Point/Counterpoint
Shoulder Ultrasound vs MRI for Rotator Cuff
Pathology
CASE SCENARIO
A 36-year-old, right-handed husband of a physician presented to a physiatrist with a
2-week history of right shoulder pain that began at the end of a downhill ski trip. He re-
ported that the pain began after he traversed a flat ski trail using ski poles for propulsion.
The pain was worsened with activities that required shoulder abduction and internal
rotation. Pain was located over the deltoid region, and he denied complaints of catching or
clicking. He felt generally weak in the shoulder. No other associated symptoms were pres-
ent such as numbness or tingling. On physical examination, passive range of motion of both
shoulders was symmetric and without limitations. Active range of motion of the left
shoulder was normal. Active range of motion of the right shoulder was limited by pain with
abduction at 90° and internal rotation at 10°. Active external rotation with the elbow held at
the side of the thorax was limited to 10° because of pain. Provocation tests, including load
and shift, active compression test, and apprehension test, were not painful bilaterally. Im-
pingement test of the right shoulder was painful. Manual muscle testing of the supraspina-
tus was 4+/5 and associated with pain. All other muscles in the upper limb were 5/5 in
strength. Upper limb muscle stretch reflexes, sensation, and cervical spine range of motion
were normal. Radiographs of the right shoulder (anteroposterior, true anteroposterior, axil-
lary, and Y-views) were normal. Ice, relative rest, acetaminophen, and physical therapy were
prescribed. A diagnostic ultrasound (US) was ordered but was not available for 3 weeks.
A week after the initial presentation, the patient complained of intractable night pain
despite using opioids. The patient’s wife pointed out that he failed to mention in the initial
presentation that he sustained a hard fall on the ski trip, although the patient still did not
relate the fall to the onset of pain. An earlier appointment for an US was still unavailable and
therefore a noncontrast magnetic resonance imaging (MRI) was ordered, which suggested a
bursal side partial thickness tear and superimposed severe supraspinatus tendinopathy as
well as subacromial-subdeltoid bursitis (Figure 1).
A subacromial lidocaine and steroid injection was then administered for pain control
and physical therapy was restarted. A week later, the patient called to cancel the US because
he was feeling better, although he still did not describe full shoulder range of motion or
normal strength. The physiatrist recommended keeping the US appointment in anticipa-
tion of using US to follow the degree of healing of the presumed incomplete rotator cuff over
time. A week later, the US was performed and a complete rotator cuff tear (vs extensive
partial-thickness tear extending into the substance of the cuff) was found. The US further
showed a markedly hypoechoic and heterogeneous supraspinatus tendon, which was
interpreted as representing intratendinous hemorrhage rather than severe tendinopathy
(Figures 2, 3).
The patient was referred to a shoulder surgeon and underwent shoulder arthroscopy. A
50% partial surface to partial undersurface articular-sided tear of the supraspinatus tendon
was noted. This was debrided from the articular side. On the subacromial side, there was a
partial-thickness tear of the bursal side with only a bridge of 10% of the tendon connected.
Because of this, the tear was completed and primary repair with subacromial decompres-
sion was performed. The patient is now 10 months postsurgery with full pain-free range of
motion and normal upper limb strength. He and his wife are now debating whether he will
participate in an upcoming ski trip.
If you were managing this patient, which initial imaging study would you have ordered?
How do US and MRI compare in assessing for complete and incomplete rotator cuff tears?
What are the values and shortcomings of each of these tests?
Guest Discussants:
Sharlene A. Teefey, MD
Diagnostic Ultrasound, Abdominal Imaging
Section, Washington University School of
Medicine, St Louis, MO
Disclosure: nothing to disclose
Brian Petersen, MD
Department of Radiology and Orthopedics,
Musculoskeletal Radiology, University of
Colorado Denver School of Medicine, Aurora,
CO
Disclosure: nothing to disclose
Senior Editor:
Heidi Prather, DO
Section of Physical Medicine and
Rehabilitation, Department of Orthopedic
Surgery, Washington University School of
Medicine, 660 South Euclid Ave, Campus
Box 8233, St Louis, MO 63110. Address
correspondence to: H.P.; e-mail:
pratherh@wudosis.wustl.edu
Disclosure: nothing to disclose
PM&R © 2009 by the American Academy of Physical Medicine and Rehabilitation
1934-1482/09/$36.00 Vol. 1, 490-495, May 2009
Printed in U.S.A. DOI: 10.1016/j.pmrj.2009.03.013
490