is icing really bad for you? What the science says on icing

Is Icing an Injury Really Bad for You? What the Science Says

Today’s article is an excellent review of the effects of cryotherapy, or ice, from my good friend Phil Page, PhD, PT, ATC, CSCS, FACSM.  Man, icing an injury sure has taken some heat (see what I did there…) lately on the internet.  There is a HUGE anti-ice movement.  I’m always amazed at how polarizing social media can be, with people screaming their black or white opinion, when in reality much of what we do is in the grey.  I get questions all the time about wether or not icing is good or bad for you, with many people quick to jump to the conclusion that we should not be icing.  Well, let’s find out what the research actually says.  Phil’s the Director of Research & Education with Performance Health, and one of the best at analyzing the research.


Is Icing an Injury Really Bad for You?

You’ve probably heard the debate on whether icing is helpful or harmful. You might be strongly on one side or the other, or maybe you aren’t sure which side you’re on because you’ve heard so many different things.

Despite what you might hear from anti-ice gurus that tend to be sensationalized on the Internet, let’s look at the facts and how we got here.

Ice isn’t the bad guy. Yes, we tend to apply ice in some situations that probably doesn’t help and claim we do so for the wrong reasons.  But the bottom line is that there are several benefits to ice, and ice has not been proven to impede the healing process as many claim.

About 30 years ago as a student athletic trainer at LSU, we frequently used ice, following the research of Dr. Ken Knight, who literally wrote the book on cryotherapy. I, as most other athletic trainers, was keenly aware of the mechanism of ice after an acute injury. As a graduate assistant athletic trainer for baseball at Mississippi State, I continued to advocate ice for my pitchers after they threw. Ice was my best friend.

Suddenly, stories came out that icing was bad for pitchers. As a matter of fact, one story back then was that it actually caused bursitis! Knowing a little about pathophysiology, I quickly dismissed that hogwash…  but the gears were in motion against using ice after pitching.

Fast forward to a few years ago. All of a sudden, ice is again demonized, but this time, it’s a vicious attack:

“Icing is wrong.”

“Ice impedes healing.”

“Icing is harmful.”

Say it ain’t so! Wha are we supposed to do?  Those are some bold claims!

The argument against ice tends to center around ice impeding the healing process as an ‘anti-inflammatory.’ Throughout the healing process (injury, inflammation, repair, remodeling), we need each of those stages to occur in order.  As an anti-inflammatory, the question was if ice actually creates an environment that does not allow the tissue to repair itself?  Interestingly, this same argument came out around the same time as people started questioning NSAIDS for the same reason!

Well, one study did get published (Tseng et al. 2013) titled, “Topical Cooling (Icing) Delays Recovery from Eccentric Exercise-Induced Muscle Damage.” The authors found increased signs of muscle damage after applying ice following eccentric exercise compared to a ‘sham’ application (although I’m not sure how you actually can apply ‘sham’ ice).

Bingo. Proof that ice impedes healing!  Right?  Hold on cowboy. That’s not the whole story.

What you didn’t hear about unless you actually read the study was that the authors concluded:

This study does not provide evidence on whether recovery from pitching-induced muscle damage would be slowed down by topical cooling.”

And while the authors found increased biomarkers in the group receiving cold therapy, there was no difference in strength or pain between the groups.  And I won’t even get into the question of adequate power with an n of 11.  You could argue that the study did not have enough subjects to have much clinical relevance.

Yet, ice was under attack again.

In addition, a few review studies of ice after ankle injuries raised more doubt on the practice of “RICE” (Rest, Ice, Compression, Elevation). The conclusion was that the quality of the research was generally poor quality, and the outcomes were inconclusive.

Note the word, “inconclusive” is not the same as “ineffective.”

And many times, effectiveness of icing was measured by the amount of swelling, rather than the actual healing process and return to activity. And while we know that ice doesn’t do much for swelling after the first 48 hours (Cote et al. 1988), modest cooling has been shown to reduce edema in animal studies (Collins 2008, Deal et al. 2002).

Yet, there we were, left to question if icing for recovery or after acute injuries was actually helping or hurting our athletes.  How did we get to this point?


The Claims Against Ice are Largely Based on Pseudoscience

The claim that ice is harmful by delaying the healing process is not supported by science. You may have seen bits and pieces of “science” in the false claim, but it’s a play on science that doesn’t give you the full picture or ability to make such a bold statement.  It’s called pseudoscience….statements that appear to be based on the scientific method, but are not.

Icing is not harmful or wrong to use.

You have witnessed a sham. Like the cup-and-ball game. It happens so fast and seems logical, but it’s a mind-trick.  Here are several things to consider.

