FROZEN SHOULDER - November 2000
LYN WATSON Manipulative Physiotherapist
The term “frozen shoulder” was first introduced by Codman (1) in 1934. It is also commonly referred to as “adhesive capsulitis”. In general terms the condition means that a patient has a very stiff shoulder that may or may not be associated with significant shoulder joint pain.
What is a frozen shoulder?
A frozen shoulder is a shoulder that has become so stiff that the patient usually cannot lift their arm up past shoulder height, cannot do up their bra or tuck in their shirt and has difficulty reaching across under the other arm (i.e. washing the opposite arm pit). The exact degree of shoulder stiffness present may be variable but in order for a shoulder to be classified as truly frozen there must be significant movement loss (i.e. less than 100 degrees of abduction or sideways movement).
Frequently it is associated with significant pain, especially at night time. This pain is usually focussed over the front of the shoulder (biceps region) or down the outside of the arm. However there may be just a general shoulder ache. The amount of pain experienced by patients with a frozen shoulder may be variable with some patients developing severe pain whilst others have only mild pain. The severity of the pain is usually determined by the extent of vascular proliferation or inflammation reaction that develops inside the joint as part of the disease process.
Most patients with frozen shoulder report that their pain levels fluctuate over time. In the beginning phase the pain is usually the most acute, especially at nighttime and this may cause significant sleep disturbance. At eight months post onset most patients will report that their pain has lessened significantly. However this figure is quite arbitrary as many patients may report a decrease in pain much sooner than this. Whilst others still report significant pain at over two years post onset.
Who gets a frozen shoulder?
Traditionally it has been cited that women get frozen shoulders. This is not entirely true. Men get frozen shoulders too. Whilst there may be a higher tendency for women to get it than men, it should not be pigeon holed as a purely female condition. In a recent study by Watson & Dalziel (2), 57% of the patient population was female and 43% was male.
Frozen shoulder does tend to affect predominantly older patients. In the study by Watson & Dalziel the mean age of patient was 52 years (range 37 to 70 years). Some authors have postulated that this may be because frozen shoulder is an inflammatory response to ageing changes in the shoulder joint and or tendons of the shoulder. However there is no definite proof of this. Certainly younger people do not tend to get frozen shoulders unless it is precipitated by some sort of major trauma to the joint (such as a fracture).
There is not proof that any particular occupational tasks pre-dispose or cause a frozen shoulder. On questioning, some patients report that their symptoms started after doing a particular task whilst at the workplace, but many more patients report that their symptoms commenced for no particular reason at all. Unlike some other shoulder conditions, there is no evidence that frozen shoulder is related specifically to either manual or repetitive work.
Traditionally it has been thought that frozen shoulder affects the non-dominant arm. However studies have been variable in their findings and overall there appears to be no dominance affect. Hence a patient is just as likely to get it in their dominant or non-dominant arm.
What causes a frozen shoulder?
The exact cause of a frozen shoulder continues to prove to be elusive. It is known that frozen shoulder is a definite disease process which occurs in three major stages (3, 4). The first stage is predominantly and inflammatory stage where increased blood vessels proliferate throughout the shoulder joint lining and capsule (refer figure 1). This stage is usually associated with a lot of pain, especially at night and some range of motion loss. Stage 2 has a lessening of the vascular inflammatory synovitis (or blood vessel formation) and more adhesion or scar tissue formation. The shoulder usually becomes less painful during this stage, only hurting at the extreme of motion or during activities. However the shoulder joint itself gets stiffer and motion becomes less. Stage 3 tends to have no inflammation left and hence very little pain. However mature scar tissue has formed truly stiffening up the joint and preventing motion.
It is not known what causes this process to start in the first place. Some patients report they experienced a mild strain to their shoulder, such as reaching into the backseat of the car, prior to the frozen shoulder commencing. Others report that it commenced after a period of increased shoulder loading, such as cleaning windows. Many patients can report no precipitating event whatsoever.
There is no evidence that there is any viral or infective cause for a frozen shoulder and currently it is classified as “idiopathic” or unknown. It has been reported in many articles that there is an increased tendency for people with diabetes to get a frozen shoulder but this does not mean that the diabetes itself causes the frozen shoulder or indeed that if you get a frozen shoulder you are going to get diabetes. It is a finding of unknown significance.
