Physical Examination and Imaging
- David Bernholt, MD
- Robert H. Brophy, MD
Abstract
Isolated tears of the subscapularis can occur through forced external rotation or by anterior shoulder dislocation. Tears can be full or partial and are associated with weakness of internal rotation of the shoulder. A variety of examination maneuvers have been described to evaluate for subscapularis injury, including the lift-off test, internal rotation lag sign, belly press test, belly-off test, and bear hug test. Although these examination maneuvers are highly sensitive for complete tears, their sensitivity is much lower for partial tears, ranging from 12% to 71%. Magnetic resonance imaging (MRI) can be used to diagnose either full or partial subscapularis tears; however, MRI is less sensitive in partial tears. Arthroscopy remains the gold standard for diagnosis of subscapularis tears.
Keywords: examination; imaging; subscapularis.
Introduction
- ▪
History:
- ▪
Most common mechanisms are forced external rotation/hyperextension and anterior dislocation.
- ▪
Common complaints are anterior shoulder pain and weakness.
- ▪
- ▪
Examination:
- ▪
Assessment should be made for increased passive external rotation.
- ▪
Focused testing of subscapularis can be done through a variety of examination maneuvers, including lift-off test, internal rotation lag sign, belly press/Napoleon test, belly-off test, and bear hug test.
- ▪
- ▪
Imaging:
- ▪
Magnetic resonance imaging (MRI) is the modality of choice.
- ▪
Ultrasonography is similar in sensitivity to MRI for full-thickness tears.
- ▪
Arthroscopy remains the gold standard for diagnosis of subscapularis tears.
- ▪
History
As with any injury, a thorough history should be obtained with emphasis on mechanism and symptoms. Traumatic injuries of the subscapularis may occur through forced external rotation or hyperextension of the shoulder ( ) or in association with an anterior dislocation of the shoulder ( ). One study reported a direct blow to the shoulder as the mechanism for 13 of their patients with either full or partial subscapularis tears ( ). Alternatively, a subscapularis tear may manifest without any known traumatic event, as is often the case for other rotator cuff tears. Patients typically complain of anterior shoulder pain ( ); however, anterior shoulder pain has neither high sensitivity nor high specificity for subscapularis tears. Patients with any rotator cuff pathology most often report anterior or lateral pain irrespective of which tendons are involved ( ). In addition to pain, patients may complain of weakness about the shoulder, especially with internal rotation. When associated with an anterior dislocation, subscapularis ruptures may also manifest as subjective or clinical instability; however, instability is not typically present in forced external rotation or hyperextension injuries ( ).
Physical Examination
Physical examination for a subscapularis injury should begin with visual inspection of the shoulder, noting areas of swelling, ecchymosis, or atrophy. Careful palpation of the shoulder can identify area(s) of maximal tenderness; tenderness may be present over the anterior aspect of the glenohumeral joint or at the intertubercular groove ( ). Active and passive ranges of motion of the shoulder should be assessed. An increase in passive external rotation greater than 10 degrees relative to the unaffected side is supportive of a complete subscapularis tendon rupture ( ). This finding was present during office examination in 10 of 16 patients in the series reported by Gerber et al and in 5 of 8 described by Resch et al, and was present in all of the Gerber group’s patients under anesthesia. If the patient sustains only a partial tear of the subscapularis, a significant increase in passive external rotation is not expected to be present.
There are six examination maneuvers described in the English literature that attempt to diagnose injury to the subscapularis tendon by isolating its motor function ( 
):
- Video 8A.1
Subscapularis Examination Maneuvers
Lift-Off Test
The lift-off test, initially described by , is performed by having the patient place the hand on the affected side behind the back with the elbow flexed to approximately 90 degrees and the dorsum of the hand resting against the midlumbar region of the back or, if possible, the lower border of the scapula. The patient is then asked to lift the hand off of the back through additional internal rotation of the shoulder ( Fig. 8A.1 ). The result is positive when the patient is unable to lift the hand away from the back. This test can be performed only in patients who have full passive internal rotation of the shoulder and who are not limited by pain when attempting active internal rotation of the shoulder.