Confirmation Bias

This is the tendency for us to accept evidence to confirm our own beliefs or theories. If you think ice is bad, you will tend to accept the information that supports your belief.  This makes us feel good because it confirms our prejudice.

False Logic

If inflammation (A) is necessary to get to healing (C), and ice (B) reduces inflammation (A), then ice (B) must reduce healing (C). FALSE. There is no direct evidence that icing reduces the healing process. In contrast, research supports the fact that ice does not impede healing (Vieira Ramos et al. 2016).  Granted, this was a study from an animal model, but who wants to be a human subject to test that theory?

Circumstantial Evidence

Evidence that attempts to prove a fact by connecting a related event or condition to a conclusion, as opposed to direct observation, is considered ‘circumstantial.’ This could be one of the most common ways science is used to incorrectly support claims. The presence of biomarkers in the blood may be an indirect measure of muscle damage, but it does not prove ‘cause-and-effect’. (Remember the DOMS study I referenced above?) Guilt by association is not the same as ‘causation.’ Using surrogate measures to make a definitive conclusion is a slippery slope.

Inconclusive Conclusions

Poor research (or no research) cannot serve as a basis for a conclusion on efficacy, let alone harm. The evidence on applying ice after an acute ankle injury is ‘inconclusive’ based on only a few studies of poor quality (Bleakley et al. 2004; van den Bekerom et al. 2012). There are no studies that applying ice after an ankle injury reduces recovery time (Hubbard et al. 2004). In fact, one study showed that early application of ice (< 36 hours) resulted in significantly faster return to play compared to delayed cryotherapy (Hocutt et al. 1982).

Comparing Apples to Oranges

Equating 2 things that appear similar, but are actually different, is not a fair comparison. Comparing DOMS to the healing process is not an accurate comparison. We know more about soft tissue healing after an injury than we do about the mechanism of DOMS, which is not a true model of an acute injury. Don’t forget, inflammation is not the same thing as swelling and edema!

Selective Science

Unbalanced reporting. Cherry-picking the literature. All signs of pseudoscience. The anti-ice movement has neglected years of research on the mechanism of ice after injury, focusing only on a select few studies that support (but in reality DON’T support) their argument. Dr. Knight explained that ice is not an ‘anti-inflammatory’ per-say (Knight, 1976); rather, it prevents the secondary injury to tissues by dampening the negative physiological effects of widespread inflammation. His position has been supported by other researchers as well (Ho et al. 1994, Merrick et al. 1999). And to top it off, one study quoted against icing (Bleakley et al. 2004) even concluded, “The sooner after injury cryotherapy is initiated, the more beneficial this reduction in metabolism will be.” Hmmm…the anti-ice crowd must have missed that statement.


The Benefits of Ice

Ice is not wrong or harmful.  The theory that ice impedes the normal healing response by limiting inflammation is not well documented in the literature. If you have been swayed by this on the internet, I would urge you to try to research this more and scrutinize the literature.  Be careful of what you see on the internet and ALWAYS seek to validate anything yourself.

Ice has plenty of benefits and clinical validation.

Proper application of cryotherapy can reduce secondary injury and reduce edema formation if applied within the first 36 to 48 hours (remember, ice doesn’t reduce swelling after the acute injury phase, and may not play a huge role in inflammation or recovery).  We do know that ice helps reduce pain, spasm, and guarding, allowing more mobility (Barber et al. 1998, Raynor et al. 2005).   More than anything, ice is a convenient and potent pain reliever, so it’s ok to apply ice to ‘chronic’ conditions as a safer pain reliever at any time. In fact, cryotherapy has been shown to decrease the amount of prescription pain medications needed after surgery (Barber et al. 1998, Raynor et al. 2005).

Sure, there are some times that ice is overused or erroneously used fort the wrong reasons, like reducing swelling after 48 hours.  The clinical research may not be conclusive, but there is no direct evidence that ice impedes healing. The argument that ice is ineffective or harmful is based on pseudoscience, and we need to be aware of this tactic.

Just be careful what you read, everyone has a bias.  #StandUp4Ice.


25 replies
    • Frederick Pauly
      Frederick Pauly says:

      This article basically says to scrutinize internet rabble debunking the value of ice. But is there an honest to god scientific study that proves ice helps healing? Do you consider icing an arm/shoulder after pitching a game to be icing an injury? Or is this just preventative care. To me, an injury is some real trauma like a sprained ankle. To me these seem to be two distinctively different scenarios. Has anyone studied preventative icing care vs. trauma care? Thanks. Rick

      • James Martin
        James Martin says:

        I don’t think there is evidence it helps healing. Healing takes as long as it takes and is likely influenced by many factors including diet, age, stress, robustness of the immune system and maybe even psychosocial factors.
        I think the point of the article was to articulate that ice may not be the Holy ggrail that it once was, but there is no evidence that actually impedes healing or that it does not serve a benefit.
        If it reduces risk of dependence on opioids, allows for cost effective and convenient pain relief and reduces guarding and improves willingness to move (when movement is not contraindicated (broken bone), then there is no significant risk or wrongness in using it.
        I tell people if they like it and it helps them feel better, have less pain and are more confident about moving and they take precautions to watch for complications (use layers, intermittent use and watch for burns) they can use it but I don’t tell them they are wrong if they don’t use it.

