This blog is probably only of interest to professionals since coracoid impingement syndrome (CIS) is thought to be a very rare, though important, shoulder pathology. CIS describes when the coracoid process is thought to physically impinge upon the lesser tuberosity of the humerus, pinching the subscapularis tendon and perhaps the long head of the biceps tendon in the process, thus causing pain and/or degeneration of said structures. Though even proponents of CIS report that it occurs infrequently, it has accumulated a fair amount of attention in the medical literature and is thought to be a potential cause of anterior shoulder pain either before or after surgery for it or other shoulder pathologies. While I wouldn’t say CIS is well known, it is at least known, which is more than I can say for what I think is the most underappreciated, easy to diagnose, and relatively easy to treat (for me at least) shoulder conditions; coracoidopathy.
This blog is a long footnote to my main work Coracoidopathy, the Missing Link in Shoulder Pain, so if you haven’t read that one, I’d recommend you do so now to better understand where I’m coming from with regards to CIS. Another footnote, but not essential to his blog is coracoidopathy and bench press shoulder pain.
What I intend to do here is go through the papers that I find most interesting regarding CIS, quote the relevant text, and compare and contrast with my thoughts regarding both CIS and coracoidopathy. I’ll also describe to what degree one diagnosis might be mistaken for the other. If I am correct in my observation that 84.7% of physical therapy shoulder referrals have some degree of coracoidopathy (and almost nobody knows about it) it stands to reason that it is more likely misdiagnosed as CIS (though I have never witnessed that happening in real life either) than vice versa.
It’s difficult for me to say anything definitive about CIS because I have never made the diagnosis, nor in the 17 years I have been a physical therapist have I ever been referred a patient with CIS either pre- or post-op. Also, as you will see, a couple of the ways CIS is diagnosed is by analgesic injection or direct observation in surgery, and as a physical therapist I can’t do either. However, since tendinopathy in general and tendinopathy of the coracoid process (which I call coracoidopathy) is one of my things, I thought it would be worth writing about if, for no other reason than to make it clear that I covered all foreseeable bases.
I’ll open up by quoting what Karim et. al. had to say about CIS in their paper describing “enthesitis of the biceps brachii short head and coracobrachialis at the coracoid process” (see why I had to give it my own name?) which shows they knew about CIS as well, but didn’t think it described at least some of the patients they saw. Without further adu:
Enthesitis of biceps brachii short head and coracobrachialis at the coracoid process: a generator of shoulder and neck pain. Karim MR1, Fann AV, Gray RP, Neale DF, Escarda JD. Am J Phys Med Rehabil. 2005 May;84(5):376-80.
“…[CIS] is hypothesized to be due to encroachment of the lesser tuberosity on the coracoid process because of overuse in flexion-internal rotation, which can lead to displacement of normal centering of the humeral head. Symptoms of coracoid impingement syndrome include anterior shoulder pain, occasionally radiating down the upper arm/forearm. Shoulder impingement signs are negative. Pain is made worse with passive flexion. On palpation there is tenderness over the coracoid process. The literature does not describe conservative treatment but rather describes confirmation of the diagnosis with local anesthetic injection between the coracoid process and the humeral head. This site of injection was different from ours, which was centered over the coracoid process. No conservative treatment was recommended or described in these studies. According to the studies, definitive treatment is coracohumeral decompression surgery, such as resection of the coracoid tip. This syndrome [CIS] seems to be similar to what we found in our patients in whom conservative treatment with steroid injection over the coracoid process improved or resolved our patients’ symptoms.
I don’t have anything to add to the above except to say it sounds right, and per the references and quotes cited below, it’s clear that Karim had a pretty good handle on the CIS data up to 2005. Afterwards there was some more guidance/suggestions as to how to conservatively treat CIS, which is fairly in line with how I would treat basic rotator cuff (RTC) tendinopathy, that being global strengthening of the scapular stabilizing muscles, focused strengthening of the RTC, and perhaps some stretching and postural adjustments (if needed).