The lift-off test places the shoulder near maximal internal rotation and extension, which suggested would best isolate the motor function of the subscapularis. Electromyography (EMG) studies have shown that the subscapularis is the primary internal rotator of the shoulder in the lift-off position ( ); however there are still contributions from other internal rotators of the shoulder as well as extensors of the shoulder and elbow ( ). Increasing the amount of shoulder internal rotation at the start point of the lift-off test helps increase the activity of the subscapularis and further isolate its function ( ). In one study, patients with a visually confirmed subscapularis rupture, as well as patients given a targeted nerve block to prevent any subscapularis function, were able to perform the lift-off maneuver with the hand at the buttocks and midlumbar region but were unable to do so with the shoulder in maximal internal rotation with the hand at the lower border of the scapula ( ).
Internal Rotation Lag Sign
The internal rotation lag sign was first described by Hertel et al in 1996. This test is carried out using the same positioning as in the lift-off test. The examiner then lifts the hand passively away from the back by internally rotating the patient’s shoulder. The patient is instructed to maintain the position of the hand away from the back, and the examiner lets go of the hand while still supporting the patient’s elbow ( Fig. 8A.2 ). The result is positive if there is an internal rotation lag, meaning that the patient’s hand falls toward the back. The amount of lag can also be measured quantitatively, and suggested recording this lag to the nearest 5 degrees. Like the lift-off test, the internal rotation lag test requires full passive internal rotation and painless active internal rotation to provide accurate results. Because this test uses the same positioning as the lift-off test, the afore mentioned EMG studies likely apply to it as well, although these studies did not assess the internal rotation lag sign specifically.
Belly Press Test or Napoleon Test
The belly press test was first described by Gerber et al in their 1996 study. They developed it to test subscapularis function in patients in whom passive internal rotation is too short for positioning for the lift-off test. To perform the test, the patient places the palm of the hand against the abdomen and then pushes the hand against the abdomen with the wrist in neutral while internally rotating the shoulder to bring the elbow anterior to the trunk. A positive result occurs when the patient is unable to keep the elbow anterior to the trunk ( ) ( Fig. 8A.3 ).
The modified belly press test and Napoleon test ( ) have been presented as alternatives to the originally described belly press test. The tests are performed the same way by the patient, but the results are determined by the amount of wrist flexion required to keep the elbow anterior to the trunk ( Fig. 8A.4 ). The modified belly press test is performed on both sides and a positive result occurs when there is 10 degrees or greater increased wrist flexion on the affected side ( ). For the Napoleon test ( ), a positive result consists of wrist flexion from 60 to 90 degrees, an intermediate result of wrist flexion from 30 to 60 degrees, and a negative result with wrist remaining neutral ( ). These three categorizations for the Napoleon test are meaningful because in the series reported by , all the patients with complete subscapularis tears had a positive result, patients with partial tears greater than 50% most often had an intermediate result, and patients with partial tears less than 50% tended to have a negative result.
Belly-Off Test
The belly-off test is another test for the internal rotation lag sign but is performed with positioning similar to that for the belly press test. The examiner performs the test by flexing the patient’s elbow to 90 degrees and passively flexing and maximally internally rotating the shoulder to bring the patient’s palm against the abdomen. One of the examiner’s hands should be used to stabilize the patient’s elbow while the other holds the hand/wrist of the patient in position against the abdomen. The examiner then releases the patient’s hand/wrist while instructing the patient to maintain the position of the hand against the abdomen ( Fig. 8A.5 ). If the hand lifts off of the abdomen, the result is considered positive ( ).
