Common Shoulder Disorders

Grand Rounds


Robert G. Marx, MD, MSc, FRCSC

Attending Orthopaedic Surgeon, Hospital for Special Surgery
Professor of Orthopaedic Surgery and Professor of Public Health, Weill Cornell Medical College

  1. Rotator Cuff Disease
  2. Physical Examination
  3. Diagnosis and Treatment (Physical Therapy, NSAIDs, Steroid Injections)
  4. Surgery
  5. Shoulder Instability
  6. Glenohumeral Osteoarthritis
  7. Adhesive Capsulitis

 

Today we will talk about some common shoulder disorders that we all see in our offices in both rheumatology and orthopaedics and about my approach to some of these common conditions.

Common shoulder problems can really be classified into four areas: rotator cuff, instability, glenohumeral osteoarthritis, and fractures. We are not going to talk about fractures much because that really doesn't overlap at all. This is just sort of a gross classification of shoulder problems.

So what else is there? There are three other things that I am going to touch on briefly: adhesive capsulitis, acromioclavicular joint problems, and biceps tendon, which I think are all fairly common causes of patients to come with outpatient complaints to our offices. There are plenty of other things that cause trouble in the shoulder that we are not going to get into. These include: inflammatory arthritis, which you are really all the experts on; musculo-myofascial pain, which will commonly presents in the outpatient setting; brachial neuritis; Parsons-Turner Syndrome, which can be very painful and is not that common, fortunately; and nerve entrapments about the shoulder and C-spine, which can commonly present with shoulder pain and, therefore, it is really important to examine the neck any time someone presents with shoulder pain. The first thing in my physical examination is to check the neck and other things -- such tumors and uncommon things that can present as shoulder pain.

Rotator Cuff Disease

Rotator cuff disease is pretty much the most common thing that I see in the office in terms of patients presenting with shoulder pain. I will talk about my approach to diagnosis, how to stage rotator cuff problems and then the treatment -- both non-operative and a little bit on the operative management of cuff problems.

The pain when people present with rotator cuff disease is generally located in the upper lateral arm. There can be associated weakness, either due to pain or due to cuff dysfunction, so that the shoulder doesn't work properly. In the setting with trauma with weakness that is new, we have to consider an acute cuff tear. That is a separate scenario in which someone has a fall and then can't lift the arm or is very weak after having been able to use the arm normally before that fall. That is different from most cuff problems, which present with gradual worsening of pain and weakness over a period of months.

Physical Examination

I am not going to go over the whole physical examination, but a couple of points need to be made. It is really important to examine the acromioclavicular joint and the biceps. These can be separate sites of pain and pathology that are fairly easy to examine, but certainly can be separate from the cuff pain.

Examine range of motion, of course, including internal and external rotation. Internal rotation can be done with the arm up at 90 degrees, and in this position you can do internal and external rotation. I also like to examine internal rotation behind the back to see how high they can reach up in terms of touching the spinous processes compared to the normal side.

Cuff strength is, of course, important to examine, both external rotation with the arm at the side and supraspinatus muscle power in the plane of the scapula. Not way back and not out in front, but at 30 degrees in front of the plane of the body so you line the scapula with the thumb down is a good way to isolate supraspinatus muscle power. Lastly, provocative tests, so-called impingement tests. The Hoffmann test can be done out in abduction like this or across the body with internal rotation to impinge the cuff up on the acromion to see if that reproduces the pain. And also the Neer sign, with internal rotation and forward elevation straight up, can be done.

These things can often be positive in other conditions, but it all sort of goes together, putting the exam together in terms of pain on palpation, motion strength and provocative tests, among other things. I get radiographs on all these patients to rule out things that you are probably not thinking of but could be there such as a fracture, osteoarthritis of the glenohumeral joint, tumor, and calcific tendonitis.