      • Mike Reinold
        Mike Reinold says:

        Hi Rick, not sure we are ever going to prove anything is perfect. The main point is that the internet debate on ice is largely based on false information and extremely biased. What we see is people hearing this pseudoscience and getting very argumentative about it, when in all reality there is benefit to ice. Much is grey in this world!

        To answer your questions, yes I would consider icing after you pitch to be an injury. It’s a microinjury, but it’s an injury to the tissue of the body.

  1. Denny Kolkebeck
    Denny Kolkebeck says:

    Mike and colleagues,

    Perhaps a different way of posing the question could be: “When should or shouldn’t you apply ice and why?”

    Perhaps we can also agree that any semblance of level 1 research is currently lacking on this topic–at least if exists I will plead ignorance of it’s existence– and that research from the Pac Man era may not be anything to hang your hat on.

    Within that framework then, a couple of uses for ice would be:
    1. Palliative relief for acute pain from “trauma”
    2. Vasoconstriction of local blood vessels–perhaps to slow bleeding.

    On flip side, couple drawbacks to using ice would be:
    1. Reduced muscle firing/activation (Krause et al, 2001).
    2. Increased force necessary to passively move the knee through range of motion (Petrofsky et al, 2013).
    3. Potential for inhibited tendon to bone healing in RCR (Cohen et al, 2006).

    As clinicians then under what scenarios do we apply ice to our clients?

    At the risk of being branded a heretic, I’ll share that I haven’t iced a post-op TKR or RCR in seven to eight years and my bias is they do better for it–“better” being operational defined as faster recovery of ROM, better muscle firing via descending and descending steps, and less post-op swelling.

    Interested in both sides of discussion and curious what future research is able to unearth.

    Be good!

    • Denny Kolkebeck
      Denny Kolkebeck says:

      Just reread what I wrote and realized sneaking in comments before your next client is always a bad idea!

      To Point #3 about inhibited tendon to bone healing, that study looked at NSAIDS.

      Mentally I was going to draw a possible theoretical parallel between the inflammation aspect of healing and the possibility of interfering with that when icing similar to NSAIDS, but had to wrap up and move on.

      Again, the discussion and different POV make for the interesting part of any topic!

    • Mike Reinold
      Mike Reinold says:

      Well said all around! I think there is still much to learn. Most of the rationale for, or against, is based on limited information. For example with TKA, if I have to choose, i’d take the pros over the cons as I think I can mitigate the cons fairly well. But the part that is hard to ignore is that many people like you never ice, and have great outcomes. So it’s clearly not mandatory to perform to get effective outcomes. That all being said, I don’t think the issue is people like you, it’s the ridiculous claims against ice that are false and biased right now that are misleading people. Thanks for your thoughts!

  2. Jared
    Jared says:

    No love for the numbing aspect? I think it’s valuable just to calm the patient down after some scar massage or PROM. The pain just jacks your BP wnd HR up and if you can get the patient to sit down for 15 minutes and numb the area-calm the patient down before they go on with the rest of their day. Give them a chance to breathe-especially shoulders and elbows.

  3. Matt Dietrich
    Matt Dietrich says:

    What are your thoughts on contrast baths? Currently doing them with an A2 pulley injury to help promote blood flow to the area. Switching between ice water and hot water. Was nervous when I saw articles about ice slowing the healing process.

  4. Kostas Sakellariou
    Kostas Sakellariou says:

    Well ther is one thing that we should take care of. If one puts ice directly after trauma, it may increase pain due to reduction of the pain threshold and can be really painful. On ther other hand let’s think about fever. What do we do to feel better? Do we ice our head? No. We just put a cool wet towel and change it every some minutes. We can use cold packs just like that not to decreace, but to control infrlammation and pain. But again, if you don’t protect your limb and load it in an uncontrolled manner it is silly to put ice or cold packs afterwards.

  5. Mark Phillips
    Mark Phillips says:

    One of the best articles I have read regarding this issue. I was a GA under Dr. Knight at ISU when he was completing his research and writing his book on cryokinetics. The misunderstanding I see most often repeated is the confusion in using ice/cold for acute care of an injury vs during rehabilitation/treatment after the secondary/hypoxic phase has passed. The application theory is distinctly different for each and should not be compared.

