Detailed guides to painful problems, treatments & more

Poisoned by Massage

Rather than being DE-toxifying, deep tissue massage may actually cause a toxic situation

Paul Ingraham • 40m read

It turns out that massage therapy may be toxic, which is both interesting science and at odds with popular beliefs about massage. People believe massage is de-toxifying, and that is not only not true, but quite possibly the opposite of true. Intense massage may be routinely harming people, especially some vulnerable populations, people who mostly don’t even know that they are vulnerable. Specifically, there is a strong scientific case that strong massage can damage muscle, spilling proteins into the blood and clogging up the kidneys — a type of literal poisoning.

Toxification, not detoxification!

Photo of a woman receiving a massage. The scene is peaceful, but biological toxic waste hazard symbol is superimposed on her back.

Massage is toxic? Technically, it may well be. (But so is good scotch. And hard exercise.)

The idea of “toxins” is often unethically exploited in alternative medicine, usually to sell detoxification treatments.1 The toxin talkers should either be specific when they talk about toxins, or stop talking about toxins. I am going to be specific, and there’s no detox solution for sale here. (And I’m probably sacrificing a small fortune. Pesky ethics!) Before we talk about any toxins, however, let’s establish that there’s anything that needs explaining…

Feel a bit sick after massage?

Weirdly, it’s common to feel a bit gross after a massage… and sometimes very gross. People routinely suffer from varying degrees of soreness and malaise following firmer massage therapy, a phenomenon known as post-massage soreness & malaise, or PMSM.23 The worst cases feel like a full-blown flu, except it doesn’t last nearly as long. Severe PMSM is really unpleasant, even serious, and so it’s probably safe to assume that something goes wrong, a significant and poorly understood side effect of massage. Most of this article is about the worst cases, which are much harder to explain than the milder ones, but I devote a section below to exploring the causes of milder cases as well.

Stronger massage tends to cause more PMSM — though the effect is as unpredictable as weather. It is often explained away as a necessary evil, a “healing crisis” that we must endure to get to the benefits on the far side. No-pain-no-gain is the message (mostly unspoken). The effect is often explained more specifically as the symptom of a healthy detoxification process with a (rather doubtful) happy result: unspecified toxins trapped in weary, tight muscles are freed and flushed from muscle, poisoning us mildly as they make their way out of the body via kidneys and livers. There are lot of problems with that mental picture, and they are discussed in detail in Why Drink Water After Massage?

The flushing model is bogus. Massage isn’t “liberating” any toxins that were there when the client walked in the door. If anything, ironically, massage may be creating them — by-products of minor muscle injury. That would mean that massage is actually a “toxifying” treatment, not a de-toxifying one. (This is actually one of the original reasons for the idea that toxins need to be flushed out from the body by drinking some extra water.) If so, it’s interesting how the detox myth has turned itself inside out over the years. Few massage therapists have considered the possibility that the client leaves with a problem they didn’t arrive with. Unfortunately, it’s a strong possibility.

And it’s not such a radical idea. PMSM is real. Something icky is clearly going on after some massages. What?

Rhabdo the Terrible: a medical emergency (which has nothing to do with massage)

Rhabdomyolysis — or just “rhabdo” for short, and for the rest of this article — is a medical emergency that mostly crops up in the aftermath of horrible, crushing traumas.4 Rhabdo can kill, even when injuries are otherwise non-lethal. It is most often seen in the wild in the aftermath of earthquakes and car accidents. It was one of the (wrong) diagnoses suggested by Dr. House in a 2009 episode, “Known Unknowns.”

When muscle is injured, cellular guts are spilled into the blood, most notably myoglobin molecules, which messes with blood chemistry a bit, poisons the kidneys, and turns your pee dark brown. If it gets to the point of kidney damage, it has to be treated with a lot of intravenous fluids to dilute the poisons in the kidneys.

Rhabdo the Terrible, the medical emergency, really has nothing to do with massage. Be aware of this if you mention a connection to a doctor — you will get scoffed at.

But … what if there’s only a little myoglobin? What about mild rhabdo? Can massage cause mild, non-emergency rhabdo? Almost certainly, it turns out.5 In fact, if you get a strong enough massage, it can even — almost — cause severe rhabdo.

An extreme-massage horror story: a case of genuine rhabdomyolysis apparently induced by massage, with kidney danger and everything

In 2006, an elderly gentleman collapsed with a fever and failing kidneys.7 There are any number of reasons that might happen to an 88-year-old man, but the diagnosis turned out to be a fairly clear cut case of rhabdomyolysis.8 And there was a smoking gun of a cause. A veteran of regular massage for years, he’d had an unusually intense massage the day before. “Intense” is actually a bit of an understatement …

The afternoon before this accident, he received a body massage session for 2 h served by two new massagists [sic] at the same time instead of one. The strength of this massage session was significantly stronger than that of the past.

Safety tip: don’t get a brutal two-hour massage from a pair of “massagists.” 🙄

Clearly this was an exceptional massage. Nevertheless, assuming the rhabdo wasn’t caused by something else,9 his case demonstrates that massage probably can damage muscle. If he hadn’t actually collapsed, he still would have felt perfectly awful — in other words, he would have had wicked PMSM.

More massage-induced rhabdo case studies

There are some other confirmed cases of post-massage rhabdo. Another weird, extreme example is from 2009: an awful experience with a reaction to infrared heat and regular massage over several days.10 In 2020, a woman was hospitalized with extremely severe rhabdomyolysis after more typical use of a vibrating massage tool — a “massage gun,” an extremely popular type of product right now.11 Bear in mind that reported cases of adverse effects like this are usually just the tip of an iceberg.

