We're back with another shoulder diagnosis: instability and dislocations. You should definitely be able to recognize complications and know what kind of imaging to request following different dislocations. We also review the Beighton score for assessing general hypermobility.
Use code FIELDGUIDE for 40% off a MedBridge subscription.
Find more resources and subscribe to practice questions at PhysioFieldGuide.com.
Use code FIELDGUIDE for 40% off a MedBridge subscription.
Support the podcast and get study guides and bonus episodes at Patreon.com/physiofieldguide.
Find more resources and subscribe to practice questions at PhysioFieldGuide.com.
Good morning, good afternoon, or good evening, everyone. We are continuing our shoulder series today with an episode on glenohumeral instability. Before we jump in, I just want to note that I plan to be at CSM in San Antonio this week. If you see me, feel free to say hi, and I might have some OCS Field Guide swag to give out—i.e., some sweet Field Guide stickers—if you want any. But that’s not why you’re here now, so let’s talk about shoulder instability. We will start with a case and some questions to get your brain primed for this episode.
A 17-year-old was playing basketball and dislocated his shoulder when he collided with another player. The shoulder was reduced on the sidelines by the athletic trainer. This is the first time he has dislocated his shoulder. He is now seeing you on day 2 after the injury as a direct access patient. He has kept his arm in a sling since the dislocation.
First question: what position was the shoulder joint most likely in at the time of the dislocation? And, as a follow up: what direction was the joint most likely dislocated in? I’m not going to provide multiple choices for these two questions; just answer for yourself: what position was the shoulder most likely in, and what direction was the dislocation most likely in?
Second question: which nerve is most likely to be injured? Is it the axillary nerve, the upper subscapular nerve, the suprascaular nerve, or the medial cord? And as a follow-up, what tests should be performed to assess for injury to that nerve?
Third question: which of the following radiographs should be performed in order to detect a Hill-Sachs lesion? Is it the scapular Y view, the external rotation view, or the Stryker notch view?
If you’re confident in your answers to these questions, then you’re already doing great. If you’re not, you should be by the end of this episode. Let’s jump in. First we will talk about categorization, then types of dislocations, complications, and imaging.
Glenohumeral instability is defined and categorized in a variety of ways—sometimes based on the direction, sometimes based on the number of episodes, and sometimes based on etiology. The categories I see most commonly, and that I think are the most useful and easy to remember, are the acronyms TUBS (T-U-B-S) and AMBRI (A-M-B-R-I).
TUBS instability stands for Traumatic, Unilateral, Bankart, and Surgery. So T-U-B-S, it’s a traumatic mechanism of injury, it results in unilateral instability, it may be associated with a Bankart lesion, and if so, it’s likely going to require a referral for surgery.
AMBRI instability stands for (A) Atraumatic, (M) Multidirectional, (B) Bilateral, (R) Rehabilitation, and (I) Inferior capsular shift. These are those patients who often have global hypermobility and can dislocate without any injury or trauma—like a softball player I saw who dislocated her shoulder reaching down for a ground ball. Because there is global hypermobility, it can often be a bilateral problem. Rehabilitation is the preferred treatment for this type of dislocation, but when surgery is decided on, the surgery that is typically performed is an inferior capsular shift.
I want to take a second and point out that Bankart is NOT a part of the AMBRI acronym—remember the B in AMBRI stands for bilateral. We will talk more about Bankarts in a couple minutes, but you should remember that a Bankart is a soft tissue tear or avulsion from an anterior dislocation. This is less likely to happen in these atraumatic dislocations, because these patients tend to have looser or more pliable ligaments as a feature of their general hypermobility. So when they dislocate, the ligaments are not so tight that they get pulled on and cause Bankarts.
You can see how these two categories capture a large number of the dislocations you’re going to see, but there is also some gray area and overlap. Someone with general hypermobility and frequent dislocations can still have a traumatic dislocation as well—it’s not always going to be atraumatic. Or someone who has had multiple traumatic dislocations can begin to develop hypermobility and recurrent instability. So if we think of TUBS and AMBRI dislocations as two ends of the dislocation spectrum, there is another vague category that covers both the middle of the spectrum and the AMBRI part of the spectrum, and that is multi-directional instability (sometimes abbreviated MDI).
