Frozen shoulder is a condition which doesn’t always have an obvious cause. It is sometimes associated with diabetes, thyroid disease and other rarer medical conditions and it may occur after a minor to moderate injury or fracture to the shoulder. It is characterised by increasing pain and stiffness in the shoulder with gradual onset of significant disability limiting normal activities of daily living. The prevalence is 2-5% of the general population, and most commonly occurs in the 5th to 6th decades and is more common in women than men. It also appears to luckily occur more often in the non-dominant than the dominant shoulder but between 6-15% of patients develop a frozen shoulder on the other side within 5 years.
One of the key features of the condition is inflammation between the tendons and ligaments in the shoulder which abnormally tethers the glenohumeral joint. This is the ball and socket joint. Instead of the normal complex shoulder movement, with prevention of elevation of the humeral head which is kept pushed down by the action of the short upper rotator cuff muscles with elevation of the arm, the abnormal tethering prevents this gliding movement and thus converts the shoulder more into a hinge joint so that there is significant impingement, or compression, of the top of the head of the humerus and the intervening tendons and bursa underneath the bony and ligamentous arch on top of the shoulder. This abnormal tethering not only limits movement with painful restriction at end range but may also possibly cause or contribute to long term damage to the upper rotator cuff muscles and tendons and certainly cause inflammation of the subacromial bursa which separates the top muscles of the rotator cuff from the bony arch above.
The natural history of this condition is slow, complete to almost complete resolution over a period as long as 2-3 years but sometimes resolution is faster. 40% have mostly mild disability and loss of range greater than 3 years, 15% have long term disability if left untreated.
Conservative management involves relieving pain by avoiding movements in the painful range, using a pillow or rolled towel around the arm to prevent the arm from being pushed into the side during sleep and the use of anti-inflammatory medication such as Ibuprofen and analgesics such as Panadol or Panadeine or Panadeine Forte as needed, particularly if there is disturbance of sleep at night. Physiotherapy supervised simple strengthening exercises within the pain free range may be helpful in retaining better function and should be developed into a simple home exercise programme. However, once the condition is established and there is significant stiffness in the shoulder and night pain, further physiotherapy, electrotherapy, acupuncture and similar treatments are rarely of any benefit except possible additional pain relief.
There is some anecdotal evidence that corticosteroid injections into the glenohumeral joint itself (which is very difficult without ultrasound guidance) if done early enough may limit the progress of the condition and shorten the disability but this hasn’t been proven in randomised controlled studies. Corticosteroid injections of the subacromial bursa once the condition is established doesn’t have any beneficial effect on the progress of the disease. Physiotherapy trying to improve range of motion once the stiffness has occurred is not helpful and is a waste of resources, but physiotherapy is very useful after the manipulation under anaesthetic to break down the adhesions to restore normal balance of strength in the rotator cuff muscles and also normal co-ordination of movement between the glenohumeral or shoulder joint and the scapulothoracic joint, or shoulder blade articulation with the back of the chest wall (restoring normal scapulohumeral rhythm).
Manipulation Under Anaesthetic and Alternative Treatments:
In many cases limitation of range may slowly improve allowing return to most activities with less pain and improving function. Some sportspeople and those with more physically demanding jobs may find the continued shoulder stiffness very limiting and may seek more active intervention but this should be timed correctly and be evidence-based and safe.
Blowing up or distending of the shoulder joint cavity with normal saline or other fluid under ultrasound (Hydrodilatation) is useful in some cases and it may help break down the adhesions. This tends to be quite uncomfortable and in my experience is not as effective as a gentle manipulation under anaesthetic.
Manipulation under anaesthetic is an effective and safe way of treating adhesive capsulitis, markedly reducing the disability time once the acute inflammatory painful phase of the condition has resolved. It is generally safe to do after the stiffness has been present for 4 months, although some clinicians suggest this should be delayed until stiffness has been proven to not adequately resolve over 12 months. The timing of this intervention should be planned according to the progress of the condition and the needs of the patient and be subject to repeat assessment and discussion between the patient and the treating doctor.
