Gregerson, DC, DACBR

Dr. Gregerson graduated from the National College of Chiropractic in 1983 and maintained a private practice in Chiropractic for 14 years before entering the NCC radiology residency in 1997. He was the recipient of the Kenneth Yochum Memorial Scholarship for radiology in 1998. Dr. Gregerson received his DACBR certification in 1998 and is currently the chief radiologist for Gregerson Radiology Consultants. He is a former faculty member of the National University of Health Sciences where he taught radiology and was a staff radiologist as well as a senior staff clinician. He has authored journal articles for the Journal of the American Chiropractic Association, Journal of Manipulative and Physiologic Therapeutics, Topics in Diagnostic Radiology and Advanced Imaging, Spine and the Journal of the Musculoskeletal System. Dr. Gregerson has also been involved in various research projects at the university and currently lectures for many imaging centers and various chiropractic groups.

Corporate Club

Article Categories