Shoulder:The Stiff Shoulder

From WikiBeemed
Jump to navigation Jump to search

Primary Frozen Shoulder

From [1], with permission.

Bullet Points


Frozen shoulder, also known as adhesive capsulitis, is an insidious painful condition of the shoulder persisting more than 3 months. This inflammatory condition that causes fibrosis of the glenohumeral joint capsule is accompanied by gradually progressive stiffness and significant restriction of range of motion (ROM), typically external rotation. However, the patients may develop symptoms suddenly and have a slow recovery phase. The recovery is satisfying in most of the cases, even though this may take up to 2 to 3 years.[2][3]


The etiology of frozen shoulder is not yet fully understood. However, some plausible risk factors have been identified:

  • Diabetes mellitus (with a prevalence up to 20%)
  • Stroke
  • Thyroid disorder
  • Shoulder injury
  • Dupuytren disease
  • Parkinson disease
  • Cancer
  • Complex regional pain syndrome


Frozen shoulder occurs in up to 5%. Women are 4 times more often affected than men, while the non-dominant shoulder is more prone to be affected.[2][3]


Frozen shoulder is usually described as fibrotic, inflammatory contracture of the rotator interval, capsule, and ligaments. However, the development of frozen shoulder remains not fully understood. Although disagreements exist, the most recognized pathology is cytokine-mediated synovial inflammation with fibroblastic proliferation based on arthroscopic observations. Additional findings include adhesions around the rotator interval caused by increased collagen and nodular band formation.

The structure usually affected first is the coracohumeral ligament the roof of the rotator cuff interval. Contraction of the coracohumeral ligament limits external rotation of the arm, which is usually first affected in early frozen shoulder. In advanced stages, thickening and contraction of the glenohumeral joint capsule develop, further limiting the range of motion in all directions.[3]


The studies of the glenohumeral capsule histopathology have confirmed a significant increase in fibroblasts, myofibroblasts, and inflammatory cells, like B-lymphocytes, mast cells, and macrophages.

History and Physical

Patients suffering from early frozen shoulder usually present with a sudden onset of unilateral anterior shoulder pain. The typical symptoms comprise passive and active range of motion restriction, first affecting external rotation and later abduction of the shoulder. In general, depending on the stage and severity, the condition is self-limiting, interfering with activities of daily living, work, and leisure activities. Functional impairments caused by frozen shoulder consist of limited reaching, particularly during overhead (e.g., hanging clothes) or to-the-side (e.g., fasten one's seat belt) activities. Patients also suffer from restricted shoulder rotations, resulting in difficulties in personal hygiene, clothing and brushing their hair. Another common concomitant condition with frozen shoulder is neck pain, mostly derived from overuse of cervical muscles to compensate the loss of shoulder motion.

The physical findings are essential for a frozen shoulder diagnosis, although pain and stiffness make it difficult for patients to comply with a complete set of physical examination.

Two physical examinations are commonly used to diagnose frozen shoulder, including tests of combined motion, as touching the scapula from behind the neck and from behind the back.

However, the most pathognomonic feature for frozen shoulder is a loss of passive range of motion (ROM). Practically, in cases of significant restriction of passive ROM, an examination of active motion can be skipped. Nevertheless, as an undetectable limitation of shoulder motion may be present in the early stage, frozen shoulder diagnosis should be reconsidered in patients who present with a gradual restriction of range of motion at follow-ups.

In general, patients with frozen shoulder usually demonstrate significant restriction in active and passive range of motion, particularly in external rotation and abduction movement. Restricted motion in every direction not only indicates the presence of a developed frozen shoulder, but it may be a “red flag” for possible underlying malignancy or fracture.


Frozen shoulder is a clinical diagnosis made by medical history, physical examination, and imaging modalities (ruling out another condition, rather than confirming the diagnosis of AC). No specific test (laboratory or imaging) alone provides the definitive confirmation of the frozen shoulder diagnosis.[2][3][4]

Diagnostic imaging of frozen shoulder can be challenging because the findings based on currently available methods (such as radiographs, ultrasound, plain magnetic resonance imaging, and computed tomography) are usually unremarkable. Imaging is therefore limited to ruling out concurrent pathologies, like rotator cuff tendon tears and glenohumeral joint osteoarthritis. The imaging tool mostly applied to patients with AC is high-resolution musculoskeletal ultrasonography (MUS), which has emerged to be the first line to scrutinize shoulder pathology. Nevertheless, until now, there is lack of specific ultrasound findings for diagnosis of frozen shoulder. To be more comprehensive, several investigators reported thickening of the coracohumeral ligament to be a sonographic characteristic in patients with frozen shoulder. Another commonly referred ultrasound finding is the presence of fluid accumulation around the long head of the biceps tendon. Although biceps peri-tendinous effusion is prevalent in shoulders with frozen shoulder, it is specific to frozen shoulder because it might be a result of other shoulder pathology (e.g., rotator cuff disorders or biceps tenosynovitis). Furthermore, ultrasound is less useful in the pathology associated with instability or superior labrum anterior posterior (SLAP) lesion. A plain radiograph is of limited diagnostic values in patients with frozen shoulder. Nevertheless, it is reasonable to obtain routine shoulder radiographs to rule out other etiologies (e.g., tumors, acromioclavicular and glenohumeral osteoarthritis). Magnetic resonance imaging (MRI) may show thickening of the coracohumeral ligament and glenohumeral joint capsule. MRI arthrography may show a volume reduction of the joint space.

