Can your spine be “out”? Can it be “adjusted”? Adjusting the spine (spinal manipulative therapy, or SMT) is complex and controversial, largely based on the chiropractic concept of a joint “subluxation.” SMT for the back probably has modest benefits and tolerable risks, while SMT for necks involves less benefit … and the rare but real risks of paralysis and death! Nearly all medical science experts and many chiropractors reject the century-old chiropractic belief that SMT can prevent or cure diseases. This article covers these topics thoroughly — it is one the most detailed online guides that I know of. I particularly discuss the pros and cons of SMT as it relates to the treatment of neck pain, back pain, and muscle pain.
In 2012, a large, credible Cochrane review of SMT science was published.1 Its conclusions were resoundingly negative and disappointing.
Historically, spinal manipulative therapy for acute low back pain has been regarded as the best example of evidence-based care routinely offered by the chiropractic profession. Even many fierce critics of chiropractic have supported this claim (or at least left it alone while focusing on other issues). Nevertheless, some critics have pointed out that even this “best” use of SMT has been damned with faint praise by the research.
This large meta-analysis supports the most critical view: the authors concluded that “SMT is no more effective in participants with acute low-back pain” than shams and placebos. Despite this, the rest of their conclusions seem crafted to prolong the controversy. They explain to clinicians why they should still refer patients for SMT (“preferences” and “costs”); they say that more research is needed to “examine specific subgroups.” (This is based on the rather faint hope that SMT might work so well for an unknown subcategory of patients that they can pull up the average.) And they say there’s a need for “an economic evaluation,” but if a treatment is not effective, it can’t be “cost effective.”
If it were possible to report good news on this topic, it would be here — and I would be happy about it. But it’s not, and the topic can probably be closed. It won’t be closed, of course — the controversy will go on for many years — but like a news network calling election results, there’s a point at which a certain conclusion is virtually inevitable. I believe we are now at that point with SMT.
The idea of “adjusting” the spine refers to many different manual therapies. The joints of the spine may be wiggled, popped, stretched, tapped and more. The correct umbrella term for these treatments is “spinal manipulative therapy” or SMT. Expert opinions on SMT range widely, with some prominent doctors and medical scientists expressing the strongest possible concern and skepticism. Its origins in chiropractic are dubious, its benefits are not major, and there are serious risks, including paralysis and even death.
…there is a “paucity of data related to beneficial effects of chiropractic manipulation of the cervical spine” and a “real potential for catastrophic adverse events.” That’s what the science says.
Controversial it may be, but the science is clear enough to have caused a major American health insurer to stop paying for cervical spine SMT in 2010.2 There has been little high quality scientific research to determine whether or not SMT really “works,” but what little does exist is discouraging indeed. Major reviews of that literature published in recent years came to “underwhelming” and half-baked conclusions. Thus, SMT fails the “impress me” test — it might work, but it can’t possibly work particularly well. An effective therapy should have no problem passing fair tests with flying colours.
And yet spinal joint popping/cracking is a sensation that people crave. I am one of them! And most clinicians — including myself, and including serious skeptics like Dr. Homola — believe that some forms of SMT can be helpful to some of their patients, some of the time. There seems to be almost no doubt that there is something of therapeutic interest going on in SMT. In rare cases involving joint adhesions, mechanical “locking” and loss of mobility, appropriate manipulation has clear value — which is primarily why physical therapists perform spinal manipulation.
It’s a complex picture, and it’s made even more complex by the messy idea of “subluxation.” This is a chiropractic concept of some kind of spinal joint dysfunction, with many shades of meaning — too many! Some chiropractors attribute great importance to subluxation, and “use spinal manipulation to treat visceral disease.”3 The chiropractic concept of subluxation has been both popular and controversial for many decades now, but it has never achieved medical respectability.45 It’s problematic that SMT is often based on such a slippery concept.
Perhaps it has too much baggage to be a useful term.
