To cut to the chase, coracoidopathy is an unbeknownst but common cause of shoulder pain. My preliminary data gathered from my shoulder pain referrals suggest it is present in up to 85% of people with shoulder pain. It is the primary diagnosis in over a third of them, and is the sole diagnosis in 10%. If my sample estimates hold up in other populations, it is more common than rotator cuff pathology for which it is often confused. It is easily diagnosed by sharp point tenderness with palpation over the coracoid process, further confirmed with weakness and/or pain with supinated shoulder flexion (aka an isometric Speed’s test). The good news is that in isolation coracoidopathy is easily treated by performing dumbbell supinated front raises. The problem is, apparently, NOBODY KNOWS ABOUT IT, which results in misdiagnosis, ineffective physical therapy, and likely unnecessary surgery after “conservative treatment” fails. Unfortunately surgery still doesn’t fix the problem because surgeons generally aren’t looking at the coracoid process either.
Ad nauseum details below, and if you hate to read, a video is at the bottom.
Abbreviations used below (FYI, I hate when all abbreviations are not given on page one of a paper)
- CP: coracoid process
- SHB: short head of biceps brachii
- LHB: long head of biceps brachii
- CB: coracobrachialis
- PM: pectoralis minor
- RTC: rotator cuff
- SLAP: superior labral tear from anterior to posterior
- SFR: supinated front raise
- PRE: progressive resistance exercise
- ROM: range of motion
- AROM: active range of motion
- PROM: passive range of motion
- MRI: magnetic resonance imaging
- EMS: electric muscle stimulation
- FYI: for your information, haha
I became aware of the condition and came up with my treatment in 2012 when working with a patient who had persistent anterior shoulder pain months after having a RTC repair/subacromial decompression. I remember thinking, “it must be the long head of his biceps tendon,” but he wasn’t tender over his bicipital groove. When I closely read his op-report I found that the long head of his biceps had ruptured prior to his surgery and thus couldn’t be the source. After weeks of telling him the standard schtick, “the rotator cuff is known to refer pain to the front of the shoulder and down the arm,” I did more palpation and finally palpated his coracoid process, which was SHARPLY/ALARMINGLY tender. I then consulted my favorite physical therapy book, not one where someone tells you what to do but rather one that lets you see what you’ve got, Netter’s Atlas of Human Anatomy, which showed the coracoid process as the origin of the short head of the biceps and coracobrachialis muscles. The short head of the biceps and more so the coracobrachialis are the type of muscles you learn about for an anatomy test, but then forget because they’re inconsequential, or so I thought. I then did a pubmed.com search and came across only one paper that addressed those muscles as a site of pain at the coracoid process (Karim 2005). Karim injected the CP in six patients with anterior shoulder pain and tenderness over the CP with an anesthetic, which resulted in immediate and complete relief of pain. So I thought, “I bet that’s it.”
With that knowledge in hand, and I already considering myself a tendinopathy specialist, I then thought my way through the mode of action of those muscles/tendons and decided a supinated front raise (SFR) would be the best way to target, tension, and strengthen the CB and SHB while minimizing deltoid contribution. The addition of the SFR to my patient’s treatment program lead to a near immediate reduction in symptoms and decreased tenderness over the CP. From then on I started to look more closely at my shoulder patients and noticed they very often had CP tenderness and frequently had the weakness and pain with supinated shoulder flexion. It wasn’t a rare case, but almost the norm! At this point, having treated scores of patients with coracoidopathy and further refining my evaluation and treatment technique I figured it’s time to share.
What is coracoidopathy?
Coracoidopathy is tendin“opathy” of the tendons/muscles attaching to the “coracoid” process. See what I did there? Tendinopathy is a newer word for tendinitis, to reflect the fact that biopsies of painful tendons generally don’t find any inflammatory cells at the site of pain, for which the “itis” in tendinitis is in reference to. Tendinosis is another new word for tendinitis, for which more recent papers are distinguishing from tendinopathy, with tendinopathy used to describe the “painful condition” and tendinosis to describe the “tendon degeneration” that may or may not be associated with pain. To muddy things up a bit, some researchers are again arguing that there still is some inflammation present, at least early on, and as such, tendinitis isn’t always a misnomer. So tendinopathy, tendinosis, tendinitis, whatever you want to call it, it all depends on whose paper you read, but regardless, I know what you are talking about. What matters in this case is that the tendon pain/damage is coming from the coracoid process, a spot on the shoulder where three tendons attach. So if you want to call it coracoiditis, or coracoidosis I won’t argue, but yeah I thought of those too. Oh and to make things more complicated Crichton & Funk 2009 found the muscle fibers of the short head of the biceps and coracobrachialis insert directly into the coracoid process bypassing the existence of a “true tendon.” However, regardless of terminology, tendon type, or tendon existence, I find the same type of exercise program I used successfully for other tendinopathies to work as well or better with coracoidopathy.