The “lidocaine test” is a subacromial injection test that may be helpful in establishing the diagnosis in ambiguous clinical scenarios, to rule out subacromial conditions. In patients with AC, passive movement limitation persists after injection of local anesthetics into the subacromial space. On the other hand, patients suffering from subacromial impingement syndrome (e.g., pathology of the rotator cuff or bursa) usually experience improved passive range of motion after injection. The injection can easily be performed with ultrasound guidance.

Treatment / Management

Despite the number of published literature on the frozen shoulder, there is no consistent consensus about its management. The majority of treatment options are non-operative and include pharmacological management and physical therapy.[5][6]

Early Frozen Shoulder

An early stage of AC is often managed as subacromial pathology. The early “freezing” AC mentioned above can be considered as inflammatory. On the other hand, the inflammation becomes less accentuated in the later stages, where ROM limitation is predominant, and inflammation-related pain is not as much pronounced. In the light of above differences, we must consider the disease stage when planning the treatment strategy. The correct recognition of the clinical stage may help to tailor treatment plans more specifically. The treatment aim in the “freezing” stage should focus on pain control, reduction of inflammation and patient education. Initial treatment options for adhesive capsulitis may include acetaminophen or NSAID. Although evidence regarding NSAID for the treatment of frozen shoulder is limited, they can be prescribed to provide short-term relief from the night pain if present. However, in severe cases, opioid analgesics may be required. Physical therapy is important for pain control and restoration of normal shoulder mobility. Those comprise manual therapies based on soft tissue mobilization and gentle stretching. Regarding physical modalities, no particular agent has shown to be superior. The patient can be prescribed, e.g., therapeutic ultrasound, cryotherapy or transcutaneous electrical nerve stimulation (TENS) unit. Physical therapy management should focus on therapeutic exercise. Although not all patients can tolerate mobilization exercise within the initial stage of frozen shoulder due to severe pain, a supervised therapeutic exercise should be conducted to slow down ROM restriction. Furthermore, a home exercise program should be given to the patients on a daily basis. In patients suffering from moderate to severe pain who are not responsive to non-operative treatments, intra-articular injection of corticosteroid should be considered. The injection should be performed under ultrasonographic or fluoroscopic guidance to ensure the correct needle placement. Last but not least, rehabilitation exercise should be prescribed after injection.

Developed Frozen Shoulder

After the inflammation-related painful period subsides, the condition progresses to a “frozen” and subsequently into a “thawing” phases. Treatment objectives in the advanced stages should focus on regaining ROM limitation. The physical therapists should provide more intensive (compared, e.g., to subacromial pathologies) mobilization exercise to restore joint mobility. In patients who do not respond well to non-operative treatments, a more invasive therapy should be considered. The addition of suprascapular nerve or interscalene brachial plexus blockage may result in further improvement. In patients with refractory cases of frozen shoulder, who do not improve after 6 months of non-operative treatment, more aggressive treatments such as capsular hydrodilatation (stretching the joint capsule by the saline injectate pressure), manipulation under anesthesia (tearing of the contracted capsule), and arthroscopic capsular release (particularly in the rotator interval) can be considered.

Differential Diagnosis

Frozen shoulder, particularly in early (freezing) stage might be a diagnostic challenge as it may mimic subacromial pathology and rotator cuff tendinopathy. Mentioned above presentations may result in the delay of AC diagnosis in the early phases. Regarding shoulder impingement and rotator cuff pathology, patients report predominantly pain with less pronounced passive range of motion. However, several facets help to distinguish frozen shoulder from other shoulder disorders. Regarding the causes other than frozen shoulder, patients often state lifting a heavy object or performing repetitive overhead movements. In contrast, frozen shoulder patients usually describe spontaneous onset without an apparent cause or a history of overuse activity. Extra precaution should be paid in case of the history of malignancy.

Common conditions that may mimic early adhesive capsulitis:

  • Glenohumeral joint arthritis
  • Post-stroke shoulder subluxation
  • Posterior glenohumeral dislocation
  • Referred pain (cervical spine or malignancy, e.g., Pancoast tumor)

Age of onset provides additional clues to diagnose AC. Frozen shoulder is unlikely in patients younger than 40 years old, and patients older than 70 years old are more likely to develop rotator cuff tears or glenohumeral osteoarthritis than frozen shoulder.