“Physical examination revealed an exquisitely tender coracoid tip.”
“Forward flexion combined with medial rotation was chosen as the position in which to identify subcoracoid impingement by CT scan…”
“We documented idiopathic subcoracoid impingement (Figs 3-7) clinically, computer-tomographically and operatively in four painful shoulders, and we had sufficient clinical evidence of its occurrence in 21 others.”
“As in most cases of chronic idiopathic impingement, the symptoms could usually be treated conservatively.”
“We carried out isolated resection of the anterolateral coracoid tip only once; this was when, at operation, we had unequivocal proof of coraco-humeral impingement whereas there was no impingement against the acromion or the coraco-acromial ligament. This patient had a completely painfree, physically normal shoulder 18 months after operation.”
Most of the above sounds reasonable to me. While, not describing what conservative treatment is, they said it was usually effective. However, I would think that if the coracoid process was impinging on the lesser tuberosity/subscapularis/long head of the biceps, I would think those areas would be exquisitely tender, more so or at least equal to the coracoid tip. If the tenderness was isolated over the coracoid process (especially if confirmed with a weak or painful Speed’s test) then I would immediately suspect coracoidopathy. This makes me wonder if the 21 painful shoulders they diagnosed based on clinical evidence, that was not confirmed operatively or with a CT scan, were in fact coracoidopathy misdiagnosed as CIS.
“The seven patients were seen over a period of six years.”
“All patients had tenderness in the anteromedial shoulder over the coracoid process.”
“All patients had anterior shoulder pain made worse by forward flexion and medial rotation combined with horizontal adduction. This manoeuvre produced a painful click in seven of the 8 shoulders.”
“Previous operations. Five patients had one or more surgical procedures, including diagnostic arthroscopy, glenoplasty, tenodesis of the long head of the biceps (four patients), and acromioplasty (two patients).”
“Four patients had arthrograms pre-operatively. None of these helped with the diagnosis.”
“After this, either the anterolateral aspect of the tip of the coracoid process or the whole tip is removed to open the coracohumeral space (Fig. 2a), about 1 to 1 .5 cm being removed. The conjoined tendon is then re-attached to the remaining base of the coracoid process (Fig. 2b). We now recommend the more radical excision.”
“Six shoulders had complete resolution of preoperative symptoms and two had mild residual discomfort during stressful activity in the overhead position, but no pain at rest.”
“Coracoid impingement is a rare cause of anterior shoulder pain, which should be considered especially after the failure of previous surgery for impingement, biceps tendon lesions, instability, or an operation on the coracoid…”
Interesting, in this paper was the rareness of the CIS, 7 patients in 6 years. I have the same reservation as in the Gerber study in that they noted tenderness over the coracoid process, but they didn’t report further to see if they were tender over the lesser tuberosity of the humerus, where the coracoid process should be impinging. Two had a prior acromioplasty and five had prior prior tenodesis (the tendon was cut off), which makes me wonder if either or both might have encouraged anterior translation of the humeral head resulting in the coracoid impingement. Also, rotator cuff weakness/tendinopathy generally precedes long head of the biceps tendinopathy, for which it is often cut, and also precedes acromioplasty. As such, I think strengthening the rotator cuff and scapular stabilizing musculature would go a long way towards treating CIS and preventing the need for coracoid process excision. It does seem the excision helped, however if any of the patients did in fact have coracoidopathy, cutting the short head of the biceps, coracobrachialis, and pec minor loose, to reattach them on a shortened coracoid process might (in a roundabout way) eliminate the muscle/tendon pain as well as the impingement. Previous conservative treatment in this study was not described but if it were me I’d definitely want to give scapular stabilizing, rotator cuff exercises a try first, most certainly combined with my targeted exercise for coracoidopathy, the supinated front raise (SFR).
“From 1984 to 1994 we operated on almost 500 patients with shoulder impingement syndrome. In reviewing the surgical records, with found that 12 patients (14 shoulders) had impingement of the rotator interval by the coracoid process or fibrous falx at the time of surgery.”