The specific diagnostic criteria for multi-directional instability has not been clearly decided, and there’s a lot of debate surrounding it—so you might find definitions that differ a little from what follows. The general definition is shoulder instability in at least two directions—often anterior and inferior or anterior and posterior or instability in all directions, like in an AMBRI scenario. In addition to covering AMBRI dislocations, this definition covers people who have had repeated, unilateral traumatic dislocations that have led to recurrent shoulder instability in multiple directions. So you might have a patient who doesn’t dislocate bilaterally, but who has developed hypermobility due to their history of injury and now has repeated dislocations and instability in multiple directions.
So that covers the most common, broad categories of dislocations and instability. Let’s start to focus on specific dislocations based on their directions.
Anterior dislocations are by far the most common, with estimates putting it at about 95% of all dislocations. So in the case we started with, when we asked in which direction the dislocation was most likely to have occurred, the answer is anterior—it is by far the most likely. Anterior dislocations typically affect the anterior inferior glenohumeral ligament, which is compromised in combined shoulder abduction and external rotation—so think about a quarterback getting hit from behind halfway through his throw or a basketball player who is hit from behind while reaching up for a rebound.
If you have difficulty remembering that it’s the anterior band of the inferior glenohumeral ligament, remember that, in the case of AMBRI dislocations, the I stands for inferior capsule shift. It’s the inferior ligament that benefits from being tightened up to prevent future dislocations. The anterior band is implicated in anterior dislocations, and the posterior band is implicated in posterior dislocations, which we will talk about in a moment.
Now, in real life, the mechanism of injury and subjective report should be plenty of information to diagnose an anterior dislocation. But if for some reason you needed to perform some tests to confirm the diagnosis, the anterior apprehension and relocation test has the best diagnostic accuracy. In the anterior apprehension test, the patient is supine and the PT moves the shoulder into abduction and then slowly into external rotation—you know, the vulnerable position for the affected ligament. Pain or apprehension are considered positive, but since there are a lot of potentially painful structures being stressed here, you should know that apprehension is more sensitive and specific than pain. The relocation test is only done if the apprehension test is positive. In the relocation test, the examiner performs the same maneuver except they provide a posteriorly directed force to the anterior proximal humerus, basically stabilizing the shoulder against a possible anterior dislocation. If the pain or apprehension is reduced, the test is positive—again, with apprehension being a more useful measure than pain. If you want to be really cruel and never see your patient again, a third test can be clustered with these two. It’s called the anterior release or surprise test. It is performed at the end of the relocation test; when the examiner gets to full external rotation, they suddenly remove the hand providing the posterior force to the humerus. If pain or apprehension is reproduced, the test is positive with 99% specificity and a 58.6 positive likelihood ratio. Which is great, except you might have completely lost your patient’s trust. But just in case the OCS asks: the best cluster for anterior instability is anterior apprehension and relocation, possibly followed by an anterior release or surprise test.
Let’s move on to posterior dislocations.
Posterior dislocations or subluxations are most likely to happen with the shoulder in a flexed and internally rotated position, like in American football when a lineman is blocking. The posterior inferior glenohumeral ligament stabilizes the shoulder in this position, so it’s most affected by posterior dislocations. About 2/3 of posterior dislocations are due to trauma, and about 1/3 are secondary to seizures.
These are much less common than anterior dislocations, and it’s much more likely to see posterior subluxations instead of full dislocations, so it might be a little trickier to tease out posterior instability in your subjective examination. And patients are going to report pain in the posterior shoulder, which might look like posterior impingement. However, patients dealing with posterior instability are more likely to report pain in shoulder flexion, adduction, and internal rotation—the position that stresses the posterior inferior glenohumeral ligament.
The most useful special test to rule in posterior instability is the posterior apprehension test. Here, the elbow is flexed, and then the shoulder is flexed to 90 and adducted. The examiner applies a posterior force through the long axis of the humerus, and apprehension is considered positive. Again, you’re basically trying to recreate the conditions of the injury and then wait to see if the patient freaks out. Keep in mind that this test can also be painful with posterior impingement, so a painful test does not necessarily indicate instability.