Gradual restriction of passive shoulder motion characterizes a natural course of frozen shoulder. The development is commonly described as progressing through 3 overlapping phases (4 stages classification can also be found in the literature).

However, from a practical point of view, we recommended using 2-stage scheme: early and developed frozen shoulder.

Freezing (2 to 9 months): Early

Frozen (4 to 12 months): Developed

Thawing (12 to 42 months): Developed

Freezing is an initial, painful phase with predominant pain that is worse at night, with gradually increased glenohumeral joint ROM restriction.

Frozen is the second phase with stiffness and persisted glenohumeral joint motion limitation, but with less pain than that at the “Freezing” stage.

Thawing is the third (recovery) phase with the gradual return of range of motion.


The duration of frozen shoulder is from 1 to 3.5 years with a mean of 30 months. In about 15% of patients, the contra-lateral shoulder becomes affected within 5 years.


General complications are:

  • Residual pain
  • Residual stiffness
  • Fracture of the humerus
  • Rupture of the biceps tendon after shoulder manipulation

Pearls and Other Issues Considering the diagnostic accuracy of frozen shoulder, researchers should keep investigating its pathomechanism. Some studies have recently reported the application of contrast-enhanced ultrasonography in the diagnosis of frozen shoulder. Application of the microbubble-based ultrasound contrast agents (increasing a liquid substance echogenicity) in musculoskeletal medicine has already been adopted for selected indications. Looking ahead, the utility of contrast agents in a frozen shoulder diagnosis seems to be promising particularly in ambiguous cases.

Enhancing Healthcare Team Outcomes

The management of frozen shoulder is usually done by a team of healthcare professionals that include an orthopedic surgeon, a rehabilitation specialist, a nurse practitioner, a pharmacist, and a pain consultant. In addition, the pharmacist must educate the patient on the pain management. Patients managed with corticosteroids will need to be additionally monitored for drug adverse side effects. For most patients enrollment in a physical therapy program is the key to recovery. Some patients may benefit from supervised home physical therapy. Finally, patients have to be told that recovery will occur but it is gradual and may take several years.[7][8][9]


Frozen shoulder does recover in most people but the recovery may take 1-3 years. In most cases, physical therapy and arm exercise will gradually result in diminishing symptoms. So far, data do not show that diabetics have worse outcomes than non-diabetics. About 10% of patients will have residual shoulder stiffness and disability. After arthroscopic surgery, there is a gradual improvement in symptoms with slow recovery. However, postoperative physical therapy is a must after surgery to ensure recovery.[5][10]

What would Codman have thought about this?


THIS is a class of cases which I find it difficult to define, difficult to treat and difficult to explain from the point of view of pathology. Yet these cases form a fairly distinct clinical entity, which I formerly described under the adherent type (Type II) of subacromial bursitis. The experience of the last fifteen years has led me to doubt the advisability of calling this entity "bursitis," for I now believe it is essentially a tendinitis, with only secondary involvement of the bursa. The typical case is one of "frozen shoulder," which shows no calcified deposit, and in which a history of trauma is absent, vague or not clearly associated with the onset of symptoms. These cases are usually diagnosed as "periarthritis" or "neuritis."

FIGURE 48. FROZEN SHOULDER The term "Frozen Shoulder" covers the cases which are the subject of this chapter, but it also applies to many other conditions which cause spasm of the short rotators or adhesions about the joint or bursa. This figure shows the characteristic limitations of a severe case. The outline of the posterior edge of the scapula and the axis of the spine of the scapula and acromion are indicated. In the "frozen shoulder" the axis of the shaft of the humerus is nearly at right angles to the axis of the spine of the scapula, and remains so in whatever direction the arm is moved. Although determining the relation of the axis of the shaft of the humerus to that of the spine of the scapula is a primary one for the understanding of the diagnosis or treatment of many shoulder lesions, it seems best to dwell on its importance here. Cases without distinct traumatic history which show a frozen shoulder, and in which the X-ray gives no information, are likely to be cases of the condition described in this chapter. Unless the other joints of the body show pronounced evidence of arthritis, you may be quite sure that you are dealing with a case of tendinitis.