And then there’s the ones I am hearing about directly from my readers…

A reader’s rhabdo story (and the chilling possibility that some people are much more vulnerable than others)

Many people experiencing PMSM are finding this article. They feel rotten after a massage — beaten up and gross, properly sick — and they start Googling. And so, for a couple years now, I have received a steady supply of e-mailed tales of the consequences of intense massages. Many of these are probably not about rhabdo,12 but others are too clear and precise to dismiss.13 It’s impossible to infer real prevalence data about this problem from my deeply skewed sample, and yet there are obviously approximately “quite a few” people out there who are not feeling good after their massages — and pissing the Coke-coloured urine that more or less confirms rhabdomyolysis.

Here’s just one good example, which also highlights another important point: that some people are probably more biologically vulnerable to this phenomenon than others. Reader Moriah Dyke of Langely, Canada (shared here with permission):

About a year ago I had a particularly rough massage therapy session and the next day at work I started feeling super sick so I went home (I never get sick, ever) so I layed on the couch whatever went to bed but when I woke up the next morning I literally felt like I had been hit by a bus. I called in sick to work as I could barely get out of bed. The body aches I was experiencing were like nothing I’ve experienced before, I could barely walk! I felt like someone was stabbing tiny knives all through out my major muscles. I stayed in the couch all day and was in so much pain I started googling to see if I could find out what was wrong and I came across rhabdo. My pee was dark brown and I had all of the other symptoms. I was in so much pain I was crying on the couch and so I called the nurses line to see if they could give me any advice. They didn’t believe it was rhabdo but said even if it was all they do is give fluids (and I’m scared of needles) so I didn’t end up going to the hospital but instead took a ton of vitamins and drank an outrageous amount of water and electrolytes to try and help.

It took almost two weeks for the body aches to subside 100%. I’ve seriously never experienced anything like this in my entire life.

The reason I bring this up is very important. I am in the process of trying to get a diagnosis of Ehlers Danlos Syndrome (a connective tissue disorder) which makes tissue extra fragile. After realizing I might have this condition it seems a lot more likely that massage could rip my tissue to the point where it causes rhabdo which means my experience was probably not in my head at all but a real situation. EDS is severely undiagnosed and some people believe that 1/200 people could have his condition with only 1/5000 being actually receiving a diagnosis. This means that a lot of people could be thinking they’re crazy when doctors tell them rhabdo isn’t possible from a massage when in fact they could have a condition that makes it a lot more likely to have this happen.

I thought my personal experience and the link between a connective tissue disorder and rhabdo might be of benefit to you in any future articles you may choose to write.

Moriah Dyke, Langley, Canada

Moriah is right: EDS is both underdiagnosed and surprisingly common,14 and it is a real and worrisome possibility that people with EDS are at much greater risk of injury when subjected to powerful massage. It’s probably even worse, in fact.

EDS is part of a spectrum of connective tissue diseases that are all basically about fragility of tissue.15 Most have genetic markers, but not all: hypermobility-type EDS, the most common sub-type of the disease, has not yet been linked to a gene. People with EDS suffer from a wide array of strange musculoskeletal pains and problems,16 many of which drive them to seek therapy, and many massage therapists are all too likely try to treat them by vigorously stretching their connective tissue (fascia) — precisely the opposite of what they need, but it’s an extremely popular trend in therapy, based on misguided enthusiasm for the idea that fascia gets stuck or deformed.17

I shudder to think how many people with pathologically fragile fascia have been subjected to treatment specifically intended to stretch fascia. But even without that specific therapeutic goal, these patients are probably much more easily harmed by any kind of “deep tissue” massage, and they may account for many or most stories about post-massage rhabdomyolysis. It’s conceivable this is a common collision of pathology and manual therapy that almost no one is aware of, patients or practitioners.

Disintegrating muscle tissue!

Rhabdo the Terrible is mostly defined by nasty muscle damage and kidneys dying from myoglobin poisoning — right before you die. It’s the kidney damage that particularly makes rhabdo a medical emergency. But there’s a lot of other things going on — a cocktail of substances that cause a lot of other symptoms (ahem, malaise) many hours before your kidneys are wounded.

Notably, frank trauma is not the only way to get rhabdo, just one of the worst and most obvious. Instructions for releasing lots of myoglobin into the blood:

You’re done! But rhabdo is not just caused by “crushed” muscle but by muscle that, in the words of Wikipedia, “rapidly breaks down” … by any means.

Such as? Space alien disintegrator rays? Giant monster fly regurgitating digestive enzymes onto your quadriceps? How, pray tell, does muscle ever “rapidly break down”?18

This is where things get a bit weirder.

Rhabdo the Merely Unpleasant (a.k.a. recreational rhabdo)

Read up on rhabdo and it becomes clear that while the worst cases are caused by things like buildings falling on people’s legs, there are plenty of other less traumatic causes:

What this odd list of causes tells us is that acute rhabdomyolysis is the tip of a physiological iceberg, the bad end of a broad spectrum.19 Muscles will obviously spill their chemical guts in response to much milder stresses and forces than overt crushing.

Relatively minor, exertional rhabdomyolysis is actually common, and even has a name: “white collar rhabdomyolysis.”20 That term was coined by Knochel in 1990 because rhabdo was striking recreationally extreme athletes — people who voluntarily work themselves into a sorry state.21 You could also call it (for fun) recreational rhabdo. Another well-known source of rhabdo cases is boot camp: “large numbers of [recruits] may have myoglobinemia … .”22

After a bit of browsing through the literature, I have the impression that you could be rhabdo-ized by an especially hard sneeze.

