SHOULDER PAIN

Lisa Sanders, M.D.

WEEK 14

Learning Objectives:

  1. Describe the differential diagnosis for shoulder pain
  2. Be able to examine the shoulder for the presence of common shoulder injuries
  3. Know when additional studies are indicated and which studies would be most effective at defining the suspected injury
  4. Apply the therapeutic options for different etiologies of shoulder pain

CASE ONE:

Mr. Omy Armegedon-Paynes is a 35-year-old construction worker who presents with a four-week history of right shoulder pain. The pain started right after the company picnic where he pitched the best game his team had ever played – the workers versus the management team. He’d really been on a roll that day – pitching a near no-hitter (only one guy ever came even close to the ball that day), and the next morning he could hardly move his arm. At first they put him on light duty, but the pain actually got worse even doing that, and he had to take last week off. After a week of rest, the shoulder wasn’t any worse but it wasn’t any better either, and now it keeps him awake at night. He has taken aspirin off and on, and that helps a little, but not enough to let him go back to work. He is concerned about missing any more days on the job.

Questions:

  1. What is in your differential?

The pain is clearly secondary to the trauma of his baseball game, but it could be a number of different entities. The rotator cuff is the most common site of injury. This includes a rotator cuff tendonitis, subacromial bursitis, impingement syndrome, or a rotator cuff tear, either full or partial. Given the circumstances of the injury, it might also be a tear of the labrum, the cartilage that surrounds the glenoid fossa. The most common of the labrum injuries is known as a SLAP (superior labrum anterior-posterior lesion, where you get a tear at the superior aspect the labrum that goes from the anterior to posterior aspect of this hardy structure.)

  1. What will you be looking for in the physical exam? Demonstrate the tests you would perform on the patient on one of your colleagues.

You should look for evidence of muscle wasting – particularly the superspinatus muscle, which would suggest a rotator cuff tear. Test for range of motion, both passive and active. The so-called Painful Arc can help make some distinctions: If the patient has pain between 60-120 degrees, that suggests an impingement syndrome; if the pain is worse between 120-180, it suggests acromio-clavicular joint involvement. The classic test for a full tear of the rotator cuff is the empty beer can test, illustrated in the Journal of Family Practice. This tests the integrity of the superspinatus, the most common location of rotator cuff tears. A positive empty can test is most reliable in those over the age of 50. There is a higher rate of false positives in younger patients. It is important to make sure that the patient doesn’t have a full thickness tear because that injury requires surgery. The Lift test, also illustrated in the Journal of Family Practice article, can look for tears in the subscapularis tendon. Tests positive for rotator cuff tears do not rule out injuries to the Labrum. The three provocative tests illustrated in the JAMA article, the Speed test, the O’Brien test, and the Jobe relocation test can be used to identify a SLAP injury. Residents should practice these tests on each other.

CASE ONE CONTINUED:

The patient has full range of motion but has pain when his arm is actively abducted above 60 degrees. Passive motion is painless through the entire arc. He has a negative empty can test. The Speed, O’Brien, and Jobe Relocation tests are negative as well.

  1. How likely is it that this patient has a full thickness rotator cuff tear?

Not very. Using the algorithm in the JFP article, the likelihood that he has a full thickness tear is about 5%.

  1. Would you order any further tests on the patient? If so, which tests? If not, how would you treat this patient?

This patient has a subacromial bursitis, a rotator cuff tendonitis, or a partial thickness rotator cuff tear. Conservative management is the mainstay of treatment. This would include NSAIDS and physical therapy (Ultrasound, heat or ice are the treatments most recommended) Rotator cuff strengthening exercises have been shown to improve symptoms in athletes and may be useful. Glucosamine and chondroitin have been shown to be useful in the treatment of osteoarthritis (although studies in shoulders have not been done) and so may be helpful to this patient if he had given a history suggestive of OA (history of previous trauma or previous episodes of similar shoulder pain). A six-week trial is usually enough to show whether this therapy will be sufficient.

CASE ONE CONTINUED:

After six weeks the patient continues to have pain. A friend had arthroscopic surgery in his knee and that worked wonders, and the patient wonders if that would be helpful for him.

  1. What would you do next?

Your next step would be to order an MRI on the patient to better define the cause of his pain. If he had a partial tear, the author of the JAMA study suggests that arthroscopic evaluation and surgery may be appropriate. If there is no evidence of a rotator cuff tear and an impingement syndrome or bursitis is the most likely cause of the pain, then steroid injections may be helpful. The JAMA author recommends using this technique no more than three times over three months. This will relieve pain and allow more time for healing. After that, he recommends discussing arthroscopy and arthroscopic surgery for pain that doesn’t resolve with time.

CASE TWO:

Ms. Acromia L. Bursidus is a 55-year-old woman with a two-month history of left shoulder pain. She has a history of diabetes and hypertension, both well-controlled on oral medications. When it started, the pain had been mild, 2-3 out of 10, and she treated it with rest and aspirin, but it has worsened over the past few weeks so that almost any movement is painful. Getting dressed is difficult, and she has trouble brushing her hair and teeth because of the pain and stiffness. She denies any trauma. She’s had trouble with this shoulder in the past, but it was never this bad.

  1. What are the most likely causes of her pain?

This could be osteoarthritis of the shoulder given her age and history of previous episodes of pain. It could also be an impingement syndrome caused by osteophytes compressing the rotator cuff. A third possibility is adhesive capsulitis, the so-called Frozen Shoulder.

CASE TWO CONTINUED:

On physical exam, she has pain with passive and active movement of her arm, and her active movement is quite limited because of pain and stiffness. She is unable to cooperate with any of the provocative tests to look for evidence of a rotator cuff tear or labrum tear.

  1. What studies if any would you do to diagnose the cause of her pain? What therapy would you recommend?

Her physical exam is very suggestive of adhesive capsulitis. If you need confirmation, you can get an MRI, which would show the typical pathology. Her diabetes puts her at higher risk of developing this problem. Patients with diabetes are five times more likely to develop adhesive capsulitis than are patients without diabetes and are much more likely to develop severe symptoms (prevalence of 10% in patients with diabetes vs. 2% in the general population). The old teaching about adhesive capsulitis was that it lasts one to three years whether you treat it or not. Physical therapy is not effective. The JAMA article suggests that steroid injections into the capsule may speed healing. Again, he recommends three injections at monthly intervals. If the patient remains symptomatic, the options of either arthroscopic release or manipulation under anesthesia should be offered. If the pain and restriction last more than one year, the author suggests that patients with diabetes, like this patient, are unlikely to regain full range of motion without arthroscopic release. Even in individuals without diabetes, arthroscopic release can dramatically shorten the duration of the symptoms.

References:

  1. Burkhart, Stephen S. A 26-year-old woman with shoulder pain, JAMA. 2000; 284 (12): 1559-67.
  2. Stevenson, J. Herbert MD; Trojian, Thomas MD. Evaluation of shoulder pain. Journal of Family Practice.2002; 51(7):605-611. (This article is good for different examination techniques and has great tables. Figure 5 is an algorithm for calculating likelihood of a rotator cuff tear based on history, physical exam and test findings.)