Perhaps it is best to attempt to describe the course of these cases clinically and to point out the ways in which they differ from arthritis on the one hand and cases of calcified deposits on the other. Perhaps it is only theoretically that they differ from arthritis, which we have already reduced to tendinitis in a previous chapter. The great clinical differences are that they are not a part of a generalized arthritis and that they run a self-limited course and clear up entirely without leaving the joint deformed or otherwise permanently damaged. They usually give the history of some slight trauma or overuse. Call some of them arthritis if you want to, but others will merge into the class of calcified deposits because the X-ray shows small flecks in the tendon. They might be called by whatever . name is best for that unsatisfactorily named group and given the same name without the "calcified" part, just as the term aleukemic leukemia is used. I have removed bits of tendon from these cases which under the microscope show the same necrotic changes which we find in the tendon substance adjacent to the deposits, or adjacent to the rents in the rupture cases. In all three there is the same change in the tendon substance as is depicted in Plate VI. It is a degeneration of the collagen without signs of inflammation as invasion of lymphocytes. One is tempted to say, "Oh, well, all three are arthritis in different phases!" Undoubtedly all three may be fundamentally tendinitis, but clinically there is great difference to the patient in point of what he may expect in the way of pain and in success of treatment, according to whether there is a calcified deposit (weeks), a tendinitis without calcification (months), or a complete rupture (years). Practically they are three different clinical entities, although they might be spoken of as the calcified, the uncalcified, and the ruptured forms of tendinitis. Last year I had four of these non-calcified cases to treat at about the same time, and it may be more instructive if I write of them together. They had in common: The condition had come on slowly; pain usually felt near the insertion of the deltoid; inability to sleep on the affected side; painful and incomplete elevation and external rotation; restriction of both spasmodic and mildly adherent type; atrophy of the spinati; little local tenderness; X-rays negative except for bone atrophy. The pain was very trying to every one of them, but they were all able to continue their regular habits and daily routine. One was a doctor of 62, an active man who takes good care of himself. Number two was a very vigorous, muscular, unmarried woman, who has been a champion golf player and still competes with much success at 46. Number three was a feeble old maid of 70. Number four was a married woman of about 50, apparently in perfect health. All of them were a little "run-down" without anything particular the matter. In each there was a possible history of some slight accident. The golf player, who was also an extremely busy lady in charitable affairs, when told that she would get well anyway in time, took no treatment and even played a little golf. The others submitted to the treatment I usually recommend to these cases; i.e., rest in bed with the arm in elevation. I do not always give an anaesthetic and break up the adhesions as I used to do many years ago, nor do I open the bursa and break them up with my finger as I did for a period. I simply put the patient to bed, apply a splint to the back of the forearm and tie each end of the splint with a loop over the railing at the head of the bed. The position of the patient's body usually acts as sufficient extension, although placing blocks under the legs at the head of the bed assures it. A light dose of morphine is given the first night to procure relaxation, but not enough to absolutely remove the pain for fear of letting the arm " go to sleep." In twelve to twenty-four hours the spasm relaxes, adhesions yield, the tuberosity passes under the acromion, and the arm becomes abducted and externally rotated and is more comfortable. In a- day or two I remove the splint and simply tie a bandage loosely around the wrist and to the head of the bed to remind the patient not to lower the arm. I get the patient up daily and begin the stooping exercises described on page 202. He stays in bed until he can freely move his arm about in any direction above his head. This is usually after one or two weeks. After discharge from the hospital he takes stooping exercises and for a few weeks sleeps at night with his arm in the hammock position. Of the three patients thus treated, the feeble, elderly lady made the most prompt and satisfactory recovery. The doctor also did well. The other lady taught me a lesson. She was wilful and charming, and persuaded me to let up on the traction too soon. She was a very busy woman and rebelled against the restraint, for she was planning to go abroad. The result of my weakness in not insisting on keeping up the traction longer was that when she got up the arm came into the sling position again; the spasm recurred, and her continued restriction marred her trip. However, when she returned in two months the shoulder was almost well again. This patient and another lady, whom I have since treated, exhibited a complication of this abduction treatment against which I wish to warn the reader. During the first few days they had intense abdominal discomfort which I now believe was due to acute dilatation of the stomach. A similar condition sometimes arises in cases where a plaster jacket is applied with the dorsal spine extended. The maintenance of the irritable shoulder joint in elevation has the effect of hyperextending the spine. I think this complication can be avoided by attention to the position of the patient's body and by permitting them to occasionally rise and walk about with the hand of the affected side held by the other hand over the head. In these two cases the suffering from the abdominal condition was worse than that from the shoulder. Another infrequent complication of the elevation treatment is difficulty in getting the arm to the side again after having kept it up for a week or two. In a few very sensitive patients this trouble has occurred, but in most cases if the patients get up each day and do the stooping exercises with my encouragement, they are able to bring the arm to the side from the first. Occasionally spasm persists to such an extent that even when the adhesions have been broken up under ether and the arm placed in elevation, the muscles will remain rigid and hold the arm in this erect position without yielding even when the patient stoops. This condition of affairs need give no anxiety, however. Strangely enough, it is not very painful. One lady in whom this occurred used to get up and walk about with the arm as erect as that of the