These are real cases of rhabdomyolysis, with quantities of myoglobin in the blood ranging from mildly oogy to alarming, and plenty of mineral and electrolyte derangement — but they are also generally less severe and less likely to cause kidney damage. They are somewhere in the middle of the rhabdo spectrum.

This is also roughly where the older man landed on the spectrum after his brutal 2-man, 2-hour massage. That case is probably the worst-case scenario for massage-induced rhabdo … but it’s also a pretty bad worst-case scenario. If that’s what a strong two-hour massage can do, what can a strong one-hour massage do? Maybe to someone with a genetic predisposition to rhabdo?23

Drawing of a runner with the caption, “Rhabdomyolysis: nature’s way of telling you to stop fucking running already.”

Disclaimer: this is obviously all hypothetical

It’s a hypothesis that mild rhabdomyolysis may explain PMSM. It’s based on no direct evidence beyond case studies and anecdotes. But there’s an awful lot of smoke!

Anecdotes are not data, but they are a fine justification for research (which probably won’t happen24). And the undeniable existence of exertional rhabdo makes it highly plausible that milder rhabdo can probably also be caused by other biological stresses. It’s not a big leap from there to realizing that painfully strong massage could be one of those stresses.

This is “just” a hypothesis, but it’s a highly plausible one.

Mild rhabdo as a common denominator in feeling cruddy after virtually any physical stress

To understand common experiences, it is often helpful to learn about their extreme and exaggerated forms. What’s going on here? Turn it up! Acute rhabdo is interesting and revealing because it almost certainly represents the extreme form of a tamer and nearly universal human experience: feeling sore and cruddy after physical stress.

Even fairly severe rhabdo can have symptoms that are basically just intense exhaustion. If that’s possible, then milder rhabdo can probably be downright subtle. One reader, hospitalized for almost two weeks for exertional rhabdo, mostly just felt like he’d badly overdone it — worn out, but not sick. (For that whole story, see Radu’s rhabdo: a story of intense exercise gone wrong.)

The lower half of the rhabdo spectrum is messy and fascinating. It is undoubtedly biochemically diverse, with the exact cocktail depending heavily on many variables like genes, fitness, and the type of stress. Regardless, if relatively normal people like runners and soldiers can get middlin’ rhabdo, then it is virtually guaranteed that less extreme stresses are routinely drivers of less extreme rhabdo. Much of this would fall well below the threshold of what would ever be diagnosed as rhabdomyolysis. Indeed, at the lowest end of the spectrum we simple have other names for it …

“Soreness,” for instance.

Rhabdo the Merely Unpleasant clearly overlaps — mostly or entirely — with the familiar phenomenon of being really sore after strong exercise (usually called DOMS, for delayed onset muscle soreness). And that, in turn, feels almost exactly like PMSM. Quite appropriately, one of the names for DOMS is “muscle fever.”

These experiences are probably just mild versions of something that can get a lot more dire, but usually doesn’t. But the nature of the mild phenomenon is nicely illuminated by the nature of its more serious versions. Although the biochemical details are absurdly complex, the general theme is not — we are poisoned by the by-products of relatively minor tissue stress.

Even a bit of massage. You can almost define massage as “muscle crushing,” and rhabdo is caused by muscle crush injuries and even much milder damage to muscles. Therefore massage probably causes rhabdo, and it is the most likely explanation for PMSM.

So what? The implications are surprisingly unexciting

This is all very interesting, but it may not be very important.

As I mentioned at the start, “Poisoned By Massage” is a sensational title. I chose it to help promote the article, of course — but I wouldn’t have chosen it if I didn’t also believe that “poisoned by massage” actually is scientifically defensible (as well as provocative and interesting). I think the evidence and reasoning is good, and I have some confidence that massage actually does “poison” us, a little.

But so what? So does exercise! And we aren’t quitting that. It is nearly impossible to progress in fitness without “poisoning” yourself with a little DOMS, almost regularly. It is not entirely unreasonable to call it a “healing crisis” — an unpleasant price to pay for some benefits.25

The issue with massage is that the benefits are much less clear. No one really knows exactly what good massage does for people, medically speaking.26 It’s rather murky. We don’t even know if it makes us “fitter” to adapt to the stress of PMSM. Being toughened up by massage might be as dubious as toughening up your feet so you can walk barefoot — that’s all fine and good, but do you need tough feet? Probably not.

Fortunately, we do know that good massage is intrinsically satisfying as a sensory experience, and that is probably a good enough reason to put up with some side effects. However, I am also sure that I don’t want to dial up the PMSM and embrace it either. It can mostly be avoided with gentler treatment, and probably should be.27

Life is stressful. It is a multi-decade process of adapting to stresses. I’m happy to tolerate a little mild rhabdo along the way — but I’m certainly going to try to minimize it.

PMSM: What about people who feel lousy after a gentle massage treatment?


Members-only area unlocked. See your account page.

I suspect it’s unusual for people to get PMSM from gentle massage, and when they do it’s probably mild PMSM, just not in the same league as the worst cases. While this article generates a lot of email from people who feel really rotten after a strong massage, I have yet to receive a report of major malaise after light massage. The combo of sensationally bad PMSM with gentle massage is clearly a rarity. While noteworthy, there’s just less to explain here.