Diagnosis and Treatment (Physical Therapy, NSAIDs, Steroid Injections)

MRI is also helpful if you a re considering that there may be a cuff tear or something else going on, e.g., other soft tissue pathology such as a labral cyst. I think that the stages of cuff disease described by Neer originally are really very helpful for the treatment, so I am going to talk a little bit about them. These are sort of the way he looked at it pathologically where initially there's inflammation in the subacromial space which, after months and months, goes on to fibrosis. When we do surgery in subacromial spaces, there is a certain amount of fibrosis where that bursa just gets thickened and scarred. Ultimately, as this disease progresses, you can have partial and then full thickness tearing of the rotator cuff. So without a tear in the cuff, the first two stages are really what we call impingement syndrome, bursitis, or tendonitis -- there are a lot of different words for this, but it is really the same thing. It's really bursitis in the subacromial space.

These stages affect how we treat things. I think the longer the symptoms have been present and the more severe the symptoms, then probably the higher the stage, the more scarring there is in the subacromial space and the less likely they are to resolve without injection or surgery. That doesn't mean that, just because someone has the symptoms for a long time, they shouldn't have a course of anti-inflammatories. But I think this has to factor into the thinking because, when someone has six months or a year of this untreated, there is more fibrosis and scarring in the subacromial space. And in my experience and in the literature, they don't tend to do as well as someone who has had symptoms for a month who generally, and in almost all cases, will resolve with physical therapy and anti-inflammatories.

So the first line of treatment for cuff disease, whether it is a tear or not, generally and depending on how long the symptoms have been going on, is physical therapy and anti-inflammatory medications. However, all physical therapy is not the same. There is really a large variation in the sophistication of physical therapists. There are some therapists at our hospital here who are just superb and really understand how to strengthen the cuff, strengthen the scapula muscle without inflaming the cuff and not pushing the patient too hard. That is really a very delicate balance. Then there are other places where patients may go for physical therapy where the therapy may be involving hot packs, ice and less sophisticated and advanced strengthening exercises. I think it is always important to ask patients what they are doing in therapy and how they feel after therapy. Particularly with shoulder patients, if they come out and say "Well, for two days after therapy my shoulder is killing me", that is a bad sign.

Now if the symptoms are not improving at six weeks, then I will generally inject the patient. If they are having some improvement, and depending on how keen they are to have a subacromial injection, I may or may not do it. Usually at six weeks, if things are not moving along, I will proceed with a subacromial cortisone injection and possibly get an MRI at that point, depending on how suspicious I am of a cuff tear.

Sometimes I will inject patients the first time I see them if they have had the pain for, say, six months. Usually, after a minimum of six months or a year and they had no treatment, then they are unlikely to get better with physical therapy and anti-inflammatory medication alone. So sometimes I will inject these patients the first time I see them.

I am going to discuss my method for injection in a minute. I use 80 mg of DepoMedrol and some local anesthetic, usually anywhere between 3 and 6 cc. I use 80 mg instead of 40 mainly because I think that you have room to inject it. So I guess if you can use it, that is what I elect to put in. I haven't had problems with it. I think it's not unreasonable to use 40, for example, in the acromioclavicular joint or when I inject the lateral epicondylitis, which I don't do very often, but in some cases I will. I will use 40 just because I don't want to put that much cortisone in there.

The subacromial space is a potential space with a lot of room and I want to try and make them better, so I think a nice high dose is safe in that area also. It is not like injecting the lateral epicondylitis where you are injecting into the tendon itself. This is really a potential space, so you shouldn't be injecting into muscle and tendon.

Question: What about the timing of the injection? Most of my patients are older and have contraindications to NSAIDs and anti-inflammatories. I tend to do injections sooner rather than later because I think it's safer. What about the balance between degenerative tears versus the benefit of anti-inflammatory injection?

A couple of excellent points here. The first thing is the elderly patient can't take NSAIDs. Yes, in that setting I am more aggressive in the first time cortisone injection because we have fewer options and surgery may be less of an option in these patients. Even repeated injections may be appropriate in those cases. The question about an MRI really relates to whether you think there is a cuff tear.