Bear Hug Test
To perform the bear hug test, the examiner places the palm of the patient’s hand belonging to the injured extremity on the contralateral acromioclavicular (AC) joint with the shoulder flexed to either 45 or 90 degrees. With the palm in this position, the elbow should be positioned directly anterior to the shoulder and should be kept directly in line with the shoulder, maintaining the same craniocaudal level. The patient is then asked to keep this position while the examiner applies an external rotation force to the patient’s wrist in an attempt to pull the wrist/palm away from the contralateral AC joint ( Fig. 8A.6 ). A positive result occurs when the patient is unable to maintain the palm against the AC joint or the position of the arm/elbow or when there is detectable weakness relative to the contralateral extremity ( ). EMG analysis of upper and lower subscapularis firing during this test has shown that the test better isolates the subscapularis muscle with 45 or 90 degrees of shoulder flexion than with 0 degrees of flexion; the analysis also found that when the test is performed at 45 degrees of flexion, it better isolates the upper subscapularis, whereas at 90 degrees of flexion it better isolates the lower subscapularis ( ).
Internal Rotation Resistance Test
The internal rotation resistance test as presented by is performed with the patient in two different positions: (1) 0 degrees of shoulder abduction and external rotation, and (2) 90 degrees of abduction and maximal external rotation. For the 0-degree test, the patient flexes the elbows to 90 degrees bilaterally with the arms adducted at the side and the fingertips facing forward to indicate 0 degrees of shoulder external rotation. The examiner then applies an internal rotation force perpendicular to the patient’s forearm while asking the patient to maintain the position. For the 90-degree test, the patient flexes the elbow to 90 degrees, and then the shoulder is abducted to 90 degrees and externally rotated to a maximal degree by the examiner. The examiner then applies an internal rotation force perpendicular to the patient’s forearm while asking the patient to maintain the position. A positive result occurs when the patient cannot maintain the position or when there is detectable weakness in comparison to the contralateral side ( ).
Utility of Subscapularis Clinical Tests
Owing to the low frequency of isolated complete subscapularis tendon ruptures, there are no well-established sensitivity and specificity values for clinical subscapularis tests. This situation is further complicated by the fact that many studies combine partial-thickness and full-thickness subscapularis tendon tears in assessing test sensitivity and specificity. The lift-off test has shown nearly perfect sensitivity for full-thickness tendon tears in many series ( ); however, when partial subscapularis tendon tears are included, the sensitivity has been reported as low as 12% to 69% ( ). When initially presented by , the internal rotation lag sign showed a much higher sensitivity than the lift-off test, but this figure has not been replicated in all studies, and the sensitivity now ranges from 20% to 71% ( ) The modified belly press and belly-off tests were reported as much more sensitive (88% and 87%, respectively) than the lift-off and internal rotation lag sign tests (71% and 40%, respectively) in one of the larger studies comparing the effectiveness of the various clinical examination maneuvers in diagnosing partial subscapularis tears ( ). Similarly, the bear hug test has been reported to have a higher sensitivity for partial tears than the lift-off test (60% vs 17%) ( ). The greater sensitivities of the belly press, belly-off, and bear hug tests for partial subscapularis tears have been suggested to result from more recruitment of the upper subscapularis than of the lower subscapularis with the test movements, because most partial tears occur at the upper portion of the tendon. Although one EMG study supported this explanation ( ), a later EMG study showed no difference in recruitment of the upper and lower subscapularis by the bear hug or belly press test and the lift-off test ( ). Despite the problematic lack of sensitivity with the various subscapularis examination maneuvers, these tests do show high specificity ( ).
Imaging
Because a subscapularis tear cannot often be diagnosed clinically, imaging is important to the diagnosis, although arthroscopy does remain the gold standard for diagnosis of subscapularis pathology. MRI is the primary imaging modality used to evaluate the integrity of the subscapularis. Unfortunately, just as with the clinical examination, MRI is very sensitive for identifying complete tears of the subscapularis tendon, but identifying partial tears comprising less than 50% of the tendon width with this imaging modality is difficult ( ). To best view the subscapularis on MRI, a combination of the sagittal oblique and axial planes is used ( ); using only one plane decreases the accuracy for diagnosis ( ). Furthermore, one study has suggested that the addition of radially sliced MR images can be used to increase sensitivity for subscapularis tears, particularly partial tears ( ). Partial tears typically occur in the anterosuperior portion of the tendon, which is susceptible to the partial-volume effect, making them hard to visualize, and addition of radially sliced images helps mitigate this problem ( ). Although use of intraarticular contrast for MR angiography (MRA) has been shown to enhance the diagnosis of supraspinatus and infraspinatus rotator cuff tears, one study showed MRA and MRI to be equivalent for diagnosis of subscapularis tears ( ). However, the ability to better assess for other shoulder pathology that may be responsible for the patient’s symptoms makes MRA the preferred choice ( ).