If you have this condition and decide on this form of treatment you will be admitted to hospital for a short stay and taken into the operating theatre where you will be put under a light anaesthetic which relaxes the muscles and completely anaesthetises you so you won’t feel pain during the procedure. The shoulder can very gently be moved through a full range of motion equal to the normal range in the other shoulder, breaking down the adhesions. Because this is only moving the shoulder through a normal range of motion it should be only significantly stressing the abnormal adhesions rather than the normal structures in the shoulder. In almost every case the amount of force required to break down the adhesions is much, much less than the force required to cause any damage to normal structures in the shoulder unless there is already damage in the shoulder such as rotator cuff tears or fractures which obviously need to be excluded and noted prior to consideration of manipulation under anaesthetic.
The use of anaesthetic allows this to be done without pain at the time and allows the procedure to be done gently with full control and without the patient resisting, which means that much less force is required. This makes the manipulation much safer.
After the manipulation the shoulder is carefully re-examined for stability and smooth movement of all of the joint surfaces. Normally full range of motion can be achieved, although sometimes there is a very slight persisting limitation of internal rotation reach.
Corticosteroid and long acting local anaesthetic is then injected into the shoulder joint and the subacromial bursa whilst you are still asleep. This should give good pain relief in addition to oral or injected pain killers (analgesics) for the first 10-12 hours. After this, normal analgesics can be used as necessary.
It is important to start supervised movement immediately after you wake up from the anaesthetic and you should book to start physiotherapy the following day. Physiotherapy is initially daily, then second daily and then once a week and normally physiotherapy is required for a period of 4 weeks, and sometimes more depending on progress. This will be supplemented by a series of home exercises which you should carry out regularly. You will be reviewed by myself and the physiotherapist during this period and treatment will be modified as necessary.
For the 4 days prior to the procedure you should shower using a medicated soap and lather up all around the neck, shoulder and underarm region, and leave the lather on as long as possible in the shower before washing off. On the morning of the procedure you should have a light breakfast which should be finished before 7am and normally you would be admitted to hospital around 12 midday and the procedure would be done in hospital that afternoon. You should be comfortable enough to be discharged from hospital early the same evening. In some cases, if you need more analgesia after the procedure, you may need to stay in overnight. It is very important that you have nothing to eat or drink after 7am on the morning of the operation unless advised otherwise by the hospital prior to admission.
Some patients experience side effects related to the general anaesthetic, particularly with nausea and vomiting. Sometimes light headedness, drowsiness and impairment of judgement can significant affect your ability to drive a motor vehicle or operate machinery. For this reason you should not carry out any dangerous activities such as this for at least 48 hours. Post-operatively, and on some occasions, even longer. Driving a motor vehicle will also be limited by shoulder pain and when you can return to this and other similar activities will depend on your post-manipulation progress. Your treating physiotherapist can help you decide when it is appropriate for you to get back to your various activities. Very occasionally patients will react to the injection of the corticosteroid with a slight flare up of their symptoms after the local anaesthetic wears off. This is normally, if it does occur, short lasting and normally only 24-48 hours. If pain increases or remains the same beyond this time you should contact me immediately or discuss this with the physiotherapist. Other complications such as infection following the corticosteroid injections are extremely unlikely with an incidence of less than 1 in 10 000. These risks are minimised by pre-operative skin preparation and protecting the skin over the shoulder area prior the procedure from cuts, scratches or grazes.
The risk of injury to other structures in the shoulder with a breaking down of adhesions is also extremely unlikely. In some cases however, if the shoulder was unstable before the adhesions formed, the shoulder will again be unstable following the manipulation until dynamic stability can be achieved by building up the muscles under the physiotherapy programme. In some such cases as pre-existing instability, if the physiotherapy fails to return dynamic stability of the shoulder, surgical shoulder reconstruction may be required. There have been instances reported in the medical research literature of fracture occurring during manipulation under anaesthetic but this is extremely unlikely and certainly hasn’t occurred in over 25 years of my personal clinical experience. Pre-procedural investigations should generally be able to identify individuals more at risk of complications.
Dr Stuart Watson, Sports and Exercise physician