The other point you bring up is subacromial injections of cortisone in the setting of a cuff tear. Is that appropriate or, if the cuff is torn and they have symptoms, should it just be fixed? First, I do not think it's reasonable to inject the cuff tear if it is someone young who has significant weakness and symptoms and probably should be fixed. The other case is someone who is perhaps in their late 70s or early 80s who really doesn't look like someone who you want to be taking to the operating room. That is someone I would inject even in the setting of significant cuff tear and probably reinject.

That is a controversial area -- as to whether partial tears should be injected -- and my sense is in the absence of evidence that it is a bad choice, I would err on the side of being conservative and inject it. We don't know that you are going to cause the cuff to tear by injecting it. Certainly not once. Maybe five injections or more is probably not a good idea and I would definitely not recommend that. But with regard to a single injection in the setting of partial thickness cuff tear, there is certainly no evidence that it is in any way going to make the cuff tear and fall apart. But that is a controversial area.

The only reason not to do a cortisone injection in my mind -- and I'm not talking about five or multiple but one or maybe two injections -- really is if you are going to be operating on the patient some time soon and you don't want cortisone bathing the tissue. Other than that, I don't think a single cortisone injection is going to make the cuff become irreparable or deteriorate significantly. So I do inject partial thickness tears because I think they can get better. I think a lot of the stuff we see on MR, a lot of us may have partial thickness tears and they may develop subacromial bursitis and the partial thickness tear starts to hurt. Now a lot of those will get better and they live fine with the partial thickness tear.

So just because we pick up a partial thickness tear on MRI doesn't necessarily mean we have to operate on it. If they don't get better, that is a different story.

I think that ultrasound is very operator-dependent. With the ultrasonographers here -- and I am going to talk about some of their work later on -- I think it is very appropriate and I will get ultrasounds for cuff tears. Usually I will use it in the setting of someone who can't have an MRI because of size, claustrophobia, pacemaker, but with our operators I think ultrasound is reasonable. I think one of the main advantages of MRI is that it gives you a lot of other information, shows you the labrum that you don't see very well and it also shows some things maybe you are not expecting, such as a lesion in the humerus further down. It shows the muscle quality a little better. So I get more information on the MRI, but if you are talking about just evaluating whether or not there is a cuff tear, I think that ultrasound in the hands of a skilled examiner is reasonable.

Let's just talk about how I inject the subacromial space. Once you have a little bit of practice, it is not that hard really. I just mark out the acromion. I actually draw the anterolateral corner of the acromion on a patient with one of our markers that we get from the operating room. So then when we prep, it doesn't rub out. Then what I do is have the patient relax so you can pull down on their arm, which is not that easy to do, because they are nervous about having the injection. But by getting them to relax and pulling down on their arm, you can really open up the subacromial space and make it easier to get the needle in through the lateral approach.

I go 1 cm down and 1 cm posterior to the anterolateral corner of the acromion and just sort of shoot across. You should be able to get into the space in most cases pretty easily.

The other thing is that it shouldn't be that hard to inject the fluid. If your needle is in the right place in the subacromial space, then you should, with a 22 gauge needle, be able to squirt the solution in without too much resistance. So if you are having resistance, you may not be in the right place and it may be worth repositioning.

I like to keep the patients around after I inject them, mainly to check their response. Hopefully, they are not having any side effects with the injection, but I like to reexamine them, check the cuff strength and check the impingement signs to see if they are having any relief, because there is local anesthesia, and theoretically you are freezing the subacromial space and eliminating their pain.

If they don't feel better 15 to 30 minutes after the injection, I think that you have to consider you were in the wrong place. If you are experienced at this, then you are probably in the right place and if they don't have any relief at all, then I think you have to reconsider the diagnosis -- that maybe it is not coming from the subacromial space and the cuff. Usually there is at least some improvement after the injection. At least they will say they feel some relief.