Overall, the reported accuracy for MRI is inconsistent across the literature. One of the biggest factors contributing to this may be the experience and training of the examiners reading the images. Multiple studies have reported very low sensitivity for detecting subscapularis tears on MRI, ranging between 31% and 39% ( ). In contrast, in two studies in which musculoskeletal radiologists read the images, the sensitivity was reported at 80% and 91% and specificity at 91% and 86% for any tear, and sensitivity at 100% for full-thickness tears ( ). In a follow-up study, whose initial study found a 39% sensitivity, were able to markedly increase the sensitivity to 73% and specificity to 94% by employing strict diagnostic criteria to assess for subscapularis tears. These diagnostic criteria require that two of the following conditions be met: (1) the presence of tendon tear on the axial or (2) sagittal images, (3) presence of subluxation or dislocation of the biceps tendon on the axial images, and (4) presence of atrophy of the subscapularis muscle on the sagittal images ( ) ( Fig. 8A.7 ). Importantly, in this study all MR images were read by five orthopedic surgeons at different institutions, suggesting that use of these criteria may be generalizable and may allow for more accurate interpretation of MRI for subscapularis tears even in the absence of a trained musculoskeletal radiologist.
Although MRI is the imaging modality of choice for evaluating subscapularis injuries, ultrasonography also has been shown to have utility. It has been validated in the diagnosis of rotator cuff tears in general ( ). There are limited data pertaining to ultrasonography for the subscapularis tendon specifically, but ultrasound has been demonstrated to be able to accurately diagnosis full-thickness tears ( ). The largest study of ultrasonography for subscapularis ruptures demonstrated a sensitivity of 86% for full-thickness subscapularis tendon tears; however, ultrasound is less sensitive for evaluating small partial-thickness subscapularis tears ( ). It should be noted that accurate ultrasound evaluation of the subscapularis tendon requires that the patient have full external rotation of the shoulder, a potential limitation of this imaging modality.
Plain radiographs are of minimal utility in the diagnosis of a subscapularis tendon injury, but shoulder radiographs should be obtained in the patient presenting with a traumatic shoulder injury or chronic shoulder pain to rule out fracture, arthritic change, and any bony lesions. In the case of a complete subscapularis rupture, it is possible to see anterior subluxation of the humeral head on a radiograph. Subscapularis tendon injuries may occur after anterior dislocation, and in this case radiographs may demonstrate a bony Bankart or Hill-Sachs lesion. Computed tomography could be used to better evaluate any of these bony lesions, but it is not helpful for diagnosis of a subscapularis tendon rupture.
Surgical treatment: Arthroscopic Repair
- Laurence D. Higgins, MD, MBA
- M. Daniel Hatch, MD
- Tyler J. Hunt, BS
- M. Daniel Hatch, MD
Abstract
Historically, subscapularis tears have been treated with an open technique and were the last of the rotator cuff tendon repairs to be fixed arthroscopically due to its unique constraints. These constraints include increased retraction, proximity to neurovascular structures, and limited visualization associated with the tendon. The surgical technique for each specific tear is varied slightly, but the common trend is Type I through III tears are treated through intraarticular visualization through the posterior portal, while type IV tears require extraarticular work. Over the past several of years, the arthroscopic repair technique has become more commonplace with improved knowlede, instrumentation, and techniques. The literature has found arthroscopic repair to be at least as successful as open repair, and it offers the added benefit of concomitant visualization and minimal invasion.
Keywords: Subscapularis; arthroscopic repair; open repair; full-thickness tear.
Introduction
- ▪
Arthroscopic repairs to the subscapularis tendon are unique because of the increased retraction, proximity to neurovascular structures, and limited visualization associated with this structure.