Nevertheless, gentler massage can cause some blahs. I often heard about these less dramatic cases during my career as an RMT, and I’ve experienced a few myself. So I am sure that it does occur, and it would be nice to understand. It’s highly implausible that gentle rubbing could cause significant biological harm, so I suspect the causes are diverse, and mostly don’t have much to do with the massage itself. For instance…

Sub-clinical infections. Many cases are probably explained away by a simple coincidence: sometimes you just happen to get sick soon after a massage, just bad timing. Viral and bacterial infections are more common than massage therapy, and they will collide occasionally.

Infections too mild to diagnose (or even inspire the attempt) are surprisingly common, but those can still increase our vulnerability to physical stresses significantly. This is surely one reason why we’ve all had days when a familiar workout feels surprisingly hard and makes us more sore than it “should.” Even a light massage in such biological conditions might have the same effect.

Chronic mild pathological vulnerability. In addition to mild infections, there are a variety of diseases and chronic illnesses that can go undiagnosed for years, or even forever, and those might also dial up vulnerability, either transiently during flare-ups, or all the time.

“Sickness behaviour.” The phenomenon of PMSM is so similar to the aching malaise that comes with viral infections that we have to consider both the possibility of actual infections… and the possibility that something about massage has the same effect as an infection, which isn’t as weird as it might sound at first. Feeling gross is technically known as “sickness behaviour,” a set of symptoms generated by an ancient biological system shared by all animals: when threatened, neuroinflammation enforces rest and retreat by making us feel fatigued and fragile.28

I do not know why mild massage would provoke a neuroinflammatory response, but neuroinflammation and sickness behaviour have such deep roots in biology that it wouldn’t surprise me if they can be provoked in surprising ways. This isn’t very solid physiological ground,29 but it’s fun to brainstorm about.

Extreme de-conditioning. Consider all the people who are so out of shape that they would feel slammed by a 10-minute jog, a dose of exercise that any fit person would blast through with no ill effect. For the truly unfit, a “gentle” massage might not be so gentle. We shouldn’t underestimate the potency of significant deconditioning. Anyone who has trained really out-of-shape people — or someone trying to claw their way back from a severe injury — knows how startlingly “fragile” they can be.

Emotional and psychological factors. Finally, there’s undoubtedly some psychological PMSM. Even a happy, relaxing massage experience may expose or highlight an exhaustion that was already there, like taking off a mask. Life is hard and many people are running on empty, pushing through the days powered by caffeine, worry, grit, and necessity. Massage is a rare chance to relax and unwind, but it may also have the potential to disarm us, to rob us of our normal defenses against the suffocating fatigue lurking just under the surface. I believe I have experienced precisely this flavour of PMSM on many occasions … and on most Sundays, frankly.30

It’s extremely unlikely that PMSM is caused by a Jarisch-Herxheimer reaction

Many massage therapists have speculated that PMSM is caused by a Jarisch-Herxheimer reaction, which is being poisoned by the corpses of dead bacteria during antibiotic treatment. The most polite response to this idea I’m capable of is that it is wildly speculative, citation needed, and good luck with that. It rests heavily on huge optimistic assumptions about the powers of massage, specifically its ability to cause such an extraordinary surge in immune system activity that it could kill off enough bacteria to cause this effect.31

And what infection is this, anyway? Are we also hypothesizing that people with serious enough infections to cause a J-H reaction are also asymptomatic enough that they are seeking out massage therapy?

I’ve been writing about this stuff long enough to avoid categorically declaring that anything is “impossible” — I always want to leave a little wiggle room — but this is about as far out in left field as massage claims can get. For some alternative medicine practitioners, I think “Jarisch-Herxheimer reaction” has just become a handy way to say “healing crisis” in a science-y way.


The flushing theory goes down the drain

This entire article started out as an attempt to clarify a point about “flushing” in my article Water Fever and the Fear of Chronic Dehydration. The effort got a little out of hand, and I ended up with a huge article about rhabdomyolysis. However, I still need to address the flushing thing …

Photograph of a glass of water.

If PMSM is basically a light poisoning, does this validate the popular practice of advising massage clients to drink extra water? Mostly no — it’s not specifically, medically relevant.32 It’s sensible to avoid compounding mild rhabdo with dehydration. But it’s also sensible to avoid actual dehydration, period.

If acute rhabdo is treated with IV fluids, doesn’t that suggest that drinking water would help? In mild rhabdo, we’re not worried about kidney damage at all.33 And drinking an extra glass or two of water wouldn’t (remotely) do the job anyway. Indeed, there’s an ironic catch-22: if you drank enough water to successfully dilute a dangerous concentration of myoglobin in your kidneys, you’d actually poison yourself. With water.34

Consider: Can you “treat” the much more familiar problem of DOMS by drinking extra water? Answer: definitely not. Try drinking three extra, big glasses of water the next time you’re super sore after a big workout. Still really sore? You can count on it.

In principle, the only medical condition that can be effectively treated by drinking more water is … dehydration. A number of disease processes and even toxins are dehydrating, which can in turn be treated with hydration. The obvious example is alcohol.35 But neither massage nor rhabdomyolsis is “dehydrating.” (I’ve actually heard the claim that massage is hydrating — but that’s too kooky to bother debunking.)