“Preoperative imaging techniques were no help in diagnosis, even retrospectively…”
“Six patients had previously undergone shoulder operations. Three had been operated on for calcific tendinitis and 2 for subacromial impingement.”
“Although we were aware of the possibility of coracoid impingement from the start, we discovered most of the impingement lesions during surgery.”
“We believe that the tests devised by Yocum and Hawkins and Kennedy are most likely to yield positive results in cases of coracoid impingement. However, as Patete has already pointed out, none of these tests are specific…”
“Contrary to authors who initially addressed the issue we believe that cases caused by an excessively lengthy coracoid process are rare and that no coracoid process can induce a cam effect.”
“We chose to resect the coracoid process no further than the extent necessary to relieve impingement; we did not follow the recommendation of Dines et al that the tip of the coracoid process be removed.”
“Three patients (4 shoulders) still had a fair amount of pain… Seven patients (9 shoulders) reported shoulder weakness…”
“Whether coracoid impingement exists as a distinct syndrome and, if so, how variations of it should be classified, has yet to be fully established. We believe that lesions of the biceps and the subscapularis should not be attributed to coracoid impingement syndrome.”
This paper all sounds very reasonable to me. 12 patients in 10 years, and 12/almost 500 impingement surgeries is “almost 2.4%” which does seem to back up the idea that CIS is uncommon, which is contrary to my findings coracoidopathy. I also like that they took a minimalist approach to excision of the coracoid process. Considering several of the patients still had shoulder pain and weakness after the procedure, it makes me wonder if they had coracoidopathy in addition to CIS, with the residual pain and weakness being secondary to the coracoidopathy still being there. As often as I find coracoidopathy with RTC tendinitis (74% of the time) I’d be surprised if they didn’t.
“Previously reported cases of failed cuff surgery are as follows: (1) infection, (2) deltoid denervation, (3) deltoid detachment, (4) failure of the cuff repair, (5) rehabilitation failure, (6) re-tear, (7) remaining subacromial roughness, and (8) loss of superior stability. However, subcoracoid impingement syndrome has seldom been mentioned as a cause of failed anterior acromioplasty and rotator cuff surgery.”
“Eleven cases of subcoracoid impingement syndrome were clinically diagnosed among 216 cases in which anterior acromioplasty and management of rotator cuff tear had been performed.”
“The radiographic findings revealed slight upward migration of the humeral head in 8 cases and bony spurs of the coracoid and humerus in 5 cases. Bilateral CT scans were obtained in 7 cases. On CT scanning, the coracohumeral distance in the affected shoulder was seen to be, on average, 57.6% of that on the contralateral side…”
“Subcoracoid impingement syndrome was diagnosed in terms of effective subcoracoid block. Effective subcoracoid block means that the symptom is alleviated after injection with 1.56 mL of 1% lidocaine and 0.5 mL of contrast medium fluoroscopically the site just posterior to the base of the coracoid.”
“Coracoplasty was performed in 9 cases in which conservative treatment, consisting of physical therapy for more than 1 year and steroid injection at the base of the coracoid, had failed.”
“Histopathologic findings of resected coracoid processes revealed hypertrophic changes to the fibrocartilage layer and the osseous trabecula at the posterior aspect. These findings suggest mechanical compression to the coracoid process of the tendinous rotator cuff or humeral head.”
Results of this paper again sound reasonable. They didn’t report much in the way of hands on tests, and myself not having any experience with analgesic blocks, I do wonder if the subcoracoid block might also numb up the tip of the coracoid process similar to the analgesic injection given by Karim. They did noted the block was considered effective if pain with horizontal adduction was alleviated, which makes might or might not distinguish it from Karim’s injection. Also, while they gave a picture of hypertrophic changes in the resected coracoid process, they did not say in what percentage of resected coracoid bone/cartilage this was found. Patients were given a year for physical therapy to work, which seems like more than enough time. However, what constituted physical therapy was not described, so it could have been progressive resistive exercise to the scapular muscles and RTC, which I think would help in the case of CIS, and be part of the treatment for coracoidopathy, but without a SFR I would expect coracoidopathy symptoms to remain. Physical therapy could have been, cold, heat, massage, stretching, ultrasound, acupuncture (the study was done in Japan), etc., all of which I expect would be relaxing but ultimately worthless (beyond placebo) for either condition.
