Research Review By Dr. Robert Rodine©

Date Posted:

May 2010

Review Title:

Cryotherapy: What does the literature tell us?

Background Information:

So why write a review on cryotherapy? It’s not exciting, it’s not new, not even clever. However, in an evidence-based healthcare environment, cryotherapy is an area of intervention where many of us are simply ‘out of the loop.’

Within day-to-day practice, many of us discuss home-based management with patients including stretching, exercises and managing positions of relief and aggravation. These recommendations often include suggestions for the use of heat and/or ice. Within my own practice, I often advise patients to use ice several times during the first 24 hours after their first manual treatment in order to reduce the effects of possible post-treatment soreness, to reduce the risk of bruising from pressure exerted to deep-tissues and to help decrease the general inflammatory cycle during acute injury.

I advise that the ice will help to reduce pain, reduce muscle spasm, improve recovery time and reduce the intensity and effect of inflammation. Seems intuitive, right? However, when challenged, I could not quote a single reference to support my points of view…in our current healthcare environment, that’s a problem! And when looking into it, one author put it best: ‘the strength of evidence supporting the use of cyrotherapy in the management of acute soft-tissue injury is generally poor… most randomized, controlled trials have focused on post-surgical patients…there is no evidence to suggest an optimal mode, duration, or frequency of ice application.’ (Bleakley 2006).

Considering this, I realized there are several important questions we need answerd regarding cyrotherapy:
  1. Does icing decrease surface temperature only, or does it decrease deep tissue temperature as well?
  2. If deep tissue temperature decreases, how long is that change sustained for?
  3. Does icing decrease pain?
  4. Does icing have other neurological effects?
  5. What is the best (and safest) way to use ice?
  6. And most importantly, does icing change outcomes for our patients?
To answer these questions, several articles will be presented and collaborated to create an evidence-based conclusion on what cryotherapy does, and whether it matters.

Articles Reviewed:

Dykstra et al. Comparisons of cubed ice, crushed ice, and wetted ice on intramuscular and surface temperature changes. Journal of Athletic Training 2009; 44(2): 136-141.

Objective: Through the use of three separate icing methods, the total amount of temperature change and length of time this change was sustained for, were evaluated.

Methods:
  • Six males (23.7 years; SD 1.0) and six females (22.8 years; SD 0.8) without history of lower leg injury within the previous one month and no known allergy to cold or ice received all three cyrotherapy methods, with at least four days between sessions.
  • Ice was contained within 22 x 40 cm polyethylene bags.
  • Methods included: 2000 ml of cubed ice; 2000 ml of crushed ice; and 2000 ml of cubed ice with 300 ml of room-temperature water.
  • Ice was applied to the posterior lower leg, at the area of largest girth and largest measured skinfold thickness. Sessions were 20 minutes in length.
  • Temperature was measured every 30 seconds for a 20 minute pre-treatment period, during the 20 minute treatment period and for 120 minute post-treatment period. Temperature was recorded at the cutaneous level via a surface thermocouple and the intramuscular temperature was recorded via a microprobe inserted into the lateral head of the gastrocnemius muscle to a depth of 2cm plus one-half of the skinfold thickness measured at the treatment area’s centre.
Results:
  • Crushed ice produced the smallest change in temperature (15.0°C at the surface and 4.3°C intramuscularly).
  • Cubed ice produced a max temperature change of 14.1°C at the surface and 4.8°C intramuscularly.
  • Wetted ice produced the greatest max temperature change of 17.0°C at the surface and 6.0°C intramuscularly.
  • The greatest change in surface temperature was achieved at 18 minutes of ice treatment for all methods.
  • Intramuscular temperature reached lowest levels post-treatment for all methods, with wetted ice producing the greatest sustained change from 2 minutes post-treatment to 75 minutes post-treatment.
Overall Conclusion: A wetted ice pack is more effective at decreasing treatment area temperature, and at maintaining this decrease than cubed or crushed ice.

Anaya-Terroba et al. Effects of ice massage on pressure pain thresholds and electromyography activity postexercise: a randomized controlled crossover study. J Manipulative Physiol Ther 2010; 33: 212-219.

Objective: To investigate the effect of ice massage on pain pressure thresholds of the quadriceps muscle following isokinetic exercise.

