Now that Austin has covered a framework for classifying and treating shoulder pain, today we'll take a closer look at what is classically known as shoulder impingement and rotator cuff tears. This episode describes the standard model for these diagnoses as well as the problems with those models and the current push to leave them behind.
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.
Hello and thanks for tuning in again to OCS Field Guide. We’re continuing our shoulder series today with an episode on another very controversial topic. So controversial, in fact, that we can’t quite settle on what to call it. I’m talking about what’s commonly called shoulder impingement, which has also been called rotator cuff-related shoulder pain, or rotator cuff syndrome, or subacromial pain syndrome—the last of these being the name the STAR-shoulder system uses, as we discussed in the last episode.
Now when it comes to highly controversial topics like this where there is a sharp divergence between what the newest literature is saying and what is commonly taught and practiced, I think it’s important to know both the traditional understanding and the new ideas for the purposes of the OCS exam. It’s important to know the new ideas, because this exam is testing whether you are up to date on the latest research in orthopedic practice. But it’s also still important to know some of the older theories, because this is, after all, a controversy. And controversies have two sides. Once one side has clearly won, it’s no longer controversial. So I’m planning to talk about our traditional understanding of shoulder impingement, because it hasn’t gone away yet, but then I’m also going to explain why it’s being challenged and what we’re moving toward. I also want to note at the outset that in both the traditional model and the newer model, shoulder impingement and rotator cuff tears have been viewed as interconnected, so we’re going to talk about them together. Let’s begin.
Shoulder impingement, as it’s traditionally understood, is broadly categorized as external impingement, where the rotator cuff muscles, tendons, and subacromial bursa rub against the anterior third of the acromion, or internal impingement, where the underside of the muscles and tendons are pinched between the humeral head and the glenoid. Over time, this may progress to rotator cuff tears. So when we talk about a bursal-sided rotator cuff tear, we’re talking about tearing on the exterior of the rotator cuff from external impingement. And when we talk about articular-sided tears, we’re talking about tears on the deep, inside layer of the rotator cuff from internal impingement.
Within the broad category of external impingement, we subcategorize impingement as primary impingement or secondary impingement. Primary impingement is directly related to the morphology of the subacromial space. It is thought to be caused by acromion anatomy or morphological changes like bone spurs on the acromion or AC joint. You should know the three types of acromions: type I is a flat acromion, type II is curved, and type III is hooked. The role of the type III hooked acromion in shoulder impingement and rotator cuff tears has been challenged more recently, as many people with hooked acromions have no pain or tears, and surgical acromioplasty is not super effective—and we’ll talk more about that later. However, cadaver studies seem to corroborate the idea that rotator cuff tears are more common in those with type III hooked acromions.
Primary impingement was classically outlined by Neer as a three-stage process. Stage I is called edema and hemorrhage. Here, the rotator cuff tendon is irritated by mechanical impingement with overhead activities. This is going to occur in a younger, athletic population, and the process is thought to be reversible with conservative management. Stage II is fibrosis and tendonitis. This stage is characterized by repeated episodes of mechanical inflammation, and thickening or fibrosis of the subacromial bursa may occur. This is going to be seen in those 25-40 years old. Stage III is called bone spurs and tendon rupture. Here, repeated mechanical compression leads to rotator cuff tears, biceps tendon lesions, and bony abnormalities of the acromion and AC joint. Now, this all describes Neer’s classic model of impingement and rotator cuff tears. In real life, impingement doesn’t necessarily follow this clinical course all or even most of the time. However, many practitioners still use this framework to visualize the progression from impingement to tears, and it would be easy to write an OCS exam question that presents a patient case and asks the test taker to identify the Neer stage of impingement. So once more, stage I is edema and hemorrhage, and it’s reversible irritation that you will see in young, athletic patients. Stage II is fibrosis and tendonitis, and you’re going to see it in 25-40 year-old patients. Stage III is bone spurs and tendon rupture, which, let’s face it, is self-explanatory.
So, just to reorient: we said the two broad categories of impingement are external and internal, and external impingement can be classified as primary, which is related to the morphology of the subacromial space, and secondary, which we will talk about now.
Secondary impingement is still a mechanical, external form of impingement, but it’s caused by shoulder instability, not morphology. The instability may be due to particularly lax ligaments, poor dynamic neuromuscular stabilization from the rotator cuff muscles, or just excessive demands placed on these stabilizers by overhead throwing activities. Secondary impingement is particularly associated with anterior instability of the shoulder, and patients may present with positive anterior instability apprehension tests in addition to the typical shoulder impingement signs.
