Muscle Weakness in the Empty and Full Can Tests Cannot Differentiate Rotator Cuff Tear from Cervical Spondylotic Amyotrophy: Pain Provocation is a Useful Finding

1. INTRODUCTION

The causes of difficulty in shoulder elevation are often difficult to identify, as this complaint might originate from various shoulder and cervical spine disorders. Additionally, the age at occurrence of pathologies of the shoulder and cervical spine is similar, and these specifically occur in the aging population [1, 2] and are often misdiagnosed. Among these, rotator cuff tears and cervical spondylotic amyotrophy (CSA) are often confused, as one of the main symptoms in both is difficulty in shoulder elevation.

CSA is classified as the proximal or distal type [3-5]. The clinical characteristics of the proximal type of CSA include muscle atrophy in the upper extremities, the absence of or insignificant sensory deficits and lower-extremity symptoms, and weakness of the deltoid and biceps muscles, which often causes dropped shoulder syndrome [3-7].

While one of the major symptoms of rotator cuff tears is difficulty in shoulder elevation, rotator cuff tears usually involve the supraspinatus tendon [8, 9]; therefore, the diagnosis of a rotator cuff tear is mainly based on the diagnosis of a torn supraspinatus tendon. The empty and full can tests are frequently used for the clinical diagnosis of rotator cuff tears, which elicit weakness or pain secondary to a torn supraspinatus tendon [10, 11]. Some authors reported the clinical usefulness of the empty and full can tests for determining the presence of a torn supraspinatus tendon [12-14].

Usually, conservative therapy is effective for CSA, but surgery is sometimes needed if conservative therapy is ineffective. The duration of symptoms is one of the risk factors for poor outcomes after surgical treatment [15, 16], so early diagnosis is very important.

If problems with shoulder elevation due to a rotator cuff tear are suspected, the empty and full can tests are performed. We hypothesized that these tests could also be used to differentiate rotator cuff tears from CSA. Thus, the aim of the present study was to investigate whether the empty and full can tests could be used to differentiate rotator cuff tears from CSA.

2. MATERIALS AND METHODS

3. RESULTS

4. DISCUSSION

In the present study, the empty and full can tests were performed in both rotator cuff tear and CSA patients to investigate whether these two tests could differentiate rotator cuff tears from CSA. Both tests showed low specificity and PPV for muscle weakness; therefore, these findings might show a high positive rate in patients with CSA. Pain provocation showed high specificity and PPV in both tests. In addition, pain in the empty can test showed high sensitivity (88.9%). Similar to previous reports, the full can test was more accurate than was the empty can test for the clinical diagnosis of rotator cuff tears; this was because avoiding of the risk of mechanical impingement can lead to pain provocation [12-14, 18]. However, pain provocation was useful in the differential diagnosis of rotator cuff tear from CSA, and is an important finding in the present study.

Rotator cuff tears usually involve the supraspinatus tendon [8, 9]; therefore, the diagnosis of rotator cuff tear is mainly based on the diagnosis of a torn supraspinatus tendon. In the present study, the supraspinatus tendon was involved in all 27 cases. Jobe et al. reported that a test to assess the ability of the affected shoulder to maintain the arm at 90º elevation in the scapular plane and in full internal rotation elicits weakness or pain secondary to a torn supraspinatus tendon (the empty can test) [10]. Subsequently, Kelly et al. proposed a new test to assess the function of the supraspinatus tendon to maintain the arm at 90º elevation in the scapular plane and at 45º external rotation and that elicited weakness or pain, and named this the full can test [11]. These two physical examinations are frequently used for the diagnosis of rotator cuff tears. Some authors reported the clinical usefulness of the empty and full can tests for determining the presence of a torn supraspinatus tendon. Itoi et al. reported that the full can test may be more beneficial in the clinical setting, considering the pain provocation [13]. Further, another author reported that pain provocation was more frequent during the performance of the empty can test than it was during the full can test, and the full can test more accurately evaluated muscle weakness [12, 14]. Graichen et al. reported that a three-dimensional magnetic resonance imaging study demonstrated that internally rotated arm abduction decreases the size of the subacromial space [18]. This avoids the risk of mechanical impingement that leads to pain, and therefore, the full can test is more accurate than the empty can test for the clinical diagnosis of rotator cuff tears.

One of the major symptoms of rotator cuff tear is difficulty in shoulder elevation. This complaint might originate from cervical pathologies. The chief complaint of patients with cervical pathologies, especially CSA, is also difficulty in shoulder elevation [3-7]. In addition, the predilection age for these two conditions is similar, especially in the aging population [1, 2]. Therefore, CSA is often misdiagnosed as rotator cuff tear. Conservative therapy is usually effective in CSA; however, if it is ineffective, surgery is needed [15, 16]. Tauchi et al. reported that early surgery was recommended for CSA patients in whom conservative treatment was unsuccessful, based on clinical status before surgery, including symptom duration [15]. Uchida et al. reported that a long preoperative period strongly correlated with less improvement in muscle power [16]. Early diagnosis is therefore vital in CSA.

Diagnosis of rotator cuff tears is comprehensively based on physical findings, disease course, and imaging study, such as with magnetic resonance imaging [12, 19, 20]. However, imaging generally causes a delay in diagnosis and entails a high cost, while physical findings are convenient and enable rapid diagnosis. Therefore, we have identified physical findings (i.e., those of empty and full can tests) that would be useful for differentiating between rotator cuff tear and CSA to prevent misdiagnosing CSA as rotator cuff tear, especially on initial diagnosis.

Our study has several limitations. First, we did not evaluate the coexistence of rotator cuff tears and CSA, i.e., patients with rotator cuff tear did not undergo examinations for CSA and patients with CSA did not undergo examinations for rotator cuff tears. Second, we did not evaluate inter- and intra-observer errors. The examiner was only one person who specialized in spine surgery. Third, we did not conduct a power analysis to determine the appropriate sample size before conducting the present study. However, as there was a significant difference for each physical finding, a type 2 error (caused by a small sample size) might not have occurred. Fourth, the examiner was not blinded from the gold standard diagnosis when performing the test. To eliminate these issues, we aim to conduct a similar prospective study in a large cohort.

We suggest that not only evaluation of muscle weakness but also pain provocation should be conducted during the empty and full can tests for the diagnosis of rotator cuff tears. If pain provocation is positive, CSA could be excluded because of the high specificity and PPV of this test. If only muscle weakness is positive, the possibility of CSA should be considered, because of the low specificity and PPV. Therefore, we can prevent the misdiagnosis of CSA as rotator cuff tear and avoid delays in diagnosing CSA.

CONCLUSION

Muscle weakness in the empty and full can tests could not differentiate rotator cuff tears from CSA because of its low specificity and PPV; however, pain provocation was useful for differentiating these two conditions because of its high specificity and PPV.

REFERENCES