Chris Bleakley, Suzanne McDonough and Domhnall MacAuley

The American Journal of Sports Medicine

2004, Volume 32, Pages 251-261

Bleakley et al have undertaken an interesting review of the scientific and medical literature to assess the clinical evidence on ice therapy for acute soft tissue injuries. As with most physical modalities, there is no definitive evidence which gives optimal parameters on the use of ice therapy. What the authors of this review did was measure the quality of each study's methodology and considered the clinical appropriateness of applied treatments within the published research on ice therapy.

Ice therapy or ‘cryotherapy' is one of the simplest and oldest therapeutic modalities in the treatment of soft-tissue injuries sustained during sports activities. Physiotherapists who use ice therapy aim to reduce inflammation by decreasing tissue temperature, which can diminish pain, metabolism, and muscle spasm. By reducing inflammation it is thought that the recovery from soft tissue trauma can be accelerated.

Generally, the selection of treatment parameters in a clinical environment continues to be made pragmatically (depending on the location of the injury), and recommendations in review articles range from 10 to 20 minutes, 2 to 4 times per day, up to 20 to 30 minutes, or 30 to 45 minutes every 2 hours. This makes it difficult for the authors to make comparisons between studies. In addition, ice therapy is commonly combined with compression and elevation, making it difficult to determine the value of ice therapy alone

Bleakley et al reviewed 22 relevant studies using a computer-based literature search on a total of eight databases. The authors reviewed randomized-controlled trials of human subjects where the subjects were recovering from acute soft-tissue injuries or orthopaedic surgical interventions. All studies had to include at least one of the following outcome measures:

  • function (subjective or objective),
  • pain,
  • swelling,
  • or range of movement (ROM).

The main findings of Bleakley et al's review were as follows.

Ice alone seems to be more effective than applying no form of ice therapy after minor knee surgery. A single study supported the application of ice immediately before a rehabilitation programme as it significantly decreased the subjects' reported pain scores. The study also reported that subjects applying ice therapy used significantly less pain killing medication and had a significantly better weight-bearing status.

Research comparing the effect of continuous ice therapy to intermittent 20-minute ice applications over the first 3 postoperative days found that subjects applying continuous ice therapy had a significantly greater decrease in pain in comparison to those using ice therapy intermittently. However this was the only study to compare the effectiveness of two different ice therapy protocols, and although it appears that continuous ice therapy should be the treatment of choice after surgery, the modes of ice application were not consistent across the two groups.

Bleakley et al found that only one clinical study compared ice and compression to ice alone. The combination of treatments appeared to be significantly more effective than ice alone in terms of reducing the amount of analgesia administered post–ACL reconstruction. These results must be interpreted with caution, however, as the mode and duration of ice treatment was not controlled across groups.

The authors reviewed the effects of ice and compression with compression alone by examining studies that attempted to seperate the effects of ice from compression by comparing a variety of treatment combinations. Eight published studies strictly controlled for the type of compressive bandages used across comparison groups and Bleakley et al concluded that there appears to be little difference in the effectiveness of ice and compression and compression alone.

One study found no significant difference in the time of restricted activity after ankle sprain in subjects treated with compression alone and simultaneous ice and compression. Another study found no significant differences in subjective pain scores and the amounts of intravenous and oral analgesics in a group of subjects who had had knee replacements.

Only two studies reported significant differences between subjects treated with ice and compression and compression alone. One of these studies found no differences between groups in knee ROM after ACL reconstruction, although there was a significantly decreased analgesic consumption reported in favour of the ice and compression group, compared to the compression alone group. Another study examined subjects post–ACL reconstruction, where two groups were treated with simultaneous ice and compression and a third with compression only. The ice and compression groups were cooled to slightly different temperatures using a commercial ice machine (5°C and 10°C). The subjects who used less extreme cooling (10°C group) with simultaneous compression had significantly lower subjective pain scores and analgesic consumption compared to those using compression alone. In contrast, there were no significant differences in subjects treated with simultaneous cooling (5°C group) and compression and those treated with compression only in terms of subjective pain scores and analgesic consumption. This would suggest the degree of cooling has an effect on the effectiveness of ice therapy.

Despite (or perhaps because of) the diligent work of the authors, the findings of most randomised controlled trials provide confusing findings which make their interpretation and implementation into clinical practice difficult. If ice and compression are more effective than ice alone, and compression alone is as effective as ice and compression – then why do we use ice at all? On research findings this is a very valid question and the authors quite rightly call for more high-quality studies to ensure that clinicians and sportsmen are following evidence-based guidelines in the treatment of acute soft-tissue injuries.

They rightly state that future research must focus on developing modes, durations, and frequencies of ice application, which will optimise ice therapy during immediate and rehabilitative care. However this panacea in ice therapy may never be achieved. The individual parameters related to the patient and their injury may prove too wide ranging to allow a consensus on optimal treatment – those pesky patients always get in the way of randomised controlled trials related to physical therapy treatments.

In the absence of definitive research evidence physiotherapists in the ‘real world' must continue to do what they've always done – make clinical decisions on the modes, durations, and frequencies of ice application, based on the individual patient and their injury. If in doubt it would appear ice and compression can do no harm if applied properly – and one or the other seems better than no treatment at all in the acute stage of injury.

Go to the journal >