Clinical library · Modalities
Cryotherapy
The most reflexively applied treatment in musculoskeletal practice. It produces statistically significant reductions in postoperative pain that sit below the minimal clinically important difference — and for acute ankle sprain, the whole literature is two trials at high risk of bias.
In one line. Local cooling by ice, cold packs, cold-water immersion or cooled-air devices, to reduce pain, swelling and metabolic demand after injury or surgery.
It is cheap, safe, and patients like it. The reason to read the evidence carefully is that the size of the benefit is consistently smaller than the confidence with which it is prescribed — and in the one indication every clinician associates it with, acute ankle sprain, it is barely studied at all.
Statistically significant, clinically marginal. A meta-analysis of 28 randomised trials in postoperative musculoskeletal care found cryotherapy significantly better for pain at immediate (MD -0.77), short (-0.84) and medium term (-0.41) — and states that these effect sizes were below the minimal clinically important difference. Range of motion showed small to medium effects; the function benefit was explicitly labelled not clinically relevant. Its conclusion: minor benefits that may not be clinically relevant. [1]
What the evidence shows
| Question | Finding | Source and certainty |
|---|---|---|
| Postoperative musculoskeletal care | Pain MD -0.77 (immediate), -0.84 (short), -0.41 (medium term) — all below the MCID | 28 RCTs; GRADE very low to moderate. Swelling and function: low certainty, no substantial benefit [1] |
| After total knee replacement, blood loss | 825 mL without cryotherapy versus 561 mL with; MD 264 mL less (95% CI 7 to 516 mL less) | 12 trials, 956 participants. Low certainty, downgraded for bias and inconsistency [2] |
| After total knee replacement, overall certainty | Low for blood loss, pain and range of motion; very low for transfusion rate, function and adverse events | 22 trials, 1,839 participants. Blinding is difficult with this intervention and most studies had high risk of performance and detection bias [2] |
| Acute ankle sprain | No enhancement of other treatment for swelling, pain or range of motion | Two RCTs, both high risk of bias. The review calls current literature lacking in evidence and urges larger high-quality trials [3] |
| Closed surgical incisions | Pain SMD -0.50 (day 1) and -0.63 (day 2); opioid consumption reduced by 7.43 morphine milliequivalents | 51 RCTs, 3,425 patients, moderate certainty. No effect on length of stay or surgical site infection [4] |
| Recovery after strenuous exercise | Cold-water immersion superior to other methods for muscle soreness; similar for power and flexibility | 28 studies. Air cryotherapy beat cold-water immersion for strength and immediate power recovery. Water temperature and duration were rarely moderators [5] |
| Delayed onset muscle soreness | Cold within 1 hour reduces pain up to 24 h (SMD -0.57); no clear effect beyond 24 h | 32 RCTs, 1,098 patients. No significant difference between cold and heat overall [6] |
Where it misleads
1. Significance without clinical importance
This is the central finding. In postoperative musculoskeletal care the pain differences were statistically significant at every time point and all of them fell below the minimal clinically important difference; the function benefit was explicitly flagged as not clinically relevant. [1] A treatment can be reliably better and still not better enough to matter to the patient.
2. Acute ankle sprain is nearly evidence-free
The indication cryotherapy is most associated with has two randomised trials, both at high risk of bias, both testing it as an addition to other treatment, and both showing it did not enhance them. The review's conclusion is that current literature lacks evidence supporting its use in acute ankle sprain and that there is an urgent call for larger high-quality trials. [3] See ankle sprain for what does have evidence.
3. The knee replacement evidence is low and very low certainty throughout
The Cochrane review is unusually explicit about why: the nature of cryotherapy makes blinding difficult and most studies had a high risk of performance and detection bias. Certainty was low for blood loss, pain and range of motion, and very low for transfusion rate, function and adverse events. [2] The 264 mL blood-loss difference has a confidence interval running from 7 mL to 516 mL — the lower bound is clinically meaningless.
4. The strongest evidence is not for the physiotherapy indication
The one moderate-certainty finding on this page concerns cryotherapy applied to closed surgical incisions, reducing pain and opioid consumption across 51 trials and 3,425 patients. [4] That is a perioperative nursing and surgical intervention, not a musculoskeletal rehabilitation one, and it should not be cited as though it validated ice after a sprain.
5. Colder and longer is not better
In the cold-water immersion meta-regression, water temperature and exposure duration were rarely exposure moderators. [5] The common practice of escalating cold or duration to increase effect is not supported by the dose-response data.
What the evidence supports — and what it does not
Supported
- Reduced pain and opioid use after closed surgical incisions, on moderate certainty. [4]
- Cold-water immersion for muscle soreness after strenuous exercise. [5]
- Cold within an hour of exercise for DOMS pain up to 24 hours. [6]
- Small improvements in range of motion after musculoskeletal surgery. [1]
- Safety and low cost across all of the above.
Not supported
- Clinically important pain relief after musculoskeletal surgery. Effect sizes below the MCID. [1]
- Acute ankle sprain. Two high-risk trials, no benefit as an add-on. [3]
- Confident claims after knee replacement. Low to very low certainty throughout. [2]
- Effects beyond 24 hours in DOMS. [6]
- Colder or longer application as more effective. Temperature and duration were rarely moderators. [5]
How certain is this?