traditional Hindu fakir. She would even read quite contentedly, or write letters with the other hand, resting the erect arm against the wall. This painless spasm persisted for three weeks, when I gave her some nitrous oxide and gently lowered the arm. No adhesions were felt to give, and there was no pain afterward. A few weeks later, she had complete use of the arm. Since this experience I encourage my patients to lower the arm at least once a day. Even with this precaution one of my recent cases, a very severe one, showed a similar degree of spasm and even at the end of six weeks could not completely attain the sling position, although she could easily place her hand behind her head. My effort is always toward keeping the ability to use the arm in elevation. If this end of the range of motion be maintained, the normal uses of the arm bring it soon enough into the sling position. Breaking up the adhesions even under ether is always an unpleasant process, and especially so in cases where the joint has been "frozen" for several months. Sometimes great force is required, and as the adhesions yield there is often a loud snapping noise which vibrates down the humerus and gives the sensation to the operator that the bone has been broken or the ligaments at the elbow torn. I have never broken the humerus in doing one of these manipulations, but I believe it might readily be done. In one case in which I had opened the bursa and ruptured the adhesions there with my finger, the joint still would not yield. With my finger still in the bursa I forced external rotation with my other hand, and distinctly felt some of the fibers of the subscapularis give way under my finger. I believe that unless great care is used such tearing in the fibers of the other rotators may thus occur. It would take but little force to bring this about when they are atrophied from months of nonuse, even if they are not actually partially necrotic. I For these reasons my present custom is to secure all the stretching I can by slow traction for forty-eight hours before I give an anaesthetic. In fact, the slow stretching usually accomplishes the purpose in this time unaided by manipulation, and, if an anaesthetic is necessary, little force is required. If much force must be used the bursa should be opened and the adhesions in it broken or cut before attempting to stretch the short rotators. A surgeon who attempts to break adhesions should know the normal motions of the shoulder described in Chapter II. The manipulations should be thorough and the joint moved to its normal extremes in all directions but not a bit beyond them. I prefer to begin with external rotation, but not to carry it to an extreme position at first. Some of the adhesions preventing internal rotation are then broken. Then abduction in the mid-position to a right angle with the body; then external rotation in abduction; then anteroposterior motions in the sagittal plane, and finally elevation to the pivotal position. After breaking some adhesions in each of these directions, perform the whole set over again, making sure that in each position the extreme is reached and not more than the extreme. During the thirty years in which I have tried to relieve these patients with frozen shoulders, I have tried and have seen tried many methods of manipulation and other forms of treatment. There have been few miracles performed and recovery is seldom rapid. A few cases already in a convalescent stage, with no pain but with restricted motion, have made prompt recoveries after snapping a few adhesions, but for most cases there have been many uncomfortable, restless nights and slow recoveries. However, recovery is always sure and may be confidently expected. None of the patients I have ever had with this condition have had recurrences in the same arm, but some have had the same trouble in the other arm a year or two later. In one case I operated upon both shoulders and removed bits of tendon for examination which showed the same pathologic changes which I have always found in these cases or in the calcified or ruptured cases. I do not think the operations did either much good or harm, but the postoperative rest in elevation gave the desired effect both times. Formerly before putting the arm in abduction I used to give ether and open the bursa. The appearances of the floors of the bursas were always the same—a congestion over the supraspinatus tendon on the base of the bursa like that of a "bloodshot eye. The congested area was usually circular with a whitish center much as in the calcified cases, but no macroscopic calcified material was present. Adhesions were often found. The congestion was in the synovia, not in the tendon beneath it. Although some cases made rapid recoveries, I gave up making an incision, not because it did any harm or caused any delay, but because I concluded that little was accomplished b}' it, and that the postoperative elevation was what really was doing the good by causing relaxation and therefore permitting the blood supply to accomplish its healing work more rapidly. In elevation also, the teres major, which has been constantly in spasm, is slowly stretched. Moreover, the old principle that adhesions will not form between the separated raw surfaces of the bursa also holds good. Perhaps, too, this position may approximate the synovial membrane at the edge of the cartilage. At any rate this treatment by rest in elevation is the best practical way to get these patients well with a minimum amount of pain so far as I know. This opinion was shared by the late Dr. Brickner of New York, who had great interest in shoulder cases. He used dumb-bell exercises afterward, while I prefer the stooping exercises as long as the joint is irritable. If a patient cannot rest in bed for a week or two I advise diathermy, although I am not convinced that it does much good. A number of my patients have preferred, as did the golfing lady spoken of above, to do nothing except the stooping exercises. As a rule they have recovered almost as quickly as the cases treated by other methods. Some cases are so mild that they do not even consult a doctor. There is great variation in the degree of severity of symptoms and in the period of disability. Even the most severe cases recover with or without treatment in about two years. Although I have devoted much time to analysis of the one hundred cases shown in Chart I, on p. 124,1 have not been able to work out any new facts which are worth recording, except those already mentioned on pages 135 to 140 in connection with the sex, the age, the occupation and the menopause. Even such analysis as I can make of the End Results is not fruitful. What knowledge I have acquired in forty years about this class of cases is not such that I can transmit it, but so far as my own practice is concerned I have diminished the relative number of cases of this kind, because