Skip navigation

Physiotherapy for soft tissue shoulder disorders



Clinical bottom line: Evidence suggests that ultrasound is not useful in alleviating the symptoms associated with soft tissue shoulder disorder. There is insufficient reliable information to assess the efficacy of other physiotherapy interventions, including different electrical therapies, cold therapy, thermotherapy and different exercise and manipulation therapies. However, based on information from poor quality trials, there is currently no evidence to support these interventions. More information from high quality trials is needed.

Pain is the primary symptom associated with shoulder disorders affecting the soft tissue. Cumulative annual incidence of shoulder disorders varies from seven to 25 per 1000 general practitioner consultations. Half of the episodes will resolve within six months, but many will persist beyond a year. Treatment includes advice, analgesics, non-steroidal anti-inflammatory drugs, steroid injection and physiotherapy.

Systematic review:

Van der Heijden GJMG, van der Windt DAWM, de Winter AF. Physiotherapy for patients with soft tissue shoulder disorders: a systematic review of randomised clinical trials. BMJ. 1997; 315: 25-30.

Date review completed: December 1995

Number of trials included: 20

Number of patients:

Control groups: active, placebo or no treatment

Main outcomes: success rate, pain, mobility or functional status.

Inclusion criteria were randomised, controlled trials of physiotherapy interventions for soft tissue shoulder disorders; physiotherapy interventions include ultrasound, transcutaneous electrical nerve stimulation (TENS), electrotherapy, pulsed electromagnetic fields, magnetic treatment, cold therapy, thermotherapy, exercises, manipulation and mobilisation; outcomes included success rate, pain, mobility or functional status; full journal publication.

Reviewers assessed trials for validity and extracted information on outcomes of interest on an intention-to-treat basis. A descriptive summary of trials was provided.


Findings:

Included trials were of mixed quality, with only half satisfying at least 50% of the validity checks. Included patients ranged from those who met specific narrow diagnostic criteria, to general soft tissue shoulder disorders. Most trials were too small to test reliably for differences between groups. Reviewers did not provide information on the blinding status of trials, making it difficult to judge whether reports of significant benefit were likely to be biased.

Ultrasound

Six trials examined ultrasound. Most of these trials have been considered in a separate review of ultrasound for musculoskeletal disorders, which concluded that the evidence suggests that ultrasound is not effective compared with placebo and active treatments.

Electrical therapies

Based on two small trials, transcutaneous electrical nerve stimulation (TENS) was no more effective than ultrasound at six weeks, or than constant voltage electrotherapy at three weeks. Of two trials looking at pulsed electromagnetic field therapy, both reported benefit compared with placebo. However, in one case reviewers were able to extract intention-to-treat data, which showed no benefit. One small trial looked at magnetic treatment plus heat and exercise, and reported no benefit compared with heat and exercise alone.

Cold therapy

One trial showed no benefit of ice packs compared with ultrasound or steroid injection. One small trial showed no benefit of ice packs plus medication and pendular exercises compared with mobilisation, facilitation exercises, steroid injection, all given with medication and pendular exercises, and no benefit compared with medication and pendular exercises alone.

Thermotherapy

Two small trials compared different types of thermotherapy with steroid injection, analgesics and/or muscle relaxants, and found no benefit.

Exercise therapy

One trial showed exercise therapy was as effective as surgery in patients with a stage II impingement syndrome at six months, and was more effective than placebo laser therapy. Three small trials showed no benefit of exercises alongside a range of other treatments compared with the treatments given alone.


Adverse effects

Reviewers did not report on adverse effects.

 

Comment

The main limitation of these trials is that very few of them were designed to assess the difference between active and control or between two active treatments. We are therefore unable to say whether negative findings reflect a lack of difference, or an inability to measure a difference. Whilst blinding of some interventions is difficult, reviewers could help by clarifying which information comes from blinded trials.

Further reading

Related topics


Ultrasound for musculoskeletal pain

Identifier

Identifier CP106 - 6371: Oct-2000