Introduction

Shoulder pain is a common symptom, often leading to medical attention in older adults. In the absence of trauma, it is usually caused by periarticular soft tissue injury and generally responds to conservative therapy. However, a tear of the rotator cuff can occur; this usually requires more aggressive management, and may result in surgical intervention. A rotator cuff tear (RCT) develops when shoulder tendons and bursae become inflamed as a result of recurrent microtrauma from repeated impingement of the cuff between the humeral tuberosity and the coracoid ligament, anterior acromium, and acromioclavicular joint.

Clinical Diagnosis of RCT

The clinical detection of such tears is important to guide further decision-making and to avoid diagnostic procedures that are often unnecessary, expensive, and/or painful. A recent article, summarized here, has described the use of bedside history and physical examination to help in the diagnosis of RCTs. The authors reviewed the charts of 448 patients with suspected RCT who were referred for arthrography over a 4-year period.

The definitive diagnosis of RCT is established using open or arthroscopic exploration of the shoulder. Double contrast arthrography has been used as the gold standard until recently, when it has been superseded by MRI. The former procedure has a sensitivity and specificity of 58% and 96% respectively, while MRI has a sensitivity and specificity of 71%-89% and 71%-97% respectively. Ultrasonography has a sensitivity in the range of 83%-98% and specificity of 57%-98%. (Sensitivity means the probability that a person not having the disorder will be correctly identified by the test, while specificity is the probability that a person having the disorder will be correctly identified by the test.) Thus MRI and ultrasonography have slightly greater diagnostic discrimination than arthrography. However, for this study, RCT was defined as either partial or complete tear of the rotator cuff demonstrated by arthrogram. The authors described several physical examination tests for rotator cuff integrity; we discuss only the most useful ones in this summary.

Results of Analyses

Of the 448 patients analyzed, 301 (67%) had RCT on arthrography. On univariate analysis, the clinical findings most closely associated with RCT were evidence of infra- or supraspinatus atrophy (p<0.001), weakness of either elevation or external rotation (p<0.001), presence of an arc of pain (p<0.004), and a positive impingement sign (p=0.01). (After the externally rotated arm is passively elevated to a level alongside the ear, the impingement sign is elicited by having the patient then internally rotate the arm; a positive sign is one that produces significant pain.) Using logistic regression, three factors had a significant association (p<0.05) with RCT: weakness on external rotation, age > 65 years, and the presence of night pain.

RCT Diagnostic Score

Based on these results, the authors constructed a model with a 5-point scoring system. Using the three most important features of RCT expression - assignment of a score of 2 for weakness on external rotation, a score of 2 for age > 65 years, and a score of 1 for night pain, they determined the probability of RCT, and tested it on half the sample. A score of 4 was optimal in differentiating patients with abnormal from those with normal arthrograms. The authors proposed the following interpretation of scores:

• Further diagnostic testing (MRI or arthrography) in subjects with a score of 4 or more should be reserved for patients who are surgical candidates and require definition of the rotator cuff anatomy in planning the operative approach.
• For those with a moderate probability of RCT (2-3 points), confirmatory testing may become necessary because uncertainty is often greatest in this group.
• Patients with a score of 0-1 are unlikely to have RCT and do not require further diagnostic testing unless they fail conservative therapy.

This scoring method can be used to avoid expensive and/or uncomfortable diagnostic procedures, and can indicate when an appropriate consultation is needed if surgical intervention seems likely.