If injection doesn't work, that is they don't feel any better afterwards -- I am not talking about in the office, but when they come back in a number of weeks and they are not better -- I think that you can consider a second injection. We kind of talked about this. I think it depends on multiple factors such as if you did the first one. If you didn't, then it is reasonable to do the second because you don't know that the quality of the first injection, the result of the injection -- whether it helped them for awhile and then they were good for three months and then kind of got worse or good for a couple of years -- then I would certainly consider reinjecting. If there is a tear present that you are probably going to be operating on if it is a young person with some weakness. For example, then reinjection is probably not indicated. Then of course there are patient demands and expectations.

Surgery

Once you have decided that non-operative management is no longer a reasonable option, surgery is then considered. That can involve a decompression where the acromion is flapping and the CA-ligament is released to make more space in the subacromial area and also bursectomy. I think the bursectomy is really important because sometimes I see the scarred, red inflamed tissue and I thing patients get relief after debriding that out with the shaver. Sometimes I will note fairly early on that they feel better even though they have had surgery, and I believe that it is because of the bursectomy and then, of course, repairing the cuff.

When is surgery indicated for cuff disease? Well -- impingement syndrome, bursitis that doesn't respond to non-operative management, or cuff tear in the same setting or acute cuff tear with significant weakness which. I think that it is important not to operate on chronic painless cuff tears with weakness. That is, someone who has weakness but no pain -- unless there is an acute injury, which is different because then they may have just pulled the cuff off and if you put them back you can give them the shape back -- but who has a chronic cuff tear that clearly has been there for a while based on MRI evidence of retraction, no history of injury, some fatty degeneration in the cuff, I think those are people that are not going to be happy if you operate on them because returning their strength is unpredictable and they really don't have pain that they are complaining of. So these people are risk of being unsatisfied.

I used one case example of a patient who is a 61-year-old man who is a very competitive tennis player and went to tennis camp a few weeks at a time during the year. He went to hit a ball and suddenly felt his shoulder go and tried to keep playing but couldn't, tried to get back to tennis but really was unable to, and here you can see his cuff tear involving the supraspinatus.

Looking at the anatomical areas, going from back to front, this is the posterior shoulder, so you see the infraspinatus coming up here. This is the teres minor. We are sort of out of the plane. These are a little tough to image in this plane. As we come forward you can see here is infraspinatus here, this is some of the injury you can see coming from below the spine. Here is supraspinatus so some of the infra is involved in this area, and here we are more anterior, we are above the spine now and this is supraspinatus. This is the acromion so this is the subacromial space, and you see that the supraspinatus tendon which is not coming out here and kind of pouring out here. We are even further anterior now, and here is the suprascapular knot with the suprascapular artery and nerve. And here you can see it is not too far from the glenoid remnant to release the cuff. We have to keep that in mind. Here is the supraspinatus muscle and there is the tendon and you can see it is clearly not attached.

So this gentleman, at six weeks after his injury when I saw him, was very weak and really couldn't lift his arm and wanted to go back to playing tennis.

[A brief video of the surgical procedure was shown.]

The cuff repair can be done arthroscopically, through portals instead of an open incision where the deltoid is being passed. The main advantage here is that it is less dissection. It takes just as long for the cuff to heal back. So the rehabilitation is the same, but it just is a little more comfortable for the patient. Sometimes they feel so good they want to move their arm and lift their arm before the biology has had a chance to work. So sometimes you have to hold them back a little bit as opposed to the open shoulder procedures.

Shoulder Instability

I am going to discuss the pathoanatomy of shoulder instability, a little bit about the natural history, which I think impacts a lot on the treatment plan, and the definitive management and surgery.

Anterior dislocations are by far the most common and most frequent dislocations seen in the emergency room, and the mechanism is usually abduction or external rotation. Here is the pathoanatomy of the dislocation, which is where the labrum gets pulled off the front of the glenoid, which is called a Bankart lesion. On the MRI in this axial cut, you can see the labrum nicely pulled off the glenoid.