The three muscles/tendons potentially involved are the SHB, the CB, and the PM. I think that Karim’s group hit the nail on the head and deserves the credit for first describing the condition as involving the SHB and CB. They specifically noted:
“No literature suggested tendons of the short head of the biceps brachii or coracobrachialis as a cause of pain.”
I agree with Karim in that I have been unable to find any papers before, or giving them credit since, implicating either the SHB or CB as causes of shoulder pain. If my numbers are correct about the incidence this is a remarkable omission in the medical literature. It’s analogous to trying to treat elbow pain, without knowing tennis elbow exists. In fact several recent review papers specifically on the subject of biceps tendinopathy and anterior shoulder pain fail to mention the short head of the biceps or coracoid process whatsoever (Ryu 2010, Tuckman 2006, Singaraju 2008, Churgay 2009). Unfortunately, I think what prevented Karim and his group from achieving acclaim was that they couched their discovery behind a terrible name. “Enthesitis of the biceps brachii short head and coracobrachialis at the coracoid process” doesn’t exactly roll off one’s tongue. Coracoidopathy on the other hand…
Certainly worthy of mention is Bhatia 2007 whose group, apparently unaware of Karim, came up with ‘almost’ the same treatment in what they described as “Bench Presser’s shoulder.” They also eliminated pain (this time in seven patients with tender CPs) with a localized injection of an anesthetic and corticosteroid at the CP. Bhatia targeted the medial aspect of the CP thinking that the PM was the primarily irritated tendon. However the CP being a small area, I expect relief of an irritated SHB and CB would also result.
While I would not rule out PM involvement, the prevalence of pain reproduced with the Speed’s test, and relief of pain following treatment with the SFR, leads me to think that Karim is more correct in his assertion that the SHB and CB is of primary importance. I do, however, think that Bhatia is correct in his association of the condition with the bench press exercise. Most patients I have treated with shoulder pain were not working out with the bench press exercise, but all of the patients I have seen who complained of anterior shoulder pain with bench presses did have coracoidopathy, for which I intend to specifically address in another blog. [update 6-5-16 to add the bench press and coracoidopathy blog and video]
If I had to limit it to one muscle/tendon as primary or most frequently irritated, I’d pick the CB. This is because resistive biceps testing at the elbow where the CB doesn’t act but which tensions the SHB plenty, is rarely provocative, while shoulder flexion, where the CB does act, most often is. Also it’s questionable how much the biceps, the long head at least, (Levy 2001) acts in shoulder flexion, and nobody has tested the short head. Also distinguishing between the SHB and CB is difficult as cadaver examination indicated they blend together at their origin at the CP, so differentiation is probably academic.
How common is coracoidopathy?
Retrospectively going through my patient charts after July of 2012 (thru the end of 2015) I had 59 shoulder referrals (after excluding post-op shoulder patients, recent fractures, or complaints related to systemic disease). I sought to determine how many had some degree of coracoidopathy. 10.2% (6/59) were diagnosed with isolated coracoidopathy, with no other apparent shoulder pathology. 35.6% (another 21/59) of shoulder patients were diagnosed with coracoidopathy as their primary diagnosis (CP the chief source of pain/Speed’s test most notable regarding pain and/or weakness) but had a secondary diagnosis. 40% (an additional 23/59) were given a diagnosis of coracoidopathy secondary to another shoulder pathology. Another way to look at it is 88% ((21+23)/50) of those with coracoidopathy also had another shoulder pathology. Of the included 59 shoulder referrals a total of 50 (84.7%) appeared to have some degree of coracoidopathy, of which 74% (37/50) had a codiagnosis of RTC pathology and 18% (9/50) also had adhesive capsulitis. Interestingly only 67.8% (40/59) of shoulder referrals were diagnosed with RTC pathology, indicating coracoidopathy may be more common. I have since started keeping track of my statistics prospectively, so it will be interesting to see if my numbers hold up, as well as what other clinicians find, as I’ll be embarrassed if my estimates turn out to be the result of confirmation bias.