So when you’re trying to tell the two apart, you’re really looking either for signs of morphologically decreased subacromial space, like bone spurs or a type III acromion, or for signs of shoulder instability. Decreased space should lead you to a diagnosis of primary impingement, and instability should lead you towards secondary impingement.
That covers external impingement. Now let’s cover internal impingement. The most common form of internal impingement is posterior impingement seen in overhead throwing athletes. In posterior impingement, the posterior rotator cuff gets pinched between the humeral head and the glenoid in the late cocking phase of overhead throwing. This is likely to happen with a combination of shoulder abduction, external rotation, and horizontal hyperabduction—again, think late cocking phase of an overhead throw. Some studies have shown a link connecting more horizontal abduction to greater supraspinatus and infraspinatus contact against the rim of the glenoid. Glenohumeral instability or hypermobility may also be implicated here, as well as a tight or contracted posterior band of the inferior glenohumeral ligament. When this type of impingement occurs frequently, it can lead to fraying and articular-sided rotator cuff tears.
Before we look at rotator cuff tears and offer a critique of this traditional impingement model, let’s examine what a typical shoulder impingement patient will look like.
A typical patient with subacromial impingement is either a younger overhead athlete or middle aged. They will report anterior or lateral shoulder pain with motions above shoulder height, they might report a painful arc—that is, pain in the shoulder abduction range of 60-120 degrees, and they may experience pain at night.
We have a clinical prediction rule for subacromial impingement by Park et al that has not yet been validated but might still show up on the exam because, well, it seems better to use a cluster of questionable tests than individual questionable tests themselves. There are three parts: a positive Hawkins-Kennedy, a positive painful arc, and a positive infraspinatus manual muscle test for weakness or pain—OR a positive external rotation lag sign. You’re looking for at least 2 out of the 3 for a 90% post-test probability. So again, positive Hawkins-Kennedy, positive painful arc, and positive infraspinatus manual muscle test for weakness or pain OR an external rotation lag sign.
From there, you will determine the difference between primary and secondary impingement based on whether the patient appears to have morphological issues or stability issues.
Posterior internal impingement will look a little different, in that the location of pain will be the posterior shoulder, and the painful activity is usually overhead throwing or being in the position of abduction, external rotation, and horizontal abduction. Patients may complain of weakness or decreased throwing performance. So pain in the posterior shoulder that is aggravated by the late cocking stage of throwing and accompanied by external rotation weakness should make you suspect posterior impingement.
Now, in a minute, we’re going to move on and talk about rotator cuff tears. But first, I think it’s important to recognize that the concept of shoulder impingement—particularly primary impingement—is currently being challenged. First, because comparisons between surgical and nonsurgical management show that surgical acromion resection is no more effective than non-operative management. Additionally, acromion resection with a bursectomy has been found to be no more effective than bursectomy alone. So if the problem is the morphology of the acromion, why doesn’t changing the shape of the acromion work? Additionally, imaging studies have found impingement present in healthy, asymptomatic shoulders, and kinematic studies have indicated that the supraspinatus comes closes to the acromion at 30-70 degrees of shoulder elevation, not 60-120. So while the mechanical framework for shoulder impingement is still being used, a lot of evidence is stacking up against it—or at least complicating our relatively simplistic model. It’s for this reason that the term “impingement” is falling out of favor, and terms like rotator-cuff related shoulder pain or subacromial pain syndrome are gaining momentum.
With that caveat about impingement in mind, let’s move on to talk about rotator cuff tears. Here again, the more traditional way of viewing rotator cuff tears is by connecting them to repeated mechanical impingement against the acromion. However, we currently understand rotator cuff tears to be the result of both intrinsic and extrinsic factors.
Extrinsic factors are factors originating outside the tendon, like acromion shape and mechanical impingement. Intrinsic factors originate in the tendon itself, like vascularity, biology, and mechanical properties. Age-related changes in the tendon are a significant intrinsic factor. Collagen thinning and loss of tendon vascularity occur with age; in fact, vascularization significantly decreases after age 40. Smoking and diabetes also affect the vascularity of the tendon and delivery of oxygen to the tissue, so they are associated with both tears and surgical repair failure. One particularly significant but often overlooked factor is hyperlipidemia. High cholesterol is significantly associated with rotator cuff tendon tears—and, fun fact, it’s associated with Achilles tendon ruptures as well, as cholesterol appears to infiltrate these two tendons and form fatty deposits that decrease the tendon’s ability to withstand forces. So, to summarize, age, smoking, diabetes, and hyperlipidemia are all associated with rotator cuff tears, and there seems to be a small genetic component as well.