This is an arthroscopy. You can see the labrum pulled off, this is the glenoid here, the probe is pulling the labrum off anteriorly and schematically where the Bankart lesion typically occurs on the pear-shaped glenoid anteroinferiorly.

The natural history is really the key when treating these injuries, because we know that people under 20 will have close to a 100% chance of recurrence, particularly if they participate in contact sports. They are almost guaranteed another dislocation at the very least. In contrast, in older patient sustaining their first shoulder dislocation, recurrences are very unlikely. In this graph you can see that in patients under 20, particularly the athletes, their chance of recurrence is almost 100% and, as we get up here in age, the chance of recurrence is very low. So the younger patients are the ones we are aggressive with in terms of operating, and the older patients I would say should never have an operation after a first-time shoulder dislocation.

I am not talking about multidirectional instability, where they have either some collagen disorder, or just very lax ligaments, whereby they begin to have instability without any trauma. These are generally patients who have either a high energy injury from a contact sport or at least your non-athlete who has fallen and dislocated their shoulder, but at least there is some level of trauma.

I think the recurrence, generally, if it is going to recur, will recur within a few years. It is uncommon to see people five to 10 years without having a recurrence. So probably the pathoanatomy of the injury and the patient's activity level will combine to yield a recurrence within a few years -- I would say within five. To have a recurrence beyond 5 years is somewhat uncommon, although it definitely happens.

Generally if the patients are not having surgery, they are immobilized. This may not make a difference. It may be more related to the injury and their activity level. Recently, there was a study presented at the American Academy of Orthopaedic Surgeons, in which a Japanese orthopaedic surgeon began to immobilize people in external rotation, believing you are pulling the capsule across and reducing the labrum back to where it should be. This actually makes sense as opposed to leaving the internal rotation and the capsule is lax and falls away from the glenoid. But this is a fairly early work, and we will have to see how this bears out.

So is surgery necessary? Well I think for recurrent dislocations, it is certainly appropriate in order to prevent further ones. It is really the only way to stop - further dislocations with activities of daily living or in people who want to play contact sports, and as I said, age is really a factor. I think that it is reasonable to repair a first-time 18-year-old college athlete. But for someone who is 40 years old and not very active in sports, definitely a first-time dislocation should be treated conservatively.

I am not aware of any influence of race or body habitus or size on recurrence rate. There has been shown to be a familial predisposition in that there are people who have a family history of a high rate of shoulder dislocations, but how that relates to race I am not sure.

I think the differences and outcomes across gender are certainly very complex and multifactorial issues, which I don't think we have a very good handle on. In the setting of shoulder instability, I think it is more common in males due to their involvement in contact sports. But as to the outcome from surgery, I don't think we understand the gender differences very well at this point.

So the surgery for shoulder instability can either be done arthroscopically or open. Arthroscopic surgery using thermal energy has fallen out of favor in most circles. We gained a lot of enthusiasm due to the fact that it is fairly easy to do; if you can arthroscope the shoulder, you can use this heat wand and theoretically shrink the capsule. But I don't think it is really very good for the tissue and hasn't worked very well overall.

So again, arthroscopic surgery can be performed for a lot of cases, but it really depends on the pathology. There are some cases where it should probably not be performed, such as when the ligaments are not detached on the labral side, such as in a Bankart, but pulled off the other side on the humerus. This is fairly uncommon but is termed a humeral avulsion of the glenohumeral ligament. If there are big glenoid or humeral sided defects, these should probably be done open as well, because they can dislocate into their bony defects and probably more soft tissue work is required in order to achieve this arthroscopically. And if there is poor quality of capsular tissue, that should probably also be done open as well.

Some cases, where maybe it should not be performed -- contact athletes. It is a little controversial as to whether they should be done arthroscopically. There are failed arthroscopic procedures, so perhaps you should do an open one for a higher degree of predictability. In glenohumeral arthritis or in a setting of multiple dislocations, where they have had quite a few dislocations and the capsule is quite stretched, you can get a better shift open and transfer more tissue to tighten it better than you can arthroscopically. If someone has had a lot of dislocations, I will recommend an open procedure.