If coracoidopathy is so prevalent, how come nobody knows about it?
That’s the big question for me. Now that I know about it, it’s obvious. However, (17 and counting now) I was a physical therapist for 13 years before I figured it out, and the standard “it’s referred pain” seemed to cover it. My data suggests there is usually more than just one problem, co-occurring with RTC issues 74% of the time, making the connection to the RTC an easy sell. Plus the RTC seems prone to tearing (perhaps because of all the cortisone injections the RTC gets whenever a person has coracoidopathy;) while the SHB/CB and PM do not, thus making the RTC the subject of more attention. Also there is the fact that the shoulder is a complicated joint, with lots of parts that can be injured, making it sometimes difficult to tell what’s what, while the CP sits just outside of what is often thought of as the shoulder joint (the glenohumeral joint). And of course, groupthink.
I do think recognizing coracoidopathy goes a long way towards making the shoulder less complicated.
How is coracoidopathy diagnosed?
If it’s just coracoidopathy, it’s easy. I test the person in resistive supinated shoulder flexion (aka Speed’s test) performed with the elbow bent ~10 degrees, and then see if they are tender with palpation over the coracoid process and/or the muscles/tendons where they attach to it. A milder case of coracoidopathy may or may not have pain or noticeable weakness with the Speed’s test but it rarely goes the other way with a painful Speed’s test and no tenderness over the CP (that finding would make me suspect a LHB/SLAP lesion). As mentioned above the problem is that more often than not there is coracoidopathy and something else. As such it’s best to do a full orthopedic shoulder evaluation to include assessment of shoulder range of motion (active and passive), resistive tests of all relevant muscles, and perhaps some special tests for which I imagine physicians and physical therapists can debate indefinitely. The tests in my video were abbreviated a bit because, in Jaydah’s case, I already knew what her injury was. Sometimes I’ll clear for cervical symptoms by looking at cervical AROM, I’ll then do shoulder AROM in standing, and if there is a limitation I’ll test PROM in supine, followed by resistive tests in shoulder abduction, shoulder internal and external rotation, biceps, triceps, followed by supraspinatus. I now always add in the Speed’s test with the elbow slightly bent. I palpate the entire scapular and RTC region and I always palpate the coracoid process.
So basically my shoulder evaluation is a fairly standard orthopedic shoulder evaluation, I just take an extra 1-2 minutes to add in the Speed’s test and palpate the coracoid process and it’s associated tendons.
How is coracoidopathy treated?
Both Karim and Bhatia reported they successfully treated the condition with a corticosteroid injection to the CP in their collective 13 patients. However, a number of studies have been published in recent years indicating that corticosteroid injections (on average) lead to short term improvement, at the expense of long term outcomes. Such that tendinopathy patients who get cortisone shots are (on average) WORSE OFF than if they had no treatment whatsoever. This is thought to be due to the catabolic nature of corticosteroids causing decreased tendon strength (Smidt 2009). Even a single injection in patients with plantar fasciitis (Lee 2014) results in an 19 fold increased risk of rupture, with another recent review concluding that corticosteroids resulted in significant long term harm to tendon tissue (Dean 2014). So I tell my patients the last thing I would do is get a cortisone shot. On the bright side, if you go to your doctor for anterior shoulder pain caused by coracoidopathy and ask for a cortisone shot, he’ll probably inject your RTC, so there’s that.
Rather than cortisone I recommended exercise. Since this blog is the first description of an exercise based program for coracoidopathy, necessarily it’s based on my experience, anecdotes if you will. However, the use of progressive resistance exercise for tendinopathy is now well established in the research, with some of my favorite papers being Peterson 2014, Kongsgaard 2009, and Arampatzis 2007. I am largely just applying that type of research to the muscle/tendons attaching to the coracoid process, and if needed, basic strength and conditioning to other muscles stabilizing the shoulder joint. So for coracoidopathy, an initial sample exercise program might be as follows:
- Standing Cable Rows (emphasizing scapular retraction)
- Lat Pulldowns, moderately narrow grip (~18-20 inches) bar to front, emphasizing scapular retraction at the bottom and full stretch at the top)
- Dumbbell Curls
- Rope Tricep Pushdowns
- Supinated Front Raise (SFR) (unilateral using the good shoulder first)
The SFR is the primary operating exercise in my program directed at decreasing pain and increasing shoulder flexion strength targeting the SHB and CB. Resistive exercises themselves, particularly the SFR, usually make the tendon feel immediately better and starts the healing process, such that if one were to wait for pain to decrease first, they will be waiting for a really long time.