    • Phil Page
      Phil Page says:

      Thanks Mark. Spot on. Even Dr. Knight wrote that ice isn’t necessarily an ‘anti-inflammatory’ or reduces swelling. But the anti-ice people are playing off the misperception that everyone has… that ice prevents inflammation & reduces swelling. I’m not sure how we got away from Dr. Knight’s original philosophies (I have a theory!). And that’s what they base their argument on, even though the premise is false. It’s actually a “straw man” logical fallacy, where their argument is based on the wrong foundation. We need to stand up for ice!

      • gary reinl
        gary reinl says:

        Hi Phil and Mike, Gary Reinl the anti-ice man here. Since well over a million people have heard my “anti-ice” message … I wanted to let you know that I have never stated that ice “prevents” inflammation (delays yes … prevents no). I would like to address a couple of your comments but decided to address only one at this time:

        “But the bottom line is that there are several benefits to ice, and ice has not been proven to impede the healing process as many claim” Phil Page

        Taken verbatim directly from the dictionary
        “Impede verb
        past tense: impeded; past participle: impeded
        delay or prevent (someone or something) by obstructing them; hinder.
        “the sap causes swelling that can impede breathing”
        synonyms: hinder, obstruct, hamper, hold back/up, delay, interfere with, disrupt, retard, slow (down), hobble, cripple; More”

        When you apply ice, you “impede” or slow down the movement of nourishment and waste … are you suggesting that slowing down the movement of nourishment and waste doesn’t “impede” (delay) the healing process?

        PS: I’d love to meet you … who knows; maybe I could lead you out of the ice-age!


        • Phil Page
          Phil Page says:

          Hi Gary. Logically, it can’t be argued that ice is wrong or harmful simply because of a lack of evidence to support it as facilitating healing, or that the subsequent reduction in circulation directly impedes the healing process … there is no direct evidence that ice is detrimental to the recovery process. To reach a ‘harmful’ or ‘wrong’ conclusion about ice would mean that harmful outcomes have been measured and conclusively found to result from ice application. All the “evidence” against ice is circumstantial and mechanistic-based reasoning at best, leading to ‘inconclusive’ results. If you’re going to cite ‘research’ to support your position against ice, you also have to use science to make the conclusion that applying ice is detrimental to healing based on cause-and-effect. We shouldn’t use pseudoscience and play on words to support a strong position against ice. The argument that ice impedes healing simply by dampening the inflammatory response is not substantiated by outcomes research. Worse yet, to claim it’s ‘wrong’ and ‘harmful’ is unfair. I wholeheartedly agree that ice is often used for the wrong reason, and may not be beneficial for everyone, but we shouldn’t throw the baby out with the bathwater just yet.

          PS: I’m glad we have the opportunity to have this discussion…I’ve heard great things about you….and Thanks to Mike Reinold for making this happen.

          • gary reinl
            gary reinl says:

            Hi Phil, I like your style (and I have also heard great things about you). Thank you. Please consider meeting with me (you can leave a note @ and I will gladly call you to set up a time and place to meet). Although I don’t want to expand our online conversation at this time, I would like to present one counter point to your response (see my re-write below … I changed one word … I replaced against with for):
            “All the “evidence” against ice is circumstantial and mechanistic-based reasoning at best, leading to ‘inconclusive’ results.”
            All the “evidence” (save pain relief) for ice is circumstantial and mechanistic-based reasoning at best, leading to ‘inconclusive’ results.

            • Phil Page
              Phil Page says:

              I had a feeling you’d play that card :) I agree much of the evidence for ice isn’t direct. I would argue that the evidence still weighs heavily in favor of ice based on the “best evidence available.” But again, my point is that we can’t condemn ice altogether with blanket statements that it’s wrong or harmful (especially taking pain relief into consideration), and expect to use ‘science’ to support that. One positive out of all of this is pointing out the fact that we need to continue to research this topic, and not just use ice “because that’s what we’ve always done.” The problem is that ice as a modality doesn’t have big bucks behind it to sponsor the research…like pharma does. Until we meet…

  6. Michael Petrarca
    Michael Petrarca says:

    I’ve been on both sides of the fence on ice of recent and more I read on it the more I realize that it like any modality if used properly can have benefits. When treating someone I find what works for them and if they are getting the desired outcome they want. That being said too often ice is used as a lazy treatment when the symptoms do not idicate for it. I believe in getting a brief history and not giving someone ice just because they want it, but because they need it.


Trackbacks & Pingbacks

Leave a Reply

Want to join the discussion?
Feel free to contribute!

Leave a Reply

Your email address will not be published. Required fields are marked *