I have subtracted from them the more pronounced calcified cases on the one hand and the more definite incomplete ruptures on the other. If the reader will refer again to page 124 he may understand that when we learn to define the borderlines more clearly, more of these puzzling tendinitis cases will be cured by the treatment suitable for calcified deposits and still more will be prevented by prompt and appropriate attention to cases of incomplete rupture. I would be more exact if I could in regard to the period these cases require to recover, but since recovery is always by degrees, it is pretty hard for even the patients to say just when they are well. They require my care about six weeks, but as their symptoms clear up gradually, they always cease coming to me before they fully recover from the last twinge. The whole course with or without treatment, from the first symptoms to the period when they forget to think of their shoulder each day, is more apt to be a year than six months. Nine months' pain and annoyance is a fair guess. One rarely sees a case until it has progressed for three months, because the onset is slow and insidious and the pain at first seems bearable. It is the restless nights rather than pain in the day which eventually drive them to a doctor. Since most of these cases follow some minor injury, it is "my hunch" that they are often due to irritation caused by the tearing away of a few fibers of the supraspinatus on the joint side, thus separating the synovial membrane from the edge of the cartilage. The irritation creates protective spasm, and efforts to use the arm or limber it up maintain the spasm. Thus this group of cases is not clearly separated from the group I have called incomplete ruptures of the supraspinatus. The main difference is in regard to the lack of distinct traumatic history. Although the type of cases spoken of under this heading of "Tendinitis" is pathologically still a little vague, it is a pretty definite clinical entity. It would require greater knowledge than is available at present to separate it pathologically from arthritis, bursitis, calcified deposits or from "rim rents" of the supraspinatus. I cannot too often warn the reader that complete rupture of the supraspinatus tendon is a very different clinical thing from these minor ruptures which set up a tendinitis or bursitis and recover after six months or so. In complete rupture cases the shoulders are not "frozen." As a clinical entity, tendinitis is diagnosed rather by exclusion than by special symptoms. All severe cases present a "frozen shoulder," and thereby complete rupture of the supraspinatus is excluded. Careful X-rays rule out a calcified deposit. The lack of involvement of other joints excludes arthritis ; so does complete recovery within a few months under appropriate treatment. These cases can only be separated from incomplete ruptures by the absence of a distinct traumatic history, and this is practically a difficult line to draw. The discovery of abscessed teeth, suppurating tonsils or other possible portals for toxic absorption would tip the balance of judgment away from trauma as a cause. Even then we must not forget that a tendon made brittle by toxins may rupture. The reader should not judge the relative frequency and importance of this group of cases by the length of this chapter compared to that of other chapters. He may, however, so judge any knowledge of the subject which I have to communicate to him. These are common cases, but it does not take a long chapter for me to tell all I know about them. I could write a whole book on my experiences with cases of this type, but it would have to deal more with human nature than with demonstrable pathology or remarkable success in treatment. The art rather than the science of medicine is called for in these cases, but in the cases of calcified deposit and in the complete rupture cases, cold, definite, prompt efficiency is more desirable. Since the symptoms in these cases vary greatly in degree, prognosis is difficult at best, and since patients vary greatly in their ability to bear the same conditions, a great variety of puzzles are offered for solution. Some patients have little restriction and much pain; others have stiff, painless joints, and there are all degrees between. I am confident that I can shorten the convalescence of any case by the elevation treatment outlined above, but in most cases I do not advise it, and the mere assurance that they will recover in time seems to be of wonderful therapeutic value. If pain is severe enough to keep them awake much at night or to prevent them from earning their living, or from following their favorite sports, I advise the above treatment. Massage is a help. Light stooping exercises seem to me to be of great benefit. Attention to the general well-being of the patient is most important. For instance, a good vacation under pleasant, healthful circumstances seems to me more desirable than any form of therapy if the pain is bearable. If the arm is not well enough to make a vacation enjoyable, it can soon be made so by the elevation treatment. Since this chapter was written a thoughtful article has appeared. (Jour. A. M. A., Dec. 31, 1932, Vol. 99, No. 27, pages 2252-2257. "Periarthritis of the Shoulder, An Analysis of Two Hundred Cases," by James A. Dickson, M.D., and Edward H. Crosby, M.D., Cleveland.) While there are many minor points on which I could not agree, I feel that this article is the most reasonable, recent attempt to cope with the problem which I have tried to present in this chapter. It is an earnest, painstaking study of an unsolved question. The especial point on which I cannot agree with the authors is that they state from a clinical point of view that there is little difference between the cases which show a calcified deposit and those which do not. My experience does not confirm this. I regard the calcified cases as very simple and easy to relieve as compared to the adherent, non-calcified cases. The results of operations on the calcified cases have, from the first, been as satisfactory as any surgical operations which I know of. Only in a few long-standing cases has the return of function been a matter of months. On the other hand, in the true non-calcified cases the convalescence is practically always a matter of months and sometimes of years. Another fair criticism of this article is that the authors seem to have assumed that because they find infected foci in many patients, the condition of the shoulder is due to these foci. It is my belief that as large a percentage (38.5%) of infected teeth would be found among any two hundred persons of the same average age and class in life. The relation to the menopause is much more striking than that to the teeth. See p. 138, Chart III. Their last paragraph, which I quote verbatim, shows an admirable degree of intellectual honesty.