Dehydration is inherently dangerous, and will particularly aggravate any condition where the concentration of toxins in your body fluids is a factor, and this can be helped by diluting the toxin with a higher blood volume — but only a little, and not enough to dilute toxins.36 Acute rhabdomyolysis is treated with intravenous fluids because, again, you literally cannot drink enough to dilute the toxins enough to protect the kidney.37

Many people will guess that drinking makes you pee more, which means more kidney action, which means more processing and filtration of some blood borne toxins — but the logic is just the same as with blood volume. It’s true-yet-trivial.38

Dehydration is worth avoiding in itself, of course, and it’s not illogical to make sure you avoid it after a massage: who wants both pre-rhabdo and dehydration at the same time? Not me. Unfortunately, in no way is extra hydration really “protective” — there’s just not that much difference between hydrated and un-hydrated.

Responses to criticisms

I certainly expected to be criticized for being “negative” about massage in this article — and I was criticized for it — but I was also called out for overconfidently presenting a theory without evidence. That’s ironic: I’ve put in plenty of time fighting hype about premature conclusions, and I would hate to be the source of one. Am I?

Whilst not disputing that your hypothesis may very well be correct and it is an interesting and plausible position, I feel that your conclusions are just a little strong without being backed by sufficient evidence.

“Inner West Mobile Massage” commenting on Facebook

I appreciated the spirit of that criticism, and I even agreed with it and immediately changed one key overconfident sentence in the conclusion. But as I reviewed this whole article looking for other signs of overconfidence, my concern eased. There are a lot of qualifiers and equivocation here! I say “maybe” a lot. It’s clear from the style that I’m hypothesizing, thinking out loud. I do like the theory, and I suspect it’s probably correct, but I don’t think I present it as a sure thing — nowhere close.

More criticisms arrived. On Twitter, for instance, @SwoleClient accused me of “unfounded claims,” arguing that I was basing my entire argument on a single case study. Others echoed the sentiment.

I do not agree with that. There are two main issues here:

  1. confusion about what a “claim” is, and where the burden of proof is
  2. levels of confidence and evidence

“Claim” need not apply here

A claim is any unverified assertion. But not all claims are created equal. In health care and health science, “claim” implies a more self-serving assertion. If a claim could be used as a bullet-point in a sales pitch, it’s more claim-y. If it makes you (or your profession) look better, it’s more claim-y. And the more claim-y it is, the more it needs to be backed up.

This special case of the word claim comes from the thorny ethical challenges with selling care to sick, hurt people. All claims need critical appraisal and verification, but it’s just not as ethically critical if it has no claim-stink. Sagan’s idea that “extraordinary claims require extraordinary evidence” is not just about alien abductions and lake monsters. It’s also, in spirit, about more mundane but self-serving and profitable claims — a more common ethical hazard than truly extraordinary claims.

I presented my theory to cast doubt on classic self-serving claims, claims that serve the interests of massage therapists: the popular and specific claim that massage “detoxifies,” and the more general claim that massage is beneficial (or at least safe). The “claim” that I made in Poisoned by Massage is more self-defeating than self-serving. It superficially contradicts my widely published opinion that people should purchase massage services and devote considerable energy to self-massage (although I explain in the article that it’s not really much of a concern). My “negativity” about massage and many other treatments does great harm to my bottom line. I could probably make much more money just by shutting up about things like this.

So the accusation of “unfounded claim” doesn’t fit well — technically correct, but definitely wrong in spirit. What I presented is just a partially supported theory.

There is exactly no evidence whatsoever that massage detoxifies. But I presented some evidence and a plausible theory that it does, in fact, do just the opposite. This is more about putting a dent in the detoxification claim than proving my own hypothesis — my own case simply has to be “reasonable” to cast significant doubt on the claim.

Levels of confidence and evidence

Many professionals believe (or fear) that evidence-based medicine is uptight and narrow, and ignores their clinical experience (even though EBM has always emphasized that evidence is only one important factor in clinical reasoning, see Science versus Experience in Musculoskeletal Medicine.).

Now, when I defend a theory with imperfect data, I got flak from some of the same people for not presenting enough evidence! Bah.

My case for rhabdo as the cause of post-massage soreness and malaise was not just based on a single case study, as several critics implied. I argued it from a few different angles, and supplied a fair bit of “circumstantial” evidence (indirect evidence). Chalk the whole thing up to “you based everything on one case study,” as a few people did, is a facile synopsis.

But neither did I say that the evidence was strong or that my conclusion is airtight. The level of evidence was modest, the reasoning reasonable — and my level of confidence matched it.

It’s a trap! An example

One crafty critic tried to trap me in an apparent inconsistency:

Earlier this year you complained that there wasn’t enough evidence for the claim that massage reduces inflammation. But now you’re claiming with weak evidence that massage causes rhabdo. What gives?

Here’s what gives: the researchers involved in that study made a big massage-boosting claim that was implausible to begin with in any number of ways, and which was poorly supported only by only a single indirect line of evidence. The claim was big, so the burden of proof was heavy, and yet all they had was some extraordinarily elaborate test tube evidence, full of problems, and far removed from any sound clinical implications. And yet they summed it up very confidently for the media and the journal very definitively as “massage reduces inflammation.”

Simply not the same thing.

A missed opportunity

Histrocially, there was something wrong with this article, but no one caught it: its most significant outstanding flaw for a long time was the absence of any discussion of alternative explanations for the phenomenon of post-massage soreness and malaise. “How else can I explain this?” is a pretty basic part of critical thinking, and it’s missing. I didn’t go there until mid-2020, when I added the section about milder cases. And there’s still more I could do. Maybe next month.

About Paul Ingraham

Headshot of Paul Ingraham, short hair, neat beard, suit jacket.