Methods:
  • 7 males and 8 females (19 years; SD 2) without prior history of trauma/fracture, lower quadrant pain or injury within the previous 12 months and without regular use of analgesic or anti-inflammatory medications participated.
  • Subjects were randomized to receive their first intervention, and received the alternative intervention a week later.
  • Interventions consisted of ice massage directly over the dominant quadriceps muscle for a 15 minute period using 200 ml of frozen water in the form of a cylinder. The ‘placebo’ treatment was a 15 minute application of a detuned ultrasound machine over the same treatment area. (Caution should be used when referring to this placebo treatment, as a true placebo intervention should resemble the experimental intervention. In this case, the interventions are not comparable from the participant’s point of view and the subject is therefore not blinded appropriately….though all things considered, it is pretty difficult to design a sham icing protocol!)
  • Pain pressure thresholds were obtained from each subject using an electronic algometer. The mean of 3 measures taken 30 seconds apart was recorded. Measurements were taken at baseline, post-exercise (isokinetic procedure paired with EMG recording) and 5 minutes post-intervention (ice vs. detuned ultrasound). Algometer readings were taken mid-belly of the rectus femoris (RF), mid-belly of the vastus lateralis (VL) and mid-belly of the vastus medialis (VM).
  • EMG was recorded while subjects performed one weight bearing isometric maximal voluntary contraction of their dominant quadriceps. Three recordings of five seconds duration were taken, with two minute rest intervals. (EMG technicians were blinded to treatment allocation). EMG was recorded at baseline, post-exercise (isokinetic procedure described below), and 5 minutes post-intervention.
  • The pre-intervention isokinetic exercise procedure consisted of subjects performing a warm-up period of three repetitions at 120°/s and three repetitions performed at 60% MVC for familiarization. Following one minute of rest, subjects performed five concentric isokinetic knee extensions on the dominant leg at 60°/s, 120°/s, 180°/s and 240°/s, all at maximum effort.
  • Subsequent to this performance, subjects received their randomly assigned treatment intervention.
Results:
  • Immediately following exercise, pain pressure thresholds over the VM decreased from a mean of 337.9 (SD 122.0) to 319.7 (SD 132.7) and over the RF from 333.4 (SD 96.7) to 308.0 (SD 108.5), a statistically significant difference. This was not noted over the VL.
  • This should be interpreted that as pain pressure thresholds decreased, no post-exercise hypoalgesia was produced from the isokinetic exercise within the recruited subjects.
  • Following ice massage, pain pressure thresholds increased over the VM (365.1; SD 129.9) and VL (356.7; SD 139.1) as compared to the detuned ultrasound group over the VM (316.1; SD 162.7) and VL (317.7; SD 147.8). These differences were found to be statistically significant. No difference was noted over the RF.
  • This should be interpreted that post-intervention hypoalgesia was produced from the ice massage over the VM and VL specifically, but not with detuned ultrasound.
  • EMG recordings were increased in the VL only, following ice massage as compared to following detuned ultrasound. This is interpreted by the authors that ice massage may increase motor unit recruitment.
Overall Conclusion: Ice massage increases pressure pain thresholds in healthy subjects following isokinetic exercise and increases EMG activity, indicating a possible activation of descending inhibitory pathways.

Herrera et al. Motor and sensory nerve conduction are affected differently by ice pack, ice massage, and cold water immersion. Physical Therapy 2010; 90(4): 581-591.

Objective: To compare the effects a resting ice pack, ice massage and cold water immersion on nerve conduction velocity in the tibial and sural nerves.