- ▪
The techniques for repair of these tears are varied, but for the surgeon it is most important to remember the principle of anatomic tension-free repair.
- ▪
The patient can be placed in the beach-chair or lateral position for this procedure.
- ▪
Repairs of isolated type I through III tears can be performed almost entirely intraarticularly, but type IV repairs require extraarticular work.
- ▪
The medial sling of the biceps tendon can be used to identify the superolateral portion of the retracted subscapularis tendon.
- ▪
Arthroscopic repair of isolated subscapularis tendon tears is as good as open repair.
Subscapularis tears were historically treated with an open technique and were the last rotator cuff tendon tears to be repaired arthroscopically owing to the unique challenges that arise from repair of that specific tendon ( ). First, the subscapularis tendon, secondary to its relative excursion, retracts more than the rest of the rotator cuff when torn, making mobilization more difficult. Second, the difficulty in mobilization is compounded by the fact that this tendon tends to scar to the coracoid near important neurovascular structures, namely the axillary artery and both the axillary and musculocutaneous nerves. Finally, visualization to mobilize the retracted and scarred subscapularis is challenging in the restricted subcoracoid space, particularly because of the close proximity to the previously mentioned neurovascular structures. Although techniques have been described for every type and variety of tear, adhering to the well-established principles of repair is more important than following any specific technique.
On the basis of preoperative imaging and intraoperative evaluation, classified subscapularis tears into five distinct types ( Table 8B.1 ). Type I tears are partial tears that involve the deep fibers of the superior one-third of the tendon. Type II through V tears are complete tears that are broken into thirds, from superior to inferior. Type II is a full-thickness tear of the superior one-third of the tendon, Type III a full-thickness tear of the superior two-thirds of the tendon, and Type IV a full-thickness tear involving the entire tendon with a well-centered head and fatty degeneration less than or equal to Goutallier stage 3. Type V is similar to type IV in dimension, but with an eccentric head that causes coracoid impingement and fatty degeneration greater than Goutallier stage 3.
| Type | Tear Description |
|---|---|
| I | Partial-thickness tear |
| II | Full-thickness tear of superior one-third |
| III | Full-thickness tear of superior two-thirds |
| IV | Complete subscapularis tear with well-centered head and ≤ Goutallier stage 3 fatty degeneration |
| V | Complete subscapularis tear with eccentric head and > Goutallier stage 3 fatty degeneration |
In the following section, we specifically detail our approach and technique for arthroscopically treating type I through type IV tears.
Surgical Options
Positioning
All patients receive an interscalene block prior to surgery. The patient is positioned in the beach-chair position (T-Max/Shoulder Positioner, Smith & Nephew, York, UK) at 70 degrees of inclination after having been intubated and receiving general anesthesia. The hips and knees should be flexed gently to relieve pressure on the sciatic nerve. The head and neck are positioned with a mask in a neutral position.
The entire operative extremity is prepared for surgery and draped free. The drapes should extend across the chest wall just above the nipple line, medially to about the sternoclavicular (SC) joint and across the lower neck, and posteriorly along the medial border of the scapula. The arm is then placed in a pneumatic arm holder (Spider Limb Positioner, Smith & Nephew) to allow free manipulation of the arm in space.
Portals
Three portals are typically used ( Fig. 8B.1 ). First, a 30-degree arthroscope is introduced through the posterior soft spot portal approximately 2 cm inferior and medial to the posterolateral corner of the acromion (viewing portal). Next, the anterior portal is created, via an outside-in technique by placement of a spinal needle into the joint just lateral to the coracoid through the rotator interval, as a working portal to place the anchor and to pass sutures through the torn tendon. The third portal is an anterolateral accessory portal placed via the outside-in technique approximately 1 cm lateral and inferior to the anterolateral acromial edge. This last portal is parallel to the lesser tuberosity and is useful to assist in pulling traction to evaluate the tear pattern, release the tendon, debride the subcoracoid space and lesser tuberosity, and place suture anchors. A cannula is optionally placed in this portal.