I am a science writer in Vancouver, Canada. I was a Registered Massage Therapist for a decade and the assistant editor of for several years. I’ve had many injuries as a runner and ultimate player, and I’ve been a chronic pain patient myself since 2015. Full bio. See you on Facebook or Twitter., or subscribe:

Related Resources

Ruth Werner, author of A Massage Therapist’s Guide to Pathology, did a recent episode of her “I Have a Client Who…” podcast about rhabdomyolysis: “A young woman has lots of muscle pain after her soccer game. A great time to receive massage, right? Not this time. Listen in for a case where massage could cause renal damage.”

Can Massage Cause Muscle and Nerve Damage?” by Ravensara Travillian explores two potential consequences of intense massage, both rhabdomyolysis and peripheral neuropathy.

The major myths about massage therapy are:

The complete list of dubious ideas in massage therapy is much larger. See my general massage science article. Or you can listen to me talk about it for an hour (interview).

And massage is still awesome! It’s important to understand the myths, but there’s more to massage. Are you an ethical, progressive, science-loving massage therapist? Is all this debunking causing a crisis of faith in your profession? This one’s for you: Reassurance for Massage Therapists: How ethical, progressive, science-respecting massage therapists can thrive in a profession badly polluted with nonsense.

What’s new in this article?

Five updates have been logged for this article since publication (2012). All updates are logged to show a long term commitment to quality, accuracy, and currency. more Like good footnotes, update logging sets apart from most other health websites and blogs. It’s fine print, but important fine print, in the same spirit of transparency as the editing history available for Wikipedia pages.

I log any change to articles that might be of interest to a keen reader. Complete update logging started in 2016. Prior to that, I only logged major updates for the most popular and controversial articles.

See the What’s New? page for updates to all recent site updates.

2020 — Cited a new case study of massage-induced rhabdo, added a link to another good quality anecdote about exertional rabdo. Although a minor change, this is now quite a lot of “smoke” and I am starting to thinking that the rhabdo hypothesis isn’t so exotic anymore, so I also changed some language to express more serious concern about the potential seriousness of the situation.

2020 — New section and sub-topic, “PMSM: What about people who feel lousy after a gentle massage treatment?” Polished and updated the article summary. Added a short section about the evidence and the plausibility of the rhabdomyolysis hypothesis.

2018 — New section and sub-topic, “A reader’s rhabdo story (and the chilling possibility that some people are much more vulnerable than others).”

2017 — Miscellaneous minor editing, and a new section, “No, PMSM is not caused by a Jarisch-Herxheimer reaction.”

2017 — Added a sidebar about how many PMSM stories I’ve been receiving from readers.

2012 — Publication.


  1. The idea of “toxins” is usually used as a tactic to scare people into buying de-toxifying snake oil. Exactly what substances and how they are to be removed is always vague, because the sellers are making it up. The body deals with unwanted molecules in many ways; the only truly detoxifying treatments help the body eliminate or disarm molecules the body cannot process on its own, like a stomach pump or an antivenom. Anything less, like mildly stimulating one normal excretion pathway in a sauna, is a detox scam.
  2. Cambron JA, Dexheimer J, Coe P, Swenson R. Side-effects of massage therapy: a cross-sectional study of 100 clients. J Altern Complement Med. 2007 Oct;13(8):793–6. PubMed 17983334 ❐ This study found that about 10% of people reported “some minor discomfort” the day after massages — which were fairly gentle, I presume. I doubt they would have tested painfully strong massage without mentioning the intensity. However, painfully strong massage is quite common “in the wild.” And the rate of soreness would likely go up sharply with the intensity. For contrast, a more general study of all kinds of manual therapy (see Carnes) found that 20-40% of treatments will cause some kind of unpleasantness, side effect or “adverse event” in medicalspeak.
  3. A few readers have challenged “often,” arguing that the phenomenon is rare. Therapists who have heard little about it from their own patients may simply be doing their jobs well: good massage is generally going to be gentle enough to avoid triggering post-massage soreness and malaise. Nevertheless, I assert that it is quite common — in large part because recklessly intense massage is also common. We’re not likely to ever get much hard data on this, so the true incidence is going to remain a mystery, but the 10–40% rates reported by Cambron and Carnes make me confident that PMSM is hardly “rare.” Even if it’s only happening to one patient in 20, that would still a huge number of cases around the world every year.
  4. Rhabdomyolysis.
  5. Posadzki P, Ernst E. The safety of massage therapy: an update of a systematic review. Focus on Alternative and Complementary Therapies. 2013;18(1):27–32. PainSci Bibliography 53974 ❐ Ernst and Posadzki found seventeen case reports of many different kinds of adverse effect, including rhabdomyolsis.
  6. Fearmongering” is the deliberate and unjustified arousal of public concern, usually to serve an ulterior motive. I am often accused of fearmongering because I express concerns about potentially dangerous treatments. I plead “not guilty”: I aim for nuance and balance whenever I write anything that might spook anyone, and I make a point of weighing risks and benefits. Raising awareness of legitimate risks is not fearmongering — it’s valuable public health education. It would be fearmongering if I was making an issue out of nothing, but it’s never about “nothing.”

    “You’re not paranoid if they’re really after you,” and it’s not fearmongering if concern is rational.