Several authors have postulated that ageing changes in the biceps tendon as the cause of the frozen shoulder but as yet there is no direct proof. Certainly much of the pain of a frozen shoulder is usually centred over the biceps tendon and the age of the population most commonly affected would fall within the parameters for ageing change. Younger patients do not tend to get frozen shoulders unless precipitated by trauma and this is not really classified as a frozen shoulder but as a stiffening response to a traumatic injury.
How long does a frozen shoulder last for?
Historically, patients have been told that frozen shoulder is a self-limiting disorder that will resolve itself in two years. Most recent articles agree that the natural history of the condition is not so predictable with several showing that the symptoms may persist for three or more years and some people may have a permanent loss of motion.
Clinical experience shows that each individual is different. Some people have a frozen shoulder that is fully resolved in 12 months, whilst other present at two years with significant pain and loss of function in their shoulder. The problem is that it is impossible to predict how long any individual’s symptoms are likely to persist for.
What treatment is available for a frozen shoulder?
Many treatments are utilised to try to combat a frozen shoulder, but unfortunately most of them have not been proven to be effective. Whilst many patients may have anecdotal evidence of things that have worked for them, very little has been shown to be any better that placebo at clinical trials. Whether or not a patient tends to seek treatment usually depends on how much pain they have (especially at night), how long it has been going for and whether or not their function is significantly impaired.
Physiotherapy, acupuncture, massage and chiropractic treatments do not tend to help in the initial inflammatory phases and in some instances aggressive treatment can make the pain worse. Anti-inflammatory medication and creams also usually do not assist in the initial phases of the disease process. Once the painful phase of the disease has settled then physiotherapy and massage techniques that facilitate a return of movement and strength in the shoulder joint appear to be much more effective, but as stated previously it is variable how long the initial inflammatory phase can last.
Corticosteroid injections or medication if given in the right place at the right dose can be effective in decreasing the pain of capsulitis. This needs to be done by a medical practitioner or specialist such as an orthopaedic surgeon, sports physician or rheumatologist, who is experienced in managing this condition. However unfortunately the literature has shown that injection alone does not always permanently fix the pain and often does not help regain the range of motion (5).
One form of injection therapy is a hydrodilatation which is an injection of saline (water) plus corticosteroids. Some patients have achieved terrific success with this technique whilst in others it has not been successful at all. Further research needs to be done on this technique to prove that it has benefits over and above other forms of injection therapy. Clinical evidence also suggests that it needs to be combined with some form of physiotherapy involving exercise and massage to fully regain good shoulder function.
Surgery is another option that patients may consider. Whilst no one wants an operation, new techniques are available that make it a much easier and reliable solution than it has been in the past. Arthroscopic techniques are available which means the surgery is done through keyhole incisions under the assistance of a microscope. The operation usually involves very specific removal of the inflammatory and scar tissue from the joint capsule (selective capsulotomy) +/- some corticosteroid medication applied into the joint at the time of surgery. Surgery usually only involves one night in hospital and most patients pain and range of motion is much better within six weeks of the surgery (2). Although it can take longer than this for patients to fully regain all of their motion and strength and some patients do require an injection in the post operative process. Research has shown that the best results are gained when the surgery is combined with a very specific physiotherapy programme in patients who are prepared to do a home exercise regime (2).
Will it re-occur?
Whilst no definite studies have been published on this subject, clinical experience has shown that frozen shoulders do not re-occur. Once a frozen shoulder episode has finished then it usually doesn’t come back. However if left untreated there is no telling just how long the episode will last and there may be some permanent restriction in joint motion and function. Hence it is best to try some of the treatment avenues outlined above.
However, at least 30% of people do experience a similar type of problem in the other side. Patients often report that one side is worse than the other but there is no guarantee of this. There is absolutely no evidence that any preventative measure can assist in minimising the chances of this occurring. The good news is that it doesn’t appear to affect other joints, only the shoulder.
In summary, it appears that frozen shoulder is a vascular based inflammatory pathology with adhesion formation in the shoulder. It affects both men and women in their middle ages and can occur in either or both shoulders for no apparent reason or with very little precipitating cause. It results in significant pain, especially at night in the early phases and causes a restriction in joint range of motion that interferes with many activities of daily living as well as hampering many occupational and sporting tasks. Its duration can be variable but without treatment may last for up to three years or longer. No one treatment regime has been proven to be absolutely the best solution, but corticosteroid application into the shoulder joint (usually via a hydrodilatation) or arthroscopic surgery appear to have the best results, if followed up with specific physiotherapy and home exercise programme.