“As opposed to sub-acromial impingement, an irritation of the subscapularis tendon seems to occur almost exclusively in shoulder joints that are pre-damaged or occurs post-operatively (subscapularis cyst or after arthroscopy, glenoplasty tenodesis of the long head of the biceps tendon, or acromioplasty)”
“…rupture of the subscapularis tendon seems to be much less frequent than the lesion of the upper/posterior cuff.”
“We found ruptured subscapularis tendons to be the exception, only to be encountered when massive defects occurred in combination with lesions of the upper rotator tendons.”
“The clinical symptoms of coracoid impingement primarily consist of pain in the anterior area of the shoulder above the lesser tuberosity with increased internal rotation and anteversion—the impingement position. Internal rotation is limited because of pain and can barely be performed actively if the subscapularis tendon is completely ruptured.”
“A comparison of the subscapularis tunnel area between shoulders with and without degenerative changes or a ruptured subscapularis tendon revealed no signiﬁcant differences (428 mm2 as compared with 427 mm2 on average).”
“There seems to be a general disposition toward coracoid impingement, as the lesser tuberosity always makes contact with the coracoid process within the normal range of motion of the shoulder.”
“…no signiﬁcant difference was found between the tunnel areas of degenerated and healthy subscapularis tendons. Thus, a relationship between the coracoid impingement and the size of the subscapular tunnel area representing the size of the subcoracoid space can deﬁnitely be excluded.”
“In conclusion, a coracoid impingement does not seem to be caused by anatomic variations as the main pathology. Instead, it seems to be the result of a functional problem, mainly anterior instability of the shoulder joint leading to a functional narrowing of the coracohumeral distance.”
Great paper, I don’t have any additional comments except to say it’s fascinating and sounds right.
That’s all for the pre-Karim papers, below are a few that were published after Karim’s 2005 paper.
“…none of the subcoracoid impingement patients clinically suspected of having subcoracoid impingement had directly evident soft-tissue impingement or abnormal focal bone or soft-tissue signal.”
“Subcoracoid impingement is clinically manifested as anterior shoulder pain after internal rotation, adduction, and forward flexion. These maneuvers are reproduced by the military parade rest position, perhaps explaining the predominance of soldiers among our patients with true-positive findings.”
“Because subcoracoid impingement clinically is more apparent in internal rotation, perhaps repositioning the patient’s shoulder in internal rotation would yield more accurate measurements. However, given the prevalence of rotator cuff abnormalities and the relatively uncommon occurrence of subcoracoid impingement, positioning the patient’s shoulder in a standard neutral or external position is more practical and efficient.”
“The average coracohumeral interval for females was 3 mm smaller than that for males. Using sex-adjusted data, we found a statistically significant difference between individuals with or without subcoracoid impingement in the axial coracohumeral interval (p = 0.01). This value, however, was poorly predictive…”
“A sex-adjusted coracohumeral interval of 10.5–11.5 mm, although statistically significantly related to subcoracoid impingement, is poorly predictive of this diagnosis when acquired via routinely performed MRI.”
“The morphology of the coracoid process and lesser tuberosity has no apparent predictive value.”
This one is interesting because all 19 subjects were confirmed to have coracoid impingement per surgery and much like prior researchers reported with x-rays or CT scans, this one found an MRI didn’t help either. No data was given with regards to other clinical tests that might or might not have helped with pre-operative diagnosis, nor what other shoulder pathologies were present at the time of surgery. They seem to think that standing in the parade rest position was maybe causative of pain but that sounds unlikely to me as while the parade rest position has a lot of shoulder internal rotation, it lacks the the flexion and horizontal adduction reported to be needed to really hit the coracoid process, maybe.