Even if you put aside all concerns about the quality of the theory, there is still not a shred of scientific evidence that any kind of spinal joint dysfunction — no matter how you define it — has any importance to your general health. In more than a century, nothing like that has ever been shown to be true.6 So do chiropractic subluxations even exist? And, even if they did, would they actually cause any problem, serious or otherwise? And how serious are chiropractors about all this anyway? I’ll address these questions over the next few sections.
My chiropractor says this is because the top of my neck attaches to my head. Is that a common problem?
from the “chiropractors say the darndest things” file, as reported by Dr. Grumpy (Only Outside Sleepy Hollow)
Joint mechanics and neurology are just insanely complex, and when that complex system fails in any way, we could (and should) call it a “joint dysfunction.” Nice, safe, generic terminology. We can cram almost any kind of theory into that label — almost anything that might go wrong with spinal joints — without going too far out on a limb. There is little doubt that spinal pain often involves some kind of “joint dysfunction.”
Or maybe there is doubt: maybe the joint just feels wonky. Pain distorts body image — our mental image of our own anatomy. An odd little 2008 paper demonstrated that people with back pain think their vertebrae are deviated to the painful side.7 Simple as it is, this phenomenon may be potent and persuasive. It could be a major reason for the belief in out-of-place vertebrae.
But if it’s not just warped sensation, an anatomical illusion — if there actually is some kind of dysfunction — what kind of dysfunction? Chiropractors believe that they know, and they’ve built a profession around it: “subluxation.” A misalignment. A “spine out of line.”8 Many doctors and scientists disagree!
Chiropractors often propose subluxation as both the main cause of back and neck pain and of disease and poor health in general.9 The chiropractic idea of subluxation has been defined in many different ways over the years, but most definitions imply some kind of “misalignment” of the spinal joints that needs to be “adjusted” because it’s causing a disproportionate amount of trouble, both pain and poor health. This ominous definition of subluxation is both the most popular and the least defensible. Although there are other definitions — for instance, definitions that might be less controversial — they probably should not be used, because they are too easily confused with the main and original chiropractic definition.
In standard medical terminology, a subluxation simply refers to a partial traumatic dislocation of a joint. Ligaments around the joint may be painfully sprained, and cause severe pain closely coupled to movement, slowly fading over many weeks, like any other sprain. Slightly subluxed spinal joints often return to a more or less normal position immediately, like a dislocated shoulder popping back in.
Although that’s not the kind of “subluxation” that this article is about, such cases do provide important perspective, in that a true dislocation can be amazingly asymptomatic. There are many case reports of true spinal joint derangement and injury that cause little or no trouble, or certainly much less than most people expect. In the low back pain tutorial, I describe one of my own former patients with a very serious lumbar dislocation — and no low back pain (she came to see me for a foot problem). Another patient with extreme scoliosis, although often uncomfortable, was generally surprisingly fine and even fit. The patient pictured here, as reported in New England Journal of Medicine, had no “pain, weakness, or parasthesia”!10
Another case report in the journal Spine describes a man who was born with part of a neck vertebra entirely missing — he also had no serious symptoms until he fell one day, and his unstable spine was dislocated “.11 After that he certainly was symptomatic: he had severe pain, but only pain.
Another fun example of congential deformity or anatomical variation is The Amazing Owl Man! You don’t see this every day. Enjoy this campy, cringe-inducing bit news reel about a dude who can “owl” his neck — rotate 180˚ degrees! — if we can believe our eyes:
What are we to make of this? Is this anything more than an eye-popping novelty? I have no idea how he can do that, but can I extract some clinical relevance from it anyway? Natürlich! (As my high school German language teacher used to say.) Many of these examples of dramatic joint injury and dysfunction actually cause less trouble than chiropractic subluxations Many of these examples of dramatic joint injury and dysfunction actually cause less trouble than much less obvious chiropractic subluxations., never mind the more dire neurological problems that most people would assume to be the case, or the diseases and organ failures that much milder chiropractic subluxations can supposedly cause.
If a painful or stuck-feeling neck or low back involves a “subluxation” in any sense other than partial dislocation, then, it must be something more subtle — something nowhere near so obvious as a traumatic injury — which puts it firmly in the category of the classic chiropractic subluxation.