"The study of the cases reported in the literature and of our series of 200 cases has shown us most strikingly that an exact evaluation of the importance of any one of the factors which may be involved in the causation of the disease, or of the relative value of the various therapeutic measures, is practically impossible. For, in our series, it has been found that in all cases, whether the evidence pointed to infection, metabolic disturbances or trauma as the important etiologic factor; whether the treatment stressed was eradication of foci, physical measures, manipulation or operation, that the time required for recovery and the total duration of the disease was remarkably constant throughout. This suggests that in these cases there must be some general physiologic disturbance as a common denominator, which cannot be explained or accounted for at present. We reiterate this statement because this fact seems to have been lost sight of by many workers in their enthusiasm for one particular type of treatment, or in their zealous endeavor to attribute the symptoms to some particular etiologic factor"

I would be inclined to swallow this paragraph whole if it were salted with just a little optimism.


Turn to those following Chapter VI, and note especially the remarks on Duplay's contributions. Introduction: Frozen shoulder is an extremely disabling condition, presenting with and remitting shoulder pain and stiffness. This was well defined by Codman in 1934, who described the first and best classical diagnostic criteria still used to this day.

These were:

1. Global restriction of shoulder movement. 2. Idiopathic etiology. 3. Usually painful at the outset. 4. Normal x-ray. 5. Limitation of external rotation and elevation.

The classification of a primary idiopathic frozen shoulder may be primary or idiopathic. This is a distinct pathological condition identified by global limitation of glenohumeral motion, with a loss of compliance of the shoulder capsule, with no specific underlying cause found.

A secondary stiff shoulder or secondary frozen shoulder, typically presents after injury or surgery. It may also follow an accompanying condition, such as subacromial impingement or a rotator cuff tear.

Diagnosis: The diagnosis of a primary Idiopathic frozen shoulder is made on the basis of:

Age: Typically occurring in females more common than males, in the 4th and 5th decade. Pain: The pain is of a constant nature, severe, affecting sleep. There is often a toothache pain at rest, with sharp pains with forceful movements. Loss of external rotation: The typical loss of external rotation is such that passive external rotation is <0 degrees from the sagittal plane. A secondary frozen shoulder usually has restriction of external rotation, which is beyond 0 degrees (i.e. external rotation of 10 degrees as opposed to -10 degrees with a primary frozen shoulder). Natural History: The natural history of a frozen shoulder has typically been described as passing through 3 stages. These stages last for approximately 2 years.

Stage: 1. Freezing phrase: This is associated with pain and loss for about 3 months. 2. Frozen phase: This lasts for approximately 3-9 months, with pain at extreme range of movement and marked stiffness. 3. Thawing phase: This last for approximately 9-18 months, usually painless and the stiffness starts to gradual resolve at this stage.

Aetiology: The frozen shoulder has been found to be more common in association with the following conditions: Diabetes (10-20% association). There is a 2-4 times increased risk for diabetics of developing frozen shoulder. Insulin-dependent diabetics have a 36% chance of developing it, 10% bilaterally and the condition is more severe in diabetics. Cardiac/lipid problems.


Endocrine abnormalities, particularly hypothyroidism. Trauma. Drugs - MMPI. Pathology: The macroscopic appearance is that of thickening of the anterior capsule, particularly the coracohumeral ligament (CHL) and middle glenohumeral ligament MGHL). Contracture of the capsular ligament restricts specific movements of the glenohumeral joint. These are: CHL = external rotation in neutral MGHL = external rotation in mid-elevation AIGHL = external rotation in abduction Inferior capsule (ICS) = abduction in neutral rotation PIC = internal rotation PSC = internal rotation in abduction

Villonodular synovitis within the rotator interval with thickening and contracture of the coracohumeral ligament is the primary pathology. The capsular thickening and contracture leads to a reduced glenohumeral joint volume.

Arthroscopic images showing a normal shoulder on the left, and a frozen shoulder on the right (note the villonodular synovitis):

Arthrogram of a normal shoulder (left) and a frozen shoulder (right) - note the reduced capsular volume:

Microscopically there are 4 stages:
Inflammatory synovitis with capsule unaffected.
Proliferative synovitis, which is hypertrophic.
Maturation of the capsule, with reduced vascularity.
Burnt out synovium with a dense scar appearance.

Tim Bunker in 1995 has shown the pathology to be similar to Dupuytren's disease with increased collagen, myofibroblast and fibroplasia. This in association with Trisomy 7 & 8, 58% of patients who had a primary idiopathic frozen shoulder had Dupuytren's disease elsewhere. The inflammatory drivers for Dupuytren's disease and frozen shoulder are similar, these being TGF-beta and PDGF.