“Patients with this syndrome typically present with a history of dull pain in the anterior aspect of the shoulder that is exacerbated by activities performed when the shoulder is in a forward flexed, adducted and internally rotated position.”
“The coracoid impingement test is performed in a manner similar to that used to perform the Kennedy–Hawkins impingement sign, except that the patient’s shoulder is placed in a position of cross arm adduction, forward elevation, and internal rotation to bring the lesser tuberosity in contact with the coracoid .”
“The coracoid impingement test is passive on the part of the patient, and this is used to differentiate it from the superior labrum test which requires the patient to resist downward pressure on the arm.”
“Another method that is used to evaluate whether a patient has coracoid impingement is to inject lidocaine into the subcoracoid region. Although it has been suggested that relief of pain with this injection can help to establish the diagnosis, this test has not been studied to determine either its validity or its accuracy. The proximity of multiple structures in the subcoracoid region, including the joint itself, makes the accuracy of these injections questionable.”
“The first line of treatment for coracoid impingement should be a program of activity modification, with avoidance of the provocative positions of forward flexion and medial rotation, and physical therapy to strengthen rotator cuff muscles and scapular stabilizer musculature.”
This is a review paper for which I am again largely in agreement. However, I’m skeptical that the O’Brien’s test (the active test they cited for superior labral pathology) is particularly diagnostic, in part because of this. I do like their use of the term “coracoid impingement test” because “passive shoulder flexion with internal rotation and horizontal adduction” is awkward to both say and write. I just don’t know if the “coracoid impingement test” works as well in practice as it sounds in theory.
My other comment would be with regards to treatment. As I have commented above, I do think scapular stabilizing and RTC muscles should be strengthened, but I would definitely add in the SFR.
The SFR being performed with active shoulder flexion and external rotation shouldn’t irritate CIS, and if ineffective would help rule out coracoidopathy and help rule in CIS. They mention activity modification to avoid provocative positions of flexion and internal rotation and I think that would help with distinguishing the two as well. With tendinopathy (in this case coracoidopathy), increased activity seems to warm up the tendon and make it feel better. This is also something I usually notice immediately with my treatments of SFRs. Physical impingement/impact however, I would expect to not warm up and feel better but rather worsen with increased stress of the offending activity.
Subcoracoid impingement syndrome: a painful shoulder condition related to different pathologic factors. Garofalo R, Conti M, Massazza G, Cesari E, Vinci E, Castagna A. Musculoskelet Surg. 2011 Jul;95 Suppl 1:S25-9.
“This study includes 13 consecutive patients suffering from subcoracoid impingement symptoms who underwent an arthroscopic treatment. Patients were identified from a database of 1678 consecutive patients treated with arthroscopy because of shoulder problems during a period of 3 years.”
“All patients underwent an unsuccessful minimum 4-month course of conservative treatment consisting of activity modification, nonsteroidal anti-inflammatory medication, physical therapy and supervised exercise regimen before undergoing surgery.”
“At one year of follow-up, VAS score decreased significantly passing from a mean preoperative value of 7.7 points (range 6-10) to a mean value of 1.2 points (range 0-3).”
“The first line of treatment for coracoid impingement is a physical therapy program consisting in avoiding provocative positions, postural exercises to modify scapular anterior tilt, and exercise to strengthen rotator cuff muscles; however when this syndrome is refractory to non-operative treatment for at least 4-6 months, the patient may be a candidate for arthroscopic surgery.”