Tears can be partial thickness or full thickness and they can be bursal-sided (like in the case of external impingement) or articular sided (like in the case of internal impingement). Acute tears may be traumatic, but many people have insidious, chronic tears that they are unaware of or can’t correlate to a specific mechanism of injury. In older patients, your suspicion of rotator cuff tear should increase even without a mechanism of injury.
Although it’s recommended to refer to an orthopedic surgeon when an acute, massive tear is suspected, conservative management is a legitimate option for partial and full-thickness tears. It’s good to know that in the case of partial tears that involve less than 50% of the tendon thickness, one study found only 14% of their subjects had tear progression nearly 4 and a half years later. So patients with tears less than 50% of the tendon thickness do not need to be overly concerned about tear progression if they opt for nonsurgical management.
Let’s talk patient presentation. Patients with rotator cuff tears will present a lot like subacromial impingement patients: they will complain of anterior or lateral shoulder pain that may be worse at night. They may also complain of loss of strength and pain waking them at night.
We have two clinical prediction rules for detecting full thickness rotator cuff tears. Neither have been validated, so—just like with the impingement rule—we need to take these with a grain of salt. The first is by Litaker et al. It includes age over 65, external rotation weakness, and night pain. You’re looking for all three to detect a tear. The big caveat here is that as many as 55% of asymptomatic individuals over 65 have asymptomatic rotator cuff tears. So age over 65, external rotation weakness, and night pain might be good for detecting tears, but it’s not necessarily good at ensuring that a tear is the source of the pain.
Park et al. produced a second rule: positive painful arc, weakness or pain with infraspinatus manual muscle test, and positive drop arm sign. Two variables led to a 69% post-test probability, and 3 variables led to a 91% post-test probability.
Now if you’ve noticed that the rotator cuff tear rule and the subacromial impingement rule are remarkably similar, you’re right: both involve painful arcs and painful or weak infraspinatus tests. The only difference is the Hawkins-Kennedy for impingement and drop arm for tear. And if that makes you suspicious of the real validity of these clusters—or the validity of the diagnoses they’re meant to detect, you’re not the only one.
In many clinical cases, the relevance of rotator cuff tears is not clear. I already mentioned that as many as 55% of asymptomatic individuals over 65 have rotator cuff tears. Even in those younger than 65, some imaging studies on asymptomatic individuals have found prevalence rates of partial and full tears as high as 40%. I suspect that’s higher than the true prevalence, but it illustrates the point that a lot of people have asymptomatic tears. A study by Barreto and colleagues in 2019 that looked at bilateral shoulder MRIs of individuals with shoulder pain found as many abnormalities in the asymptomatic shoulder as they did in the symptomatic one—with the exceptions of a small difference in full-thickness supraspinatus tears and glenohumeral OA.
This makes our job both easier and more difficult. It’s great to know that people with rotator cuff tears can live pain-free, asymptomatic lives. When I have a belligerent patient ask me how I’m going to fix the tear in his rotator cuff with exercise, I can show him the numbers and say, “We don’t have to fix the tear. We just have to help you become one of the people who has a tear but has no symptoms.” At the same time, this ambiguity can cloud our evaluations and treatment plan. That’s why the STAR-shoulder classification system that Austin covered in the last episode is gaining popularity.
*[To reiterate, the STAR-shoulder system prefers the term “subacromial pain syndrome,” and suggests matching interventions to typical impairments based on irritability level. For decreased external rotation strength associated with this syndrome, high irritability patients should be treated with isometrics in neutral, moderate irritability patients should be treated with isotonic and eccentrics below shoulder height, and low irritability patients should be treated with higher load exercises in abducted positions. For the impairment of limited glenohumeral mobility or posterior shoulder tightness, high irritability patients should receive low grade manual therapy to the shoulder, manipulation to the spine, and physiological mobility exercises in mid ranges. Moderate irritability patients should receive higher grade and larger amplitude glenohumeral mobilizations, and those with low irritability should receive grade IV mobilizations at end range with sustained stretching. For the impairment of decreased pec minor length, those with high irritability should receive scapular retraction and positioning exercise and manual therapy techniques. Those with moderate irritability should receive supine stretches over a foam roll with arms below 90 degrees, and those with low irritability should receive terminal stretching in a doorway with arms at 90/90. Finally, for those with scapular muscle weakness or motor control deficits, high irritability patients should receive scapular clock or scapular setting exercises, those with moderate irritability should receive low rows and resisted scapular retraction exercises, and those with low irritability should receive prone horizontal scapular plane resisted abduction exercises.