Treatments: The natural history of frozen shoulder is not necessarily that of complete resolution. The original quotation that most recover from frozen shoulder originates from Codman in 1934. However, he stated that most frozen shoulders recover compared to tuberculosis. Many studies have shown frozen shoulder not to be an entirely self limiting condition and most patients still have some restriction of shoulder movement on resolution of the frozen shoulder but no functional disability (Grey JBJSA 1962, Shaffer JBJSA 1992, Bunker JBJSB 1995, Miller Orthopaedics 1996).

The treatment options range from: 1. Nothing. 2. Physiotherapy. 3. Distention injections. 4. Locally acting steroid injections. 5. Manipulation under anaesthetics. 6. Open/Arthroscopic capsular release.

Non-Operative Treatment: Studies on non-operative treatments for frozen shoulder have shown that physiotherapy improves range of movement but not necessarily pain relief. Steroid injections have a benefit for short-term pain relief only but no long-term pain relief (Ryan's Rheumatology 2005, Haslan and Celiker Rheumatol int 2001, Bulgnann Ragum Dis 1984).

Surgery: Early surgery has been shown to be of significant benefit for a faster recovery of pain, quicker recovery of function and earlier return to work (Hill Orthopaedics 1988, Doddonhoff JSES 2000). Arthroscopic capsular release has been shown to avoid the complications of manipulation under anaesthetic. It has the advantage of being able to identify any other associated pathologies. Numerous studies have demonstrated the benefit of capsular release, although no study that we are aware of randomises patients directly comparing MUA to arthroscopic release. Our own study of 48 patients undergoing capsular release followed for 18 months, demonstrated significant improvement by 5 months postoperatively for resistant frozen shoulders. The Constant score improved from 14 preoperatively to 66 postoperatively, with 80% satisfaction. Range of movement improved significantly by 5 months (Boutros, Snow and Funk 2006).

Treatment Algorhythm. Treatment algorhythm is based on the ability of patients to cope and the significance to cope with the pain and stiffness during the stages of frozen shoulder. This is summarised below:

Conclusion: Primary idiopathic frozen shoulder is an extremely disabling condition, which does pass through a typical 3 stage progression. However, full recovery at the conclusion of the 3 phases is not common. Early surgical intervention has significant benefits for those patients that are unable to cope with non-operative treatment.


  1. Mezian K, Coffey R, Chang KV. Frozen Shoulder. In: StatPearls [Internet]. Treasure Island (FL): StatPearls Publishing; 2021 Jan-.[Updated 2020 Sep 3].
  2. 2.0 2.1 2.2 Hubbard MJ, Hildebrand BA, Battafarano MM, Battafarano DF. Common Soft Tissue Musculoskeletal Pain Disorders. Prim. Care. 2018;45:289-303
  3. 3.0 3.1 3.2 3.3 Xiao RC, DeAngelis JP, Smith CC, Ramappa AJ. Evaluating Nonoperative Treatments for Adhesive Capsulitis. J Surg Orthop Adv. 2017,26:193-9
  4. Wu F, Kachooei AR, Ebrahimzadeh MH, Bagheri F, Hakimi E, Shojaie B, Nazarian A. Bilateral Arm-Abduction Shoulder Radiography to Determine the Involvement of the Scapulothoracic Motion in Frozen Shoulder. Arch Bone Jt Surg. 2018;6:225-32
  5. 5.0 5.1 Oderuth E, Ali M, Atchia I, Malviya A. A double blind randomised control trial investigating the efficacy of platelet rich plasma versus placebo for the treatment of greater trochanteric pain syndrome (the HIPPO trial): a protocol for a randomised clinical trial. Trials. 2018;19:517
  6. Wong CK, Strang BL, Schram GA, Mercer EA, Kesting RS, Deo KS. A pragmatic regional interdependence approach to primary frozen shoulder: a retrospective case series. J Man Manip Ther. 2018;26:109-18
  7. Lee SH, Yoon C, Chung SG, Kim HC, Kwak Y, Park HW, Kim K. Measurement of Shoulder Range of Motion in Patients with Adhesive Capsulitis Using a Kinect. PLoS ONE. 2015;10:e0129398
  8. Yamada E, Thomas DC. Common musculoskeletal diagnoses of upper and lower extremities in older patients. Mt. Sinai J. Med. 2011;78:546-57
  9. Robb G, Arroll B, Reid D, Goodyear-Smith F. Summary of an evidence-based guideline on soft tissue shoulder injuries and related disorders--Part 2: Management. J Prim Health Care. 2009;1:42-9
  10. Lamplot JD, Lillegraven O, Brophy RH. Outcomes From Conservative Treatment of Shoulder Idiopathic Adhesive Capsulitis and Factors Associated With Developing Contralateral Disease. Orthop J Sports Med. 2018;6:2325967118785169