Copyright © 2006 by The Journal of Bone and Joint Surgery, Inc.
Commentary & Perspective
Commentary & Perspective by
Thomas J. Gill, MD*,
Department of Orthopedic Surgery, Massachusetts General Hospital, Boston, MA
The diagnosis of traumatic anterior shoulder instability can be complex. While some patients present with a history of instability in which a dislocation occurred that required a physician-aided reduction, this is not generally the case. A chief complaint of pain is often more common than a chief complaint of instability.
To confirm the diagnosis of anterior instability, a variety of physical examination and imaging tests have been advocated. Far too often, clinicians rely on magnetic resonance imaging or computed tomography scans for the primary assessment of shoulder disorders. Unfortunately, there is a high prevalence of "pathologic findings" on magnetic resonance imaging scans of even asymptomatic shoulders; these findings include partial-thickness rotator cuff tears or tendinopathy, labral tears, biceps disorders, or acromioclavicular joint arthrosis. As a result, the physical examination of the shoulder, perhaps more than in any other joint, plays a critical role in making an accurate diagnosis.
In their study, "Clinical Assessment of Three Common Tests for Traumatic Anterior Shoulder Instability," Farber et al. seek to identify whether the anterior apprehension test, relocation test and/or anterior drawer test are predictable tools to diagnose traumatic anterior shoulder instability. The criterion used to make the diagnosis of traumatic instability was either radiographic documentation of an anterior shoulder dislocation after trauma or demonstration of a Hill-Sachs lesion, a Bankart lesion, or a humeral avulsion of the glenohumeral ligament at the time of arthroscopy. Forty-six of 363 patients who underwent shoulder arthroscopy within a five-year period served as the study group, and the remainder served as the control group. The control group included patients with atraumatic anterior instability and multidirectional instability, which may have made comparison with the control group somewhat more difficult.
The apprehension test was performed with the patient standing, and was deemed positive if the patient became apprehensive about having an episode of instability. The relocation test was considered positive if a posteriorly directed force on the humeral head of the supine patient relieved the sense of apprehension. The anterior drawer test was performed with the patient supine, with the arm abducted 60° to 80° with neutral rotation, and with the amount of humeral head translation over the glenoid recorded. If a patient could not relax enough for this test to be performed, that fact was noted.
The apprehension test was found to have a sensitivity of less than 81%, while the specificity depended on whether pain or true apprehension was used as an end point. The same was true for the relocation test. Given that patients with superior labral tears, rotator cuff tears, subacromial impingement, or biceps tendonitis often have pain with the apprehension test, the findings of the authors confirm the teachings of Rowe and Zarins that only a sensation of true apprehension should be taken as a positive result. If true apprehension is present, it has been my experience that the apprehension test is even more sensitive and specific than reported in this study, and it remains my physical examination test of choice. In addition, although the authors state that they found no difference in the outcome of the test between standing or supine positioning of the patients, I prefer to perform this test with the patient supine, as it is easier to get the patient to relax and allow the test to be performed more accurately. Muscle-guarding often occurs when the patient is standing, which can give a false-negative result.
The anterior drawer test in this study was performed successfully for 87% of the patients. The remainder of the patients could not relax enough for a reliable test, due to pain or fear of instability. The usefulness of the test seemed to be related to the degree of laxity. Patients with grade-I laxity (translated to the anterior glenoid rim) had a much lower reproduction of instability symptoms with the test than did patients with grade-II laxity (humeral head can be translated over the rim). Forty-three percent of patients had more laxity when they were under anesthesia than when they were awake. Thus, despite the conclusion presented in the study, it does appear that performing an examination under anesthesia is beneficial for almost half of the patients who underwent shoulder stabilization. I typically try to tailor the degree of capsular shift and/or tightening during surgery to several parameters: the degree of shoulder laxity under anesthesia as compared with the contralateral shoulder, the history and chronicity of instability, and the ease of dislocation or subluxation. Thus, although the anterior drawer test is not always diagnostic for traumatic anterior shoulder instability, the results of that test can give valuable confirmatory information and can be helpful when deciding the type or degree of stabilization to be performed.
There are modifications to the anterior drawer test that are also useful when evaluating for anterior instability. If tests such as the anterior drawer test or load-and-shift test are only performed at higher degrees of abduction, the contributions of the superior or middle glenohumeral ligaments to the diagnosis of instability may not be fully appreciated. Performing the test with the arm in 20° of abduction, 45° of abduction, and 70° to 90° of abduction can give additional information in this regard.
This study dealt solely with traumatic anterior shoulder instability. In general, the diagnosis of anterior instability is more accurately made in this group than in atraumatic groups, such as overhead athletes, patients with a family history of the disorder, or patients with multidirectional instability, for which imaging is often not diagnostic and history and physical examination are paramount. It would be of interest to expand this study to include this group of patients and to compare the results to the traumatic group.
The authors conclude that if there is no radiographic evidence of a dislocation, the results of their study do not allow them to rely on physical examination tests alone. They recommend a diagnostic arthroscopy for any patient with suspected anterior instability of the shoulder. While I agree that physical examination is not perfect, I believe that in certain circumstances, physical examination can be more accurate than arthroscopy alone for the diagnosis of symptomatic anterior instability. Not all patients with fraying of the anterior labrum, or even small detachments, have symptomatic instability. For example, it is not uncommon to see some anterior labral fraying or minor tearing when repairing a superior labral tear. According to their criteria, do the authors routinely add a Bankart repair in this setting? I have taken the approach that in the absence of apprehension preoperatively or grade 2 or 3 laxity on anterior drawer testing under anesthesia, I typically repair only the superior labrum. The same is true with regard to small Hill-Sachs lesions. A primary reliance on arthroscopy means that patients with capsular laxity or interval lesions will often be missed, whereas they are usually identified on physical examination.
In summary, the diagnosis of traumatic anterior shoulder instability can be complex. This realization only serves to strengthen the classic teaching that in order to make an accurate medical diagnosis, the history must be correlated with the physical examination, which in turn must be correlated with the imaging findings. All of these data must then be correlated with the pathoanatomy, or lack thereof, at the time of surgery. By keeping in mind that there is no absolute pathognomonic or "essential finding," the diagnosis of traumatic anterior instability can be made more accurately.
*The author did not receive grants or outside funding in support of his research for or preparation of this manuscript. He did not receive payments or other benefits or a commitment or agreement to provide such benefits from a commercial entity. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the author is affiliated or associated.
|