Evidence grade: Low to moderate, depending on indication.
The grade differs by question more than for any other modality on this site. Postoperative incisional pain carries moderate certainty from 51 trials. [4] Recovery after exercise is supported by a well-conducted meta-analysis with meta-regression. [5][6] Musculoskeletal postoperative care is very low to moderate certainty with effect sizes below the MCID. [1] Total knee replacement is low to very low certainty throughout. [2] Acute ankle sprain has essentially no usable evidence. [3]
The recurring methodological problem is stated plainly by the Cochrane authors: this intervention cannot be blinded convincingly, so performance and detection bias are structural rather than incidental. [2] That caps how good this evidence can get.
What would change the grade: adequately powered sham-controlled trials in acute soft tissue injury, and reporting against the MCID rather than against zero.
Common questions
Should I still recommend ice for an acute ankle sprain?
You can, on comfort grounds, but not by citing evidence. The systematic review found two randomised trials, both at high risk of bias, showing cryotherapy did not enhance the effects of other treatment on swelling, pain or range of motion, and concluded that current literature lacks evidence supporting its use, with an urgent call for larger trials. [3] It is inexpensive and safe; that is a different argument from effective.
Does it help after knee replacement?
Possibly, on low-certainty evidence. Across 22 trials and 1,839 participants, blood loss was 825 mL without and 561 mL with cryotherapy — a difference of 264 mL, with a confidence interval from 7 to 516 mL. Certainty was low for blood loss, pain and range of motion and very low for transfusion rate, function and adverse events, largely because the intervention cannot be blinded. [2]
How much pain relief should I expect after surgery?
Real but small. Across 28 postoperative musculoskeletal trials the mean differences were -0.77, -0.84 and -0.41 at immediate, short and medium term — every one below the minimal clinically important difference. [1] For closed surgical incisions the picture is better: standardised mean differences of -0.50 and -0.63 on days 1 and 2 with moderate certainty, plus reduced opioid consumption. [4]
Colder water, or longer?
Neither, on the available dose-response data. In a meta-regression across 28 studies, water temperature and exposure duration were rarely exposure moderators of the recovery effect. [5] Note also that air cryotherapy outperformed cold-water immersion for recovery of muscular strength and immediate power.
Ice or heat for muscle soreness?
Both work and neither is clearly better. In a meta-analysis of 32 trials, cold applied within an hour reduced pain up to 24 hours (SMD -0.57) with no clear effect beyond that, while heat reduced pain both within and beyond 24 hours — and there was no significant difference between the cold and heat groups overall. [6] See superficial heat.
References
- Miranda JP, Figueiredo RCC, Saragiotto B, et al. Effectiveness of cryotherapy on pain intensity, range of motion, swelling and function in the postoperative care of musculoskeletal disorders: a systematic review and meta-analysis of randomised controlled trials. British Journal of Sports Medicine. 2026 Mar 17;60(6):440–449. doi:10.1136/bjsports-2024-109497 PMID 41047148 Systematic review and meta-analysis
- Aggarwal A, Adie S, Harris IA, et al. Cryotherapy following total knee replacement. Cochrane Database of Systematic Reviews. 2025 Oct 30;10(10):CD007911. doi:10.1002/14651858.CD007911.pub4 PMID 41165130 Cochrane systematic review
- Miranda JP, Silva WT, Silva HJ, et al. Effectiveness of cryotherapy on pain intensity, swelling, range of motion, function and recurrence in acute ankle sprain: A systematic review of randomized controlled trials. Physical Therapy in Sport. 2021 May;49:243–249. doi:10.1016/j.ptsp.2021.03.011 PMID 33813154 Systematic review
- Muaddi H, Lillie E, Silva S, et al. The Effect of Cryotherapy Application on Postoperative Pain: A Systematic Review and Meta-analysis. Annals of Surgery. 2023 Feb 1;277(2):e257–e265. doi:10.1097/SLA.0000000000004987 PMID 34856580 Systematic review and meta-analysis
- Moore E, Fuller JT, Bellenger CR, et al. Effects of Cold-Water Immersion Compared with Other Recovery Modalities on Athletic Performance Following Acute Strenuous Exercise in Physically Active Participants: A Systematic Review, Meta-Analysis, and Meta-Regression. Sports Medicine. 2023 Mar;53(3):687–705. doi:10.1007/s40279-022-01800-1 PMID 36527593 Systematic review, meta-analysis and meta-regression
- Wang Y, Li S, Zhang Y, et al. Heat and cold therapy reduce pain in patients with delayed onset muscle soreness: A systematic review and meta-analysis of 32 randomized controlled trials. Physical Therapy in Sport. 2021 Mar;48:177–187. doi:10.1016/j.ptsp.2021.01.004 PMID 33493991 Systematic review and meta-analysis
About this resource
Using this in clinic
Every figure here is traceable to its source.
This page reports each effect against the minimal clinically important difference rather than against zero, because for this modality that is where most of the apparent benefit disappears. Where a value could not be verified against the paper it came from, it is not on this page, and the omission is stated rather than filled with a number from a secondary source.