I thought this was a good paper, again demonstrating rarity of the condition, 13 out of 1678 shoulder arthroscopies, is 0.77%. The post-operative rehab program was well described and reasonable, as were the pre-operative rehabilitation suggestions. HOWEVER, again I would suspect a fair number of patients with coracoidopathy to not have their pain satisfactorily resolved with an exercise program that didn’t directly target and strengthen the short head of the biceps and coracobrachialis muscles. Thus, with the above recommendations, not appreciating coracoidopathy as a diagnosis in and of itself, might result in an unnecessary arthroscopic surgery for which no signs of CIS were found, nor much of anything else was found. The result might be an acromioplasty while you are in there, because what else are you going to do? And this still wouldn’t fix the coracoidopathy! Also unnecessary acromioplasty may leave the shoulder less stable, perhaps predisposing one to legitimate RTC, CIS or other shoulder problems in the future. So I definitely wouldn’t let “anterior shoulder pain” be good enough as a complaint. Among other things the evaluation should including range of motion in planes and resistive tests to all relevant muscles. I’d definitely perform a Speed’s test (as a muscle test), as I think the short head of the biceps and coracobrachialis are particularly relevant. Plus directly palpating the coracoid process, adding SFRs to the exercise program if indicated.
“Goldthwait first described coracoid impingement in the anterior aspect of the shoulder in 1909.”
“Gerber et al were the first to describe surgical management of coracoid impingement.”
“In asymptomatic patients the coracoid itself may be tender to palpation. Such tenderness is not a reliable sign of coracoid impingement…”
“A patient also may present with weakness of the subscapularis and pain on biceps testing (eg, Speed test…”
“It is important to note, on clinical examination and imaging studies, associated pathology that may be causing the anterior shoulder pain, such as lesions of the subscapularis, rotator interval, biceps tendon.”
Freehill writes a good history of CIS and points out what I would expect to find with CIS, which is tenderness in the interval between the coracoid process and lesser tuberosity. The downside, is that it’s pretty difficult to determine if that tenderness is due to impingement (pinching) between the two bones, or just tendinopathy of the subscapularis and maybe the long head of the biceps. He correctly notes that tenderness over the coracoid process is not a reliable indicator of CIS but does not appear to have read Karim’s paper to suspect tendinopathy at the coracoid process. He even notes, there may be pain with a Speed’s test. He’s so close! If he had only pluralized “biceps tendon.” Oh well, I guess we can’t all be visionaries.
Overall from my research I saw further confirmation diagnosis coracoidopathy (tendinopathy of the muscles attaching to the coracoid process) was not a consideration in any of the papers prior to Karim 2005, or in any of the papers after Karim. So the word definitely did not get out.
In general, CIS appears to be such a rare condition that it’s never hitherto been a consideration for me. I was well into developing my diagnosis and treatment for coracoidopathy before I gave it much thought. Preparing for this particular blog, I have given CIS more thought and just recently I was treating a patient for post-op lumbar fusion and she said “you know, my shoulder’s been hurting a lot too.” I asked where in the shoulder the pain was coming from and she had trouble saying where but finally pointed to her lateral deltoid (so much for all that talk about anterior shoulder pain). I did my basic shoulder tests, to include palpating the CP, which was very tender (while nothing else was) and my Speed’s test was both weak and painful (while nothing else was). I thought to add on a test of passive shoulder flexion/internal rotation/horizontal adduction, which was not provocative. I’m not sure if the coracoid impingement test is worthwhile or not, but I thought to throw it into my shoulder evaluations for a while and see if I could learn anything from it, and if it leads to any greater accuracy in shoulder diagnosis.
So, if I had a patient with painful passive shoulder flexion/horizontal adduction/internal rotation, but with a strong and painless Speed’s test combined with tenderness over the coracoid process, the lesser tuberosity of the biceps and intervening soft tissue, and they didn’t respond favorably the SFR as part of a comprehensive scapular, UE, and RTC strengthening program then CIS might be a consideration. Nine times out of ten (or more) I’d expect that anterior shoulder pain and tenderness of the coracoid process is going to be coracoidopathy and my exercise program will work. I would expect if there is CIS there is likely coracoidopathy and RTC dysfunction as well, thus my SFR combined with my comprehensive RTC-type treatment program sounds right on the money for conservative treatment for CIS anyway. If that doesn’t work, it’s probably time for an MRI or a surgeon to go in there and fix what he finds.
Next, in this hotly anticipated series will probably be coracoidopathy vs long head of the biceps tendinopathy and SLAP lesions. Anyway, review that peers.
Chad Reilly, Physical Therapist