  7. Lai MY, Yang SP, Chao Y, Lee PC, Lee SD. Fever with acute renal failure due to body massage-induced rhabdomyolysis. Journal of Nephrology, Dialysis and Transplantation. 2006 Jan;21(1):233–4. PubMed 16204282 ❐ PainSci Bibliography 54301 ❐
  8. The key indicator molecule is creatine phosphokinase (CPK). Rhabdo is “official” when kidney damage starts around 20,000 U/I of CPK, and he had just about 8,000 around day 3 when it was first measured. It had probably peaked higher.
  9. I think that assumption is quite safe. It’s hardly guaranteed, of course. A single case study is always prone to misinterpretation. There are other causes of rhabdo, and elderly people have many vulnerabilities. If the rhabdo was caused by something else, then this could be a classic case of post hoc ergo propter hoc — the logical fallacy of assuming that something was caused by whatever event it follows, which is not necessarily so. But the fallacy is the assumption, and in fact causes do precede their effects. And here we have a specific effect with a likely cause immediately preceding it.
  10. Tanriover MD, Guven GS, Topeli A. An unusual complication: prolonged myopathy due to an alternative medical therapy with heat and massage. South Med J. 2009 Sep;102(9):966–8. PubMed 19668045 ❐ The trouble started after several days. His neck and arms became swollen, the pain “unbearable,” and his “serum muscle enzymes were increased” — rhabdo? Some degree of it, very likely. The presence of rhabdo also implicates the massage itself as a significant mechanism of injury.

    (It’s hard to imagine any sane amount of infrared heating causing such a horrible reaction, and I don’t think it would increase muscle enzymes in the blood under any circumstances. Heat exhaustion would, but that’s a more obvious threat that wasn’t even mentioned. As an interesting side note, some people also believe that infrared (and other) saunas are “detoxifying,” and yet here we have a case where it clearly failed to do that.)

  11. Chen J, Zhang F, Chen H, Pan H. {Rhabdomyolysis After the Use of Percussion Massage Gun: A Case Report}. Phys Ther. 2020 11. PainSci Bibliography 51936 ❐


    Percussion massage guns are commonly used by professional athletes and nonathletes worldwide for warmup and physical recovery; however, there are no published clinical or evidence-based reports on percussion guns regarding their benefits, indications, contraindications, and even side effects. The purpose of this case report is to describe the first case of rhabdomyolysis as a severe and potentially life-threatening illness following use of a percussion gun. A young Chinese woman with untreated iron deficiency anemia presented with fatigue and pain in her thigh muscles for 3 days and tea-colored urine for 1 day, after cycling and subsequently receiving percussion gun treatment by her coach for the purpose of massage and relaxing tired muscles. Muscle tenderness and multiple hematomas were found on her thighs, and her urinalysis indicated hemoglobinuria. Her serum creatine kinase was reported as “undetectably high,” a hallmark of serious muscle damage leading to a diagnosis of severe rhabdomyolysis. Aggressive intravenous fluid resuscitation, urine alkalinization via intravenous alkaline solution, assessment of urine output, and maintenance of electrolyte balance were administered during hospitalization. The patient’s clinical presentation gradually improved with the decline of creatine kinase, and she recovered well during follow-up. A case of severe rhabdomyolysis after percussion massage should alert caregivers, sports professionals, and the public to suspect and recognize the potentially serious adverse effects of percussion guns and to ensure that percussion massage guns be used appropriately and safely in rehabilitation therapy, especially in individuals with an underlying disease or condition. Research is needed to examine the benefits, indications, contraindications, and adverse reactions of percussion guns.