These are all just examples of some interventions that match impairments to irritability. If you think through these, they make sense. And if you’re applying this kind of clinical reasoning to your shoulder cases in the clinic, then you should be able to reason through selecting an intervention on the exam.]
I want to wrap up by mentioning a great little article by Paul Salamh and Jeremy Lewis called, “It is time to put special tests for rotator cuff-related shoulder pain out to pasture” from JOSPT in May of 2020. Salamh and Lewis argue that the special tests we use for shoulder impingement and rotator-cuff related pain don’t give us a lot of information aside from knowing, hey, that position hurts. They don’t do a great job of telling us what hurts or why, so they argue that we should be moving away from them. Now, I think the test is still going to include shoulder special tests. But I do think this is an increasingly common view, and they authors lay out a nice little suggested list of examination elements that we should use to diagnose rotator cuff-related pain outside from provocative tests. I wouldn’t necessarily memorize this list. But for the 10% of the exam that’s testing “Orthopaedic Physical Therapy and Practice”—that vague, difficult-to-define-section—it might be good to recognize some of these steps in case they show up in a question.
Salamh and Lewis start with a comprehensive patient interview. They suggest identifying changes in loading history that may support a diagnosis of rotator cuff-related pain. Next, identify the impact of the symptoms on the individual, the patient’s beliefs and expectations, and what valued activities the patient wants to return to. Finally, identify psychosocial factors, lifestyle factors, current activity levels, medications, and supplements.
The second step is to screen for serious pathology. (You should be good at that by now.)
The third step is to use disability questionnaires, including general functional questionnaires, shoulder-specific questionnaires, and psychosocial questionnaires.
Fourth, assess impairments. They suggest conducting a neuro screen if appropriate and excluding referred pain as much as possible—so your cervical screen does here. They say to conduct a bilateral assessment, assess range of motion, assess strength and repetitions to pain or fatigue, assess response to changes in load on the muscle-tendon units, assess lower-limb and trunk range of movement and function, and, lastly, “appreciate that nociception is not necessary for the experience of pain. In other words, be aware of central sensitization and pain science factors.
The fifth step is patient education and advice regarding the condition and its management. They say to engage in shared decision making that incorporates harms, benefits, and the requirements of the main management options: rehab, injections, surgery, or no intervention. They write, “aim to encourage low-risk, high-value, evidence-informed care both for the individual and the sustainability of health care provision”—which I think is kind of a funny way of them saying that they think PT is good and we should recommend it.
Finally, we have the nonsurgical management step, and I think this a god part to pay attention to. For treatment, they say to “prescribe a graduated rehabilitation progression for at least 3 months, with activity modification as indicated, with the aim to exceed the patient’s functional expectations. Include all functional activities in rehabilitation: open-chain and closed-chain, precision, and ‘chaotic’ activities. Address lifestyle issues: smoking, nutrition, sleep, [and] stress. Appreciate that there is no cure and that attention to lifestyle, together with a range of whole-body exercises and activities, should be maintained and, if possible, incrementally increased, with no end date.” And finally, if you’re “not achieving desired outcomes, or if the condition worsens, consider other management options—but only after harms and potential benefits have been discussed and understood.”
I want to emphasize a few things here. They suggest a rehab plan of at least 3 months, which should help you form your treatment plan. They include open-chain, closed-chain, precision, and “chaotic” activities—which I think is a much more fun term than perturbation training. They explicitly suggest addressing lifestyle issues, like smoking, nutrition, sleep, and stress, which, as a profession, I think we’re not doing as much as we should. That may involve referral to other providers, but we should not be ignoring it. And finally, they focus on long-term management with no end date, emphasizing the importance of lasting lifestyle change over the idea that someone is “cured” once they have experienced relief from symptoms.
I’m going to conclude here for today. I have no doubt that the way we discuss and treat impingement and rotator cuff-related pain is going to keep evolving over the coming years, so keep an eye on it. Until we have a consensus, I think you’ll want to understand both sides of the debate as you prepare for the OCS. Lastly, I want to apologize to any authors of articles whose names I may mispronounce in this or in other episodes. As a fellow bearer of a difficult last name, I am truly sorry.
Thanks as always for listening, and until next time, study hard.
*The section in brackets was removed from the episode for the sake of time, but left in the transcript for the benefit of those studying the transcript later.