Two case presentations of some shoulder instability can demonstrate cases that are a little out of the ordinary. We did an open surgery on a youngster who dislocated twice snowboarding. He does crazy things with his snowboard, and after his second one he decided he wanted to have surgery. His MRI was done at an outside facility, but they did it with gadolinium and actually did a great job. You can see the labrum that maybe is a little detached there, but here you can see the anterior capsule, where the labrum looks okay but the capsule looks like it's pulled off the humerus, which again is very uncommon. It is usually almost always on the labral side, and here this is pretty convincing.

So when we took him to surgery, we went in first with the plan to repair this humeral avulsion. You can see that the labrum is detached here from the glenoid and here as well, which we repaired arthroscopically with a couple of anchors. So we had planned to go ahead with the open procedure and did so. We looked over here for the lesion on the humeral side. You can see a little erythema, but you really can't see that it is very well detached off the humerus which is important. You can't really see these lesions that well arthroscopically

This is an open surgery, and this is the deltopectoral incision in the front of the shoulder. His arm extends down here, and this is his left shoulder from the front, so it is the front of the left shoulder. The pec is over here and the deltoid here, so we're in the deltopectoral interval. And this is the subscapularis tendon, which we are dissecting off the humerus to find the capsule below. Over here, there is no capsule, there is just a hole and you can see we dissected the subscapularis off. Further, there is a bigger hole, and when I got the capsule off altogether, you can see that the edge is very rounded and that wasn't going to heal back. Here is the subscap over here, tagged and retracted, and this was not going to go back to the humerus. Actually, when we tried to sew this tissue, it was very friable. Here is the completed dissection.

So now we are viewing with the humeral head retracted out of the way, looking into the glenoid. You can see our anchors here from our fixation; this is the capsule now completely detached around inferiorly, and we can now use this capsule. However, we want to pull it and shift it up in the humerus and tighten up the shoulder. Just an example where shoulder instability isn't always a Bankart lesion.

Another case is a 30-year-old man who had two prior operations, one arthroscopic and one open, and had over 50 dislocations. On the CT scan, you can see it looks like someone took a bite out of his humeral head here, and this is a Hill-Sachs lesion in the posterolateral humeral head. With these lesions, when they abduct and externally rotate, they can dislocate into that bony defect. A Hill-Sachs is essentially a fracture but recurrent dislocations sort of grind down the humeral head. The first one is really an impaction fracture, and then they just have multiple recurrent impaction fractures. , In the end, they have an indent in the posterolateral humeral head, which you can see.

This is sort of a chronic setting, where the bone, over time, has been impacted and remodeled into this position. This is cancellous bone; it is not healed like a cortical shock factor. This is lower down, and you don't see the defect. But basically you can see that you already have some chondral wear from these repeated dislocations, and this is a Hill-Sachs lesion. This is the edge of the glenoid, and you can see that the capsule isn't attaching. Actually, the anchors were placed a little too far medially, so we revised this open and removed the previous suture material. Here at the end of the procedure are the anchors that were used for the lateral shift. You can see the previous anchors were too medial. This one came out altogether, and our anchors are now reapproximating it more anatomically. He went on to do very well and has not had any recurrences about a year out.

Glenohumeral Osteoarthritis

Glenohumeral osteoarthritis, which I think you are all fairly familiar with, is somewhat uncommon compared to hip and knee arthritis. Certainly shoulder replacements are done -- about 1 for every 50 knee replacements. Generally, the treatment is activity modification, anti-inflammatories, and physical therapy, and then total shoulder replacement if these things are not effective. This is an example of a severely arthritic glenohumeral joint with complete joint space loss. I treated him with a shoulder replacement with a prosthesis in the humerus and a cemented glenoid on the glenoid side. These procedures are very effective for people with any stage of glenohumeral disruption.