Last, let’s talk about inferior dislocations.
Inferior dislocations are also very uncommon, so inferior instability is more frequently associated with atraumatic dislocations. These are going to be your multi-directional instability or AMBRI cases.
To test for inferior instability, you can use the hyperabduction test. (Note that this is a different hyperabduction test than the hyperabduction test used for thoracic outlet syndrome.) In this hyperabduction test, the examiner stands behind the patient, applies an inferior force to the scapula, and then passively abducts the shoulder. If apprehension is reported with abduction beyond 105 degrees, the test is positive.
You will notice that there are some common tests that were not recommended here, like the load and shift test or the sulcus sign. That’s because those tests are really joint laxity tests, and—just like in the ankle—joint laxity and joint instability are not exactly the same thing. A joint can be lax but the individual might have adequate neuromuscular control so that the joint is not unstable. That’s why apprehension tests are recommended above joint laxity tests.
Before we move on, I want to talk about multi-directional instability again. By definition, this form of instability has to involve at least two of these directions, so there can be multiple positive tests. We don’t yet have a clear diagnostic criteria for MDI, but it has been suggested that positive apprehension tests in at least two directions plus the presence of a positive Beighton score may strongly indicate possible multi-directional instability.
Remember that the Beighton score is a measure of global hypermobility, and I would bet money that it shows up somewhere on the exam. Not only does it help diagnose multi-directional shoulder instability, but it’s associated with other hyper-mobility or connective tissue conditions, like Marfan’s or Ehlers-Danlos syndromes. So you should know how to perform and score it.
There are nine possible points. You gain a point for each of the following: passively extend the little finger as far as possible. If it extends beyond 90 degrees at the MCP joint, you get a point—one point for each side that is positive. Next, passively flex the wrist. Grasp the thumb and attempt to passively pull it toward the forearm. If it can touch, you get one point for each side. Next, fully extend the elbow. If it hyperextends beyond 10 degrees, you get a point for each side. Do the same with each knee. If the knee hyperextends beyond 10 degrees, you get a point for each side. Last, have the patient stand up and try to bend forward and put their palms flat on the floor while keeping their knees straight. If they can, that’s one point.
For adolescents and adults, 5/9 is considered positive. For prepubescent children, 6/9 is considered positive. So one more time, you’re looking for little finger extension past 90 (one point for each side), the ability to passively pull the thumb to the anterior forearm (one point for each side), 10 degree hyperextension of the elbow (one point for each side), 10 degree hyperextension of the knee (one point for each side), and the ability to put the palms flat on the floor while standing with knees straight.
A positive test indicates general hypermobility.
Let’s move on now to complications.
A few common complications may follow dislocations. Recall that the anterior inferior glenohumeral ligament attaches to the inferior two thirds of the anterior labrum. When an anterior dislocation causes an avulsion of this ligament and the associated anterior/inferior labrum, that avulsion is called a Bankart lesion. Sometimes the avulsion also contains a fragment of the glenoid, which can be seen on radiograph and is called a Bony Bankart. The opposite can happen with posterior dislocations, and the posterior inferior ligament together with the posterior inferior labrum can pull away from the glenoid, which is called a Reverse Bankart. When a bony avulsion occurs at the same time, it’s called a Reverse Bony Bankart. At least those terms make sense, right?
A second injury commonly associated with dislocations and Bankart lesions is a compression fracture of the posterosuperior humeral head. This happens during anterior-inferior dislocations when the articular surface of the humerus impacts the glenoid rim. This compression fracture on the posteriosuperior humeral head is called a Hill-Sachs lesion. During a posterior dislocation, a compression fracture can occur on the opposite side of the humeral head—the anterosuperomedial surface—which is called—you guessed it—a Reverse Hill-Sachs.
If for some reason—in the heat of the exam—you’re tempted to mix up Bankart and Hill-Sachs, remember that *H*ill-Sachs affects the *h*umerus. The Hs go together.