  12. For instance, many of the reports of alleged rhabdomyolysis I have received are probably just about coincidental infections. Infections and massages are both common, so sometimes they happen on the same day.
  13. If someone reports fever and malaise after a massage, if they confirm that they did not have respiratory symptoms, and they confirm that they did have dark urine, only then do I take rhabdomyolysis seriously as a diagnostic possibility. At this point, I have a few dozen reports that are fairly clearly in this category, and at least a dozen that are slam dunks — far from proof, but as good as anecdotal evidence gets.
  14. Rodgers KR, Gui J, Dinulos MBP, Chou RC. Ehlers-Danlos syndrome hypermobility type is associated with rheumatic diseases. Sci Rep. 2017 Jan;7:39636. PubMed 28051109 ❐ PainSci Bibliography 52757 ❐
  15. Most notably hypermobile-type Ehlers–Danlos syndrome (EDS), and several more sub-types of EDS, and then several more poorly understood conditions under the umbrella of hypermobility spectrum disorders (HSD). “The essential difference between HSD and hEDS lies in the stricter criteria for hEDS compared to the HSD.” But it’s very tricky, and those criteria are all very new: see the 2017 EDS International Classification. See the PainSci guide to this topic: Hypermobility and Ehlers-Danlos syndrome.
  16. Scheper MC, de Vries JE, Verbunt J, Engelbert RH. Chronic pain in hypermobility syndrome and Ehlers-Danlos syndrome (hypermobility type): it is a challenge. J Pain Res. 2015;8:591–601. PubMed 26316810 ❐ PainSci Bibliography 52758 ❐ Hypermobility is “highly prevalent among patients diagnosed with chronic pain.”
  17. Many massage therapists are selling “fascial therapy” to patients. The main idea is that fascia — sheets of tough connective tissue found throughout the body — can get tight and restricting, and needs to be “released” by pulling on it. Fascia science is considered an exciting frontier in manual therapy. Unfortunately, although some fascia biology is interesting, the stuff does not seem to have any properties that are actually relevant to healing and therapy. Key examples of fascia research either fail to support fascial therapy or actually undermine it. Enthusiasm about fascia seems to be an unjustified fad. See Does Fascia Matter? A detailed critical analysis of the clinical relevance of fascia science and fascia properties.
  18. I knew this article was going to be awesome when the learning curve forced me to ask those questions.
  19. Landau ME, Kenney K, Deuster P, Campbell W. Exertional rhabdomyolysis: a clinical review with a focus on genetic influences. Journal of Clinical Neuromuscular Disease. 2012 Mar;13(3):122–36. PubMed 22538307 ❐
  20. Knochel JP. Catastrophic medical events with exhaustive exercise: "white collar rhabdomyolysis". Kidney International. 1990 Oct;38(4):709–19. PubMed 2232508 ❐
  21. As opposed to, say, doing it because you’re a peasant farmer and you’re really screwed if you don’t get the harvest in.
  22. Olerud JE, Homer LD, Carroll HW. Incidence of acute exertional rhabdomyolysis. Serum myoglobin and enzyme levels as indicators of muscle injury. Arch Intern Med. 1976 Jun;136(6):692–7. PubMed 1275626 ❐ PainSci Bibliography 54278 ❐
  23. Yes, they surely exist, of course — again, see Landau et al. In fact, I suspect I am one of those people — which is one reason why I’ve written this article.
  24. Tough to get research funding for this even at the best of times. Tough even to get research enthusiasm! Who’d do it? 90% of massage research has been done by people assuming its benefits and eager to prove them—none would have been keen to investigate serious harm.
  25. Clarification: I don’t think it’s correct to imply that the DOMS is actually part of or the cause of “healing.” There’s a fairly big ethical difference between telling someone, “This DOMS stuff is an unavoidable side effect” versus “DOMS is unpleasant, but it’s for your own good.” No. It’s just a price to pay.
  26. Ingraham. Does Massage Therapy Work? A review of the science of massage therapy … such as it is. 23729 words.
  27. Ingraham. The Pressure Question in Massage Therapy: What’s the right amount of pressure to apply to muscles in massage therapy and self-massage? 6686 words.
  28. For a full discussion of neuroinflammation — and it’s a fascinating topic — see Chronic, Subtle, Systemic Inflammation: One possible sneaky cause of puzzling chronic pain.
  29. Woodburn SC, Bollinger JL, Wohleb ES. The semantics of microglia activation: neuroinflammation, homeostasis, and stress. J Neuroinflammation. 2021 Nov;18(1):258. PubMed 34742308 ❐ PainSci Bibliography 52006 ❐ This paper argues persuasively that the idea of “neuroinflammation” applies only to quite a well-defined and serious phenomenon, which occurs only in the presence of very serious injury and illness.
  30. No joke. I have often been amazed by the way I feel both relaxed and lousy on Sundays. For many years I have treated Sunday as a (non-religiously) sacred day, a guaranteed reprieve from work. That habit is so well-established now that I suspect every nerve in my body knows that it’s an opportunity to not be “on”… and as soon as I stop pushing myself, what I mainly notice is that I’m utterly exhausted! A weariness so deep it can easily be confused with malaise.
  31. Here’s one of the most straightforward examples I found of this kind of thinking:

    When you receive a massage, various body systems are given a temporary boost. Blood circulation is enhanced, lymph flow is improved and the immune system starts to become a bit more efficient. If you are a person who gets little or no exercise (not uncommon in our culture these days), who has never gotten a massage, or only gets a massage once or twice a year, your body will be stimulated just enough to stir things up a bit.

  32. If mild rhabdo is indeed the cause of PMSM, the mechanism is probably just the circulation of a variety of molecules, all of which will have different processing pathways, and drinking extra water just has little or nothing to do with how fast that happens. In other words, hydration does not accelerate metabolic processing of a variety of miscellaneous molecules.
  33. This a perfect example of why you really need to know your biochemistry and get specific if you’re going to talk about toxins. Myoglobin itself is not toxic, and can circulate more or less harmlessly through your blood. It doesn’t become a problem until it dissolves in acidic urine in the kidneys, because one of its molecular parts poisons the kidneys on its way through.
  34. You would dilute everything else you need in your blood — and end up with water toxicity, which is just as dangerous. When you treat acute rhabdo, you have to flush the kidneys with lots of fluid, but also include a smart cocktail of electrolytes and glucose. Otherwise the cure is just as bad as the disease.
  35. Many of the symptoms of hangover are caused by dehydration. But not alcohol poisoning proper, which can kill but not be dehydration, and you can’t “flush” the alcohol from your system. The only way to prevent the poisoning is to throw up or pump out unabsorbed alcohol.
  36. Dehydration starts causing headaches and nausea at only about 5%, and 15% is where people start dying. So at best, even for a dangerously dehydrated person, drinking a lot of water is only going to boost your blood volume by a few percent: nowhere near enough to meaningfully “dilute” any toxin. For more information about the relationship between hydration and headaches, see my full headache guide: The Complete Guide to Chronic Tension Headaches: A detailed, science-based tour of stubborn headache diagnosis and treatment, for both patients and professionals.
  37. There are limits to how much you can drink and absorb. Cholera provides a colorful example. You literally can’t drink enough water to treat cholera because you actually lose more fluid out of the rectum than you can possibly ingest to replace it. Think about that.
  38. There’s just not that much variation in how much filtering the kidneys can do. They are vital organs, and their work is critical to your survival, and must always proceed at a reasonable pace — even if there’s not enough fluid in the system. As with overall fluid volume, you can’t stray very far out of the normal range without getting into trouble.


linking guide

9,500 words

PainSci Member Login » Submit your email to unlock member content. If you can’t remember/access your registration email, please contact me. ~ Paul Ingraham, PainSci Publisher