The pain relief is predictable and the survival is similar to the total hip and total knee. Potential complications are significant, such as infection, instability in the shoulder, and potential for nerve injuries. So these are generally reserved for people who really fail conservative treatment and have pretty advanced glenohumeral arthritis.

As the X-ray showed, this can often be made very easily in end-stage arthritis because the shoulder motion is so limited and you just feel that horrible sense of crepitus as you move it. Basically, we do extensive soft tissue releases to recreate the anatomy and balance the soft tissue at the time of surgery. Often you don't have normal motion, but in some cases they can have near normal motion after the surgery, even though beforehand they had no motion. The lack of motion is due to soft tissue contractures, large osteophytes and pain and, with those things resected and the tissue released and put back in its anatomical position, the motion can be greatly improved and the pain also.

So-called rotator cuff tear arthropathy, which is also known as Milwaukee shoulder, occurs where a massive rotator cuff tear will allow the shoulder to migrate up and just degenerate and become very severely arthritic. It is very uncommon to get end-stage arthritis in the setting of a massive rotator cuff tear arthropathy. So for all the cuff tears out there, there are a very few that develop that and we don't really know why they do. Those are sort of subsets. By far the most common primary osteoarthritis is what we call wear-and-tear arthritis that we will see in hip and knee patients, and we don't really know why. It is probably genetic, but possibly related to activity. Sometimes we will see it in people who have had hips or knees replaced, and it is probably related to the genetic basis for that disease.

In osteoarthritis, I would definitely not do surgery to regain motion. You do it for pain. Generally I wouldn't do shoulder surgery just for motion, almost across the board. That is obviously a blanket statement, but for the most part, you want to be operating for pain, because that is going to leave the patient satisfied. These long- standing glenohumeral OAs develop significant soft tissue contractures around the shoulder. Releasing these contractures at the time of surgery, we completely release the capsule around the inferior pouch and dissect it off the humerus. We test it at the time of surgery to make sure we have adequate motion to take the front of the capsule totally off -- so that they should regain close to a normal shoulder motion. Now these are for arthritis, not the fracture situation, which is different.

But I wouldn't have someone come in with a little bit of pain, a little arthritis and say my shoulder is stiff and recommend a shoulder replacement. That is an important point.

Adhesive Capsulitis (Frozen Shoulder)

Adhesive capsulitis, also known as frozen shoulder, is really important to know about in the office setting. I am talking about primary adhesive capsulitis, not iatrogenic post-surgical things. We don't really know the etiology of this condition. It may be viral, and there are four stages that I think are really critical to understand, because they relate to the prognosis and the treatment.

Now these stages are somewhat controversial, but I believe quite strongly based on my experience and what I have learned here at this hospital, that this is really the way this condition works. Stage 1 is synovial inflammation. Classically, this is a woman in the 30 to 50 age range who wakes up one day and her shoulder is killing her. It is terribly painful and what they have is just an inflammation of the lining of the glenohumeral joint. Subsequently, this goes on to develop fibrosis over weeks to months, although the time frame can vary somewhat. Eventually, they get a fibrotic stiff shoulder which eventually thaws out.

But the point is that early on they often will just have synovial inflammation and, if their glenohumeral joint is injected with cortisone and lidocaine in the office, after the injection we often have marked pain relief from the freezing. That motion will return while the freezing is in indicates that they have not gone on to fibrosis. And in those patients, that acute injection of corticosteroid can often cure them. So if you can recognize that early stage and inject the glenohumeral joint with cortisone, you can really have a marked impact on that patient's quality of life for a number of months.

Now how do you diagnose it? There are a million patients who come in with sore shoulders. The hallmark is loss of external rotation. They really have a lot of trouble externally rotating because the main area of inflammation is around the coracohumeral ligament and external rotation pulls on that. So these patients won't be able to rotate externally at all. Again, it can be extremely painful.


^ Back to Top
Request an Appointment