I have my patients perform the SFR standing, one arm at a time (to better focus and prevent excessive body sway/back extension), and using the good shoulder first to obtain good technique before attempting it with the painful side. The elbow is slightly bent to lessen elbow stress and to increase SHB recruitment. The arm is fully supinated (palm up, trying to get the pinky higher than the thumb) to bring the SHB and CB muscles into what is likely their most advantageous position. The shoulder flexed to 90 degrees or where the upper arm is parallel to the floor. I think shoulder flexion to 90 degrees leads to the greatest tension and strengthening of the SHB and CB muscles, with minimal risk of shoulder impingement. Definitely check out the video, it’s way easier to just see it.
Over time the entire program will be progressed in intensity and then some of the following exercises may be added as needed.
- Standing Cable Internal Rotation
- Standing Cable External Rotation
- Dumbbell Pullovers (of my exercises I think pullovers best tension PM tendon, plus it’s a great shoulder stretch)
- Bent Over Lateral Raises
- Standing Cable Press (or Bench Press)
- Dumbbell Military Press
For rehab I start light and on each exercise, with the first set being a kind of warm-up. I have my patients do 3 sets (easy-medium-hard) of 15 reps. While I want most of the exercises to be painless, with the SFR I want the exercise intense enough to increase pain and if form is good I will continue to increase the weight, even if painful so long as it is not severe. If the person gets 15 reps with a full range of motion, pain or not, I increase the weight an increment for the second set. If they get 15 reps with full range of motion then, I again move the weight up for the third set. If the person gets full range of motion on the 3rd set, then the next day I have them start out with their ‘medium’ weight from before and move up from there. I want to quickly (over 2-3 days or so) get them to where they are unable to get all 15 reps on the third set. Only then do I stop increasing the weight until their strength progresses.
Besides the SFR, I don’t usually want my patients working through pain. I’m particularly cautious with pressing motions. However, if both the patient and I are sure that their pain is coming from the CP I’ll sometimes let them work through it and see what happens. If the pain lessens over time, it’s a good sign, if it worsens, it’s bad and I’ll reduce weight or remove the exercise. It’s a bit subjective and those calls are why I get paid the big bucks, and also why I’ll usually introduce the pressing exercises, one at a time, later into the program after giving the SFR a chance to work.
Once the patient is comfortable with the exercise, I recommend the SFR be performed daily (3 sets of 15 reps) with a home exercise program between therapy visits (preferably 3 times per week) where technique is observed and weights adjusted. My observation that SFRs performed daily speeds improvement, which was recently backed up by Lee 2014 who found resistance exercise performed 6 days per week superior to a frequency of 3 days per week in the treatment of lateral epicondylitis, which if you ask me is the same injury, just in a different location. The more comprehensive program (everything besides the SFR) if needed is performed at a lesser frequency (2-3 days per week) to build overall shoulder complex strength.
To start off I’ll select a weight on the light side of what I expect to be appropriate for the patient based on their overall fitness level and strength/sensitivity assessed when I do my Speed’s test on them. So a weaker and/or more sensitive patient might start with just a 1 lb dumbbell, while a stronger/less painful weightlifter might use as much as 10-15 lbs to start. Too heavy and they might strain something, too light and I might not get the analgesic effect of the exercise, so I would rather select a weight that is too light on day one and maybe add an additional set or two to get them to a good resistance level. If I start too heavy I might lighten the weight or repeat a weight on subsequent sets.