But is there even any such thing? There is a decades-old argument between doctors and chiropractors about the existence or nature of chiropractic-style subluxations. Many chiropractors still believe in them, in one form or another, and some do not.12 Sam Homola writes:
What I read in books written by orthopedic and physical medicine specialists made more sense to me than what I was being taught in chiropractic college. I rejected the vertebral subluxation theory as a basis for use of spinal manipulation.
Sam Homola, Doctor of Chiropractic, in “Can Chiropractors and Evidence-Based Manual Therapists Work Together?”
As far as most doctors and scientists are concerned, and even for many chiropractors, the argument is long over. They believe that either subluxations do not exist at all, or they only exist in some clinically insignificant form. They argue that chiropractors still can’t prove they can even find alleged subluxations reliably, let alone treat them. They doubt that a spinal joint can be literally “out” in a clinically significant way, and the belief that it can “keeps chiropractic marginalized and subject to ridicule by the scientific community”.13 Edzard Ernst writes, “The concepts of chiropractic are not based on solid science and its therapeutic value has not been demonstrated beyond reasonable doubt.”14
In all the years that they have been talking about them, chiropractors have never been able to furnish proof of these mysterious subluxations which they alone are able to see. They may convince their clients, but never have they provided proof of their pretensions to men of science.
Even if subluxations exist, chiropractic subluxation theory depends heavily on one premise, a conceptual bottleneck through which the entire theory must pass: that spinal nerve roots are terribly important to your health, and interfering with them spells T-R-O-U-B-L-E.
In fact, interfering with the nerve roots can only do so much harm to a person. Most of our health and vitality is regulated by mechanisms other than nerve signals travelling through nerve roots — by other kinds of nerves, and by hormones (see sidebar for details). Common trouble with nerve roots is limited to the garden-variety symptoms like pain, tingling, numbness, weakness, and so on — not organ failure and disease.17 That interference with nerves does not cause a wide variety of health problems is plainly evident in people with spinal injuries: they are paralyzed, not diseased.18 And that is only when the nerve roots are severely damaged or even severed. Slightly pinched nerve roots only spell P-A-I-N and A-N-N-O-Y-A-N-C-E, at worst — no fun, no indeed, but hardly disease.
In any event, it’s also anatomically impossible for spinal joints to physically “pinch” nerve roots with anything less than extreme deformities of the spine.19
Some chiropractors go further with subluxation theory than others, and the further they go, the further out on a scientific limb they go. Most will argue that subluxations are at least responsible for chronic low back pain.
It’s extremely important to appreciate that minor variations in spinal anatomy are normal.20 Vertebrae are not Lego bricks, each one exactly like the other — they are biological and imperfect. Just because there is a slight variation in the lumps and bumps of your spine — something many of my patients worry about — does not mean that something is “subluxed.” And any such variation that is too subtle to be easily agreed upon is also unlikely to be clinically significant — and chiropractors, when tested, routinely cannot agree which joints in a given patient are actually “out.”21 One chiropractor will say it is the L3/4 joint that needs correcting, and the next will name another. This does not inspire confidence in the idea of subluxation.
There may be some truth somewhere in the chiropractic idea of subluxation, but it is clearly loaded and controversial and variable in meaning — too much so to be clinically useful. Instead, I encourage you to think in terms of the broader concept of “joint dysfunction” — that spinal joints can “misbehave,” perhaps in a variety of ways. If this sounds vague, you’re right — but that’s appropriate, because strong confidence in anything more specific than that is not really justified by the science. For instance, an intervertebral joint might move unevenly, as opposed to being out of place, and this could be clinically significant — but no one really knows.
So joint dysfunction almost certainly does not involve partial dislocation, misalignment, and serious interference with nerve root function, as implied by the traditional definition of “subluxation.” But many chiropractors have argued that this is all they really mean by “subluxation” these days anyway …
… we [chiropractors] have successfully distanced the concept of a chiropractic subluxation from that of an orthopedic subluxation.
A Rosner, The role of subluxation in chiropractic, 1997
I’ve heard many chiropractors scoff dismissively that