Methods:
  • 18 males and 18 females (20.5 yrs; SD 1.9) were recruited. 3 were not included, leaving 36 participants. No baseline differences were noted between subjects. Subjects were randomly allocated, 12 per treatment group.
  • Participants received cold modalities at the same time of the day (between 2 and 6 pm) to reduce circadian influences. A 15 minute pre-cooling rest period within a temperature controlled room initiated the study protocol, acclimatizing the subjects.
  • Tissue cooling was performed on the posterior lower right leg, within an 18 x 8 cm rectangular area.
  • Cooling was applied for a 15 minute duration via an ice pack containing 279g of crushed ice in an 18 x 8 cm plastic bag, ice massage via 279g ice block measuring 8x10x5 cm, or through cold water immersion via a 20x35x30cm plastic tub filled with crushed ice and water, reaching a steady temperature of 10°C.
  • Nerve conduction velocity of the sural and tibal nerves was conducted immediately before and immediately after cold application.
  • Skin temperature was measured using an infrared thermometer, recorded before and after cold application from the centre of the 18 x 8 cm area.
Results:
  • Post cooling, there was a statistically significant difference in skin temperature after all three cold applications.
  • The ice pack reduced skin temperature 24.43°C (SD 2.87), ice massage reduced the temperature by 27.6°C (SD 1.32) and the cold water immersion reduced the temperature by 18.23°C (SD 1.46).
  • All three cold modalities decreased nerve conduction velocity of the sural and tibial nerves.
  • No differences in nerve conduction changes were noted between the application of ice massage versus an ice pack.
  • Cold water immersion however produced a greater effect on nerve conduction velocity when compared to ice massage. Specific values of NCV will not be presented in the interest of space. Please refer to the original article for greater detail.
  • To recap, the authors state within their paper that skin temperature has a linear relationship with nerve conduction velocity. Therefore, as skin temperature decreases, so will nerve conduction. However, in this study cold water immersion created the smallest change in skin temperature and the greatest change in nerve conduction velocity. The authors postulate that as cold water immersion affects a greater surface area it leads to a faster/stronger thermoregulatory response from the body which aims to protect against strong temperature changes. Therefore, surface temperature would demonstrate a smaller change compared to the other cold modalities.
Overall Conclusion: The application of cold therapy decreases nerve conduction velocity, and increases the latency and duration of the action potentials, thus nerve current is reduced and the refractory period is increased. This explains the hyoalgesic effect of cryotherapy. While ice massage produces the greatest change in skin temperature, cold water immersion produces the greatest effect on nerve conduction velocity and most likely to produce therapeutic effects.

Bleakley et al. Cryotherapy for acute ankle sprains: a randomized controlled study of two different icing protocols. British J Sports Med 2006; 40: 700-705.

Objective: Through randomized controlled methods, to compare a standard icing protocol to an intermittent protocol in the management of acute ankle sprains.

Methods:
  • Subjects were included if they suffered a mild/moderate ankle sprain within the previous 48 hours.
  • Subjects were excluded if there was evidence of bony injury, if they were under 16 years of age, suffered multiple injuries suffered cold aversion/allergies.
  • Participants were stratified according to sports and general population and then randomized to receive either the standard icing protocol or the intermittent protocol.
  • Ice pack modality consisted of filling a 20x20cm bag with water, freezing it solid, running it under hot water for 30 seconds, wrapping it in a single layer moist towel. The ice pack was then placed directly on the skin.
  • The standard protocol consisted of 20 minutes of continuous icing to the injured area performed every two hours for a 72 hour period.
  • The intermittent protocol consisted of 10 minutes of icing, 10 minutes of rest and 10 minutes of icing. This procedure was repeated every 2 hours for a 72 hour period.
  • All subjects were instructed on range of motion and proprioception exercises to be performed during recovery. Compliance was monitored through the use of a treatment diary.
  • Outcome measures included the Binkley lower extremity functional scale, a 10 cm Visual Analogue Pain Scale and ankle swelling as determined by the figure eight method when compared to the non-injured ankle.
  • Outcome measures were recorded at baseline, and one, two, three, four and six weeks post-injury.
Results:
  • A total of 89 subjects participated (58 male; 29.9 years of age SD 10.32).
  • No baseline differences were noted treatment groups, with comparable division of general population and sport stratified subjects.
  • LEFS scores, pain at rest and degree of swelling improved in all groups, over time. No significant differences however were noted over time between groups.
  • Pain on activity improved in both groups over time. However, a univariate analysis of covariance revealed that between the baseline measure and one week post-injury, the intermittent icing group had greater reduction in pain on activity.
Overall Conclusion: Short and intermittent applications of cryotherapy may promote greater pain relief when compared to continuous application in acute ankle sprain. This may be transferable to other acute soft-tissue injuries.

Bleakley et al. The use of ice in the treatment of acute soft-tissue injury: a systematic review of randomized controlled trials. Am J Sports Med 2004; 32:251-261.