Reducing anxiety with painful exercise:
Sometimes patients have anxiety about lifting weights with a painful tendon. As such I give them ‘my talk’ telling them what I expect will happen. As mentioned above I tell them that tendinopathies generally warm up with exercise, which is why if I’m giving them a comprehensive exercise program I’ll usually put the SFR towards the middle or end of their program. Naugle 2012 specifically talks about exercise induced hypoalgesia (pain lessening), which I expect that is part of the equation. However, I think there is more specific pain reduction secondary to directly and progressively tensioning the painful tendon with a progressively higher load. I tell them, “This is what happens in ~95/100 of my tendinopathy patients.” I say they may have increased pain during their first set of their focused exercise (in this case the SFR) which usually returns to baseline as soon as they are done. I tell them if their technique is good and they get all their reps I intend to increase the weight on the second set. Many patients are apprehensive about this because the first set with the lighter weight caused some pain. I tell them that’s what I ALWAYS do, and 95/100 of my patients report the second set, even though heavier, doesn’t hurt more than the first. I tell them if everything goes as I expect, the second set has good form, and they can get all 15 reps, I ALWAYS increase the weight an increment for the third set. I say people sometimes have anxiety about further increasing the weight for the last set, but 95/100 of my patients say, “You know the third set hurt less than the first,” and pain afterwards is usually less than at baseline. Telling my patients that always puts them at ease, and they almost always confirm that the third set felt better than the first.
What to expect if it’s working:
With tendinopathy in general there are two common responses. The first is that the pain goes away completely in a few weeks (maybe 3-6), the person is a lot stronger and able to do everything without pain. The second common response is that the pain gets 90% better in 3-6 weeks, the patient is a lot stronger, but there is still a little nagging pain (maybe a 1-2/10 with activity) that can last maybe 3-6 months. While the second is the worst case, education from the start helps patients believe that it is still a good outcome. I used to rack my brain trying to figure out what to do with the residual 10% pain. Now I just accept it as what it is, STILL A GREAT OUTCOME, and if the person continues with their exercise program (if strength is fully restored and pain minimal I’ll reduce the SFR frequency to 2-3 times per week) eventually they’ll realize they don’t hurt anymore. However if they wimp out and get a cortisone shot, all bets are off.
What about the cases where the person has coracoidopathy plus RTC tendinopathy, they are a post-op RTC repair and/or they have adhesive capsulitis?
All of that goes beyond the scope of this blog, however in general, I’ll treat just as I would for those conditions (with some combination of stretching and strengthening as is appropriate) often very similar to my comprehensive exercise program above, plus stretches if ROM is limited. So basically I do my program for whatever else they have and add on the SFR. If it’s a post-op RTC repair, I wouldn’t add SFRs until 8 weeks post-op or so, when resistance exercises become appropriate, and my progression would be slower. If they are acute post-op I’m not going to even know if they have coracoidopathy, because palpation everywhere is going to be painful and I wouldn’t be doing any resistive tests.
I am a big fan of electric muscle stimulation (EMS) and I do use it for coracoidopathy and other tendinopathies but I think it’s benefits are definitely secondary to the exercises. I’ll usually place one electrode just inferior to the CP targeting the CB paired with the other electrode on the scapula targeting infraspinatus and teres minor, and I’ll usually put a second channel over the biceps and triceps. I figure on average EMS is good for maybe a 1 point drop in pain on a 10 point scale, which is less than I expect to get out of it for other conditions like back pain. So it’s certainly not essential.
What if the treatment doesn’t work?
The treatment responses I get with tendinopathy in general, and coracoidopathy in particular are consistent enough that if things don’t work out as I expect (the same and/or lessening pain as the person progresses from set to set, particularly by the third set, as well as relatively consistent pain relief day to day and week over week) then I begin to question my diagnosis. Also, if the patient notices catching pain that doesn’t warm up and feel better, or it makes them wince then I’ll start to suspect something worse, like a RTC tear or SLAP lesion in need of further diagnostic tests such as an MRI, and perhaps surgical correction. The Speed’s test has been shown to be a poor indicator of LHB or SLAP lesion pathology (as are all manual tests for LHB tendinopathy or SLAP lesions) however I think that’s because a positive Speed’s test is such a good indicator of coracoidopathy. However, if the patient has weak and/or painful Speed’s test, and they do not respond favorably to general strengthening as described, and focused strengthening with the SFR, it’s likely that something else is wrong in the shoulder with a LHB/SLAP lesion or RTC tear being at the top of the list.
As promised, here’s a video:
As always, if you have any further questions or need for clarifications, please don’t hesitate to ask. Being aware that some of my blog ideas are contentious and occasionally a bit out of the field of my expertise, I encourage my readers to come forth with any questions/comments that are of interest or concern. Your comments are valued and welcomed.
Chad Reilly, Physical Therapist