Nerve injuries are also common. The nerve most commonly injured during a dislocation is the axillary nerve. Axillary nerve injuries are going to present with deltoid weakness and sensory loss or paresthesias around the inferior deltoid region. So a full upper extremity neurovascular screen is warranted after a dislocation, but be on the lookout for the axillary nerve specifically. Now, most of the time, these injuries will resolve spontaneously, but you will want to keep an eye on it, communicate with the other providers, and consider referral—particularly if there is significant motor loss or things are not improving.
Another potential complication is rotator cuff tear, but this really only applies to dislocations in those older than 40. Most dislocations are going to occur in younger patients—usually males in their teens and twenties. But dislocation frequency peaks again for females in their 70s—which is thought to be due to dislocations associated with falls. So when we’re dealing with dislocations and instability, we’re often thinking about young, athletic populations, but don’t forget dislocations in the elderly as well.
Last, I want to talk about recurrent dislocations. This isn’t exactly a complication, but it is potential sequelae following a dislocation. Our current data seems to indicate that about half of people who suffer traumatic dislocations will recover with conservative management and never have a subsequent dislocation. For those who do have a reoccurrence, about 90% of recurrent anterior dislocations will happen within 2 years of the initial dislocation. Males who are younger at the time of the initial injury are at higher risk for developing recurrent dislocations and instability.
I want to wrap up by mentioning imaging. As we’ve said before, the OCS exam is testing whether you have advanced knowledge in the field of orthopedics, and part of that process is testing your ability in advanced practice settings or in states with imaging privileges—or at least your ability to communicate intelligently and make recommendations to other providers.
If your patient goes to an urgent care or maybe a chiropractor and gets pure AP and pure lateral images of the shoulder, that’s often enough to visualize any significant shoulder dislocation. However, as PTs, we know that a direct lateral view is not ideal for visualizing the shoulder joint because of the 30 degree tilt of the scapular plane. So it’s recommended that a scapular Y image be taken following a suspected dislocation or subluxation, because that image is taken at an angle perpendicular to the plane of caption—so you’re looking straight at the glenoid. This angle makes it very easy to catch any subtle subluxations or less obvious dislocations. So again, the scapular Y view is going to be best at catching any dislocations or subluxations. This view is part of the standard shoulder series, so if a standard series is ordered, the Y scapular view should be included. It’s also sometimes called the lateral scapula view, so you should know those are the same thing if they pop up on the exam.
To catch a Hill-Sachs lesion, one of the common images ordered is the Stryker notch view (S-T-Y-K-E-R). Here, the patient’s arm is abducted and resting on top of their head. (I don’t think you need to know how to set up the patient—I’m just trying to paint a picture for you.) The other view recommended for catching a Hill-Sachs lesion is the shoulder internal rotation view, where the patient internally rotates and places their hand behind their back. Both of these help get clearer images of the posterior humerus, where the Hill-Sachs lesion would be located.
So, one more time: the best view to visualize a subluxation or dislocation is the scapular Y view (or lateral scapula view), and the best images to detect a Hill-Sachs are the Stryker notch or the internal rotation view.
For treatment, until we get a clinical practice guideline or something really powerful and groundbreaking, I’m once more going to point you back to the STAR classification system we covered a few episodes ago. So let’s wrap up by reflecting on the earlier case.
We had a 17-year-old with a primary traumatic dislocation. We know it is most likely anterior, since 95% of dislocations are anterior. And the most likely position in which to have an anterior dislocation is shoulder abduction and external rotation. Our next question was concerning the nerve that was most likely to be injured. As we’ve now discussed, that’s the axillary nerve. So you should do a full neurovascular screen, but pay special attention to shoulder abduction strength and sensation around the inferior deltoid. Our last question was concerning imaging. What view on radiograph would help detect a Hill-Sachs lesion? Out of the choices we provided, the correct answer was the Stryker notch view, but the internal rotation view works as well.
That’s it for this episode. I do hope to see some of you at CSM, but if you choose to stay home and study, that’s great too. Study hard, and I’ll see you next time.