Objective: Through systematic review, to determine if the use of ice during the acute phase of soft-tissue injury results in improved outcomes.

Methods:
  • Randomized controlled studies, published in the English language, utilizing human subjects with either acute soft-tissue injury or have received an orthopaedic procedure who received cryotherapy were included in the study.
  • MEDLINE, ProQuest, SINAHL, AMED, Cochrane and Web of Science databases were searched.
  • Methodological quality was evaluated using the PEDro scale.
  • Two reviewers conducted methodological appraisals.
Results:
  • 55 articles were retrieved for review, narrowed down to 22 inclusions.
  • PEDro scores ranged from 1 to 5, with a mean of 3.4.
  • No studies evaluated subjects with muscle contusions or strains.
  • 5 studies evaluated subjects with acute ligament sprains.
  • The remaining 17 studies evaluated subjects recovering from operative procedures.
  • The mode of cryotherapy varied, as did the duration and frequency of its application.
  • Based on the high variability of studies, presentation of results is difficult.
Overall Conclusion: Conclusions are difficult to draw from such a heterogenous group. This being said, the review did conclude that cryotherapy is effective at decreasing pain levels in those suffering acute injuries.

Now come the big-dogs!

Brosseau et al. Thermotherapy for treatment of osteoarthritis. Cochrane Database of Systematic Reviews 2003, Issue 4. Art No.: CD004522.
&
French et al. Superficial heat or cold for low back pain. Cochrane Database of Systematic Reviews 2006, Issue 1. Art No.: CD004750.


As per usual, reviews written for the Cochrane Database are quite extensive, meet high methodological standards and point out more flaws and holes within a research area than provide conclusions derived from extensive high-level evidence.

However, these two reviews to provide some information for us to consider clinically when referring to cryotherapy.

Brosseau et al. surrounded osteoarthritis of the knee specifically, aiming to examine outcome measures such as pain, swelling and range of motion. Reviewers were able to identify three RCT’s, consisting of 179 subjects. Obviously the cryotherapy methods differed between studies, making it difficult to pool data.

One study used 20 minutes of continuous icing, 5 days per week. Following 3 weeks of treatment, the authors found a significant difference in quadriceps strength in the interventional group, when compared to the control group. As for range of motion and functional status, a statistically significant difference was noted in favour of the interventional group, however this difference was not found to be clinically significant.

The second study evaluated 10 sessions of either hot or cold treatment. Details were not provided. The investigators evaluated swelling via knee circumference and found a statistically significant difference at the end of the study, subsequent to 10 sessions. Clinical significance of this change is unknown.

The third study compared ice administered 3 times per week for 3 weeks to that of a control group. No further details were provided. The investigators found no difference in pain levels at the end of the 3 week treatment plan, nor at 3 month follow-up.

French et al. sought to evaluate superficial heat or cold in patients with low back pain. The authors identified nine trials including a total of 1117 patients. Of these nine trials however, only three related to cryotherapy. Two compared ice massage to superficial heat packs and the third compared ice massage to TENS therapy. No comparisons were made between cryotherapy and placebo.

From this review, the conclusions are difficult to draw with respect to cryotherapy. When compared to other interventions (in this case TENS), cyrotherapy was found to be equally effect as TENS at reducing pain in patients with chronic low back pain. When compared to heat however, there was insufficient evidence to draw firm conclusions. This being said, one trial found heat and cold to be equitable at reducing pain in a mixed subset of back pain sufferers, while the other found cold to be superior at reducing pain in chronic low back pain patients.

Clinical Application & Conclusions:

Okay, so we have reviewed several articles now and you have spent a lot of time reading this! Are you better off?

Let’s review our questions from the beginning.

1) Does icing decrease surface temperature only, or does it decrease deep tissue temperature as well?

According to Dykstra et al., cyrotherapy does reduce the temperature of deep tissues. This study found that wetted ice was the best method, as it created the greatest drop in intramuscular temperature, rated at a 6.0°C change. Most importantly, the intramuscular change was noted maximally following 18 minutes of continuous icing.

The idea behind wetted ice created a greater temperature change was that the wetted bag of ice would better conform to the surface of the skin and more efficiently conduct thermal energy away from the surface of the skin. With this thinking, one would expect that cold water immersion would be the best method of decreasing both surface and intramuscular temperature.

Contrarily however, Herrara et al. found that cold water immersion decreased surface temperature to a smaller degree when compared to using an ice pack or ice massage. It was thought that this resulted from a large surface area of the body receiving cold exposure and therefore reacting with a greater thermoregulatory response. The degree of intramuscular change however was not measured, and is therefore unknown with cold water immersion. Also, despite the higher surface temperature noted with cold water immersion, this method of cryotherapy surpassed the others with respect to nerve conduction parameters.

2) If deep tissue temperature decreases, how long is that sustained for?

Dykstra et al. demonstrated that the deep tissues held a decreased temperature the longest following the use of wetted ice, noting that a significant difference held for up to 72 minutes.

3) Does icing decreasing pain?

The Cochrane reviews provided minimal information on the effect of cryotherapy for analgesia, as this was not studied in osteoarthritic patients and proved inconclusive in low back pain patients.

Increased pressure pain threshold, indicating an analgesic effect, was noted by Anaya-Terroba et al. Here it was found that healthy subjects without previous pain or injury were able to sustain higher pressure thresholds subsequent to ice massage as compared to those who received the ‘placebo’ treatment.

Bleakley et al. (2004) found that cryotherapy was effective at decreasing pain in acute soft-tissue injuries. It should be noted however that acute myogenic pain was not studied and most trials reviewed pertained to post-surgical patients.

4) Does icing have other neurological effects?

Based on the work of Herrara et al., it is clear that neurological effects are produced as a result of cyrotherapy based on the change in nerve conduction parameters. The authors conclude that this becomes most clinically relevant with respect to a decrease in muscle spasm and hypertonicity, in addition to the perception of pain.

5) What is the best (and safest) way to use ice?

The best method of icing remains elusive as there is insufficient evidence of comparisons. As a result we have to interpret the literature for ourselves and consider the patient’s preference and capabilities.

With respect to improving long-term outcomes, we cannot definitively choose one method over another. However, with respect to reducing pain levels, swelling and nerve conduction parameters, icing modalities that have a greater conformity to the skin seem to perform much better than cruder applications such as solid ice packs or cubed ice.

Ice massage was also shown to be effective in some studies, though this method can be messy. Trials that evaluated ice massage tended to use a frozen block or cylinder of ice and rub it directly on the skin. As the ice melts, this results in water….everywhere! Therefore, patients are likely to find this method practically inferior.

One method that seemed to demonstrate the most congruency with the skin was cold water immersion. However this exposes a much greater surface area of the body to colder temperatures. Additionally, depending on the body-part affected this method may be quite difficult. Again, patients are likely to find this method practically inferior.

The best practical solution therefore is wetted ice. When combined with water the modality tends to better conform to the skins surface. Better conformity apparently leads to a greater decrease in both superficial and deep tissue temperature, having a superior effect on nerve conduction parameters and pain. This method of cryotherapy is less messy for patients use at home and relatively easy to repair.

While comparable to the conformity of wetted ice packs, the use of gelled packs or bean bags has not been fully evaluated. Therefore, we cannot be truly confident about the effectiveness of their use. Though, I think we can all agree this is an easy, clean and practical way for patients to adhere to cryotherapy at home. As patients are not required to repeatedly prepare a bag of wetted ice, the gel pack may increase compliance.

As far as duration goes, there is still little agreement between studies. However, Bleakley et al. (2006) did demonstrate that intermittent icing using a 10-10-10 principle was more effective at reducing pain upon activity as compared to a 20 minute continuous icing protocol. This is a research area that should be evaluated more extensively in order to determine the full effect of different modalities and protocols.

6) And most importantly, does icing change outcomes for our patients?

Unfortunately, knowledge of long term outcome following cryotherapy for acute soft-tissue injury is minimal and insufficient. Also, there is a lack of information with respect to return-to-participation, return-to-work and disability levels.

However, based on the presented papers, we can confidently say that cryotherapy does have a positive effect on pain levels. Whether the use of ice helps to reduce the rate of mild-moderate adverse effects of manual therapies such as soft-tissue treatments and spinal manipulation, remains to be studied.

Now this has not been an extensive review of cryotherapy literature. There are still relevant articles out there for review! Regardless, with the information gained from the included articles you are now better informed and can consider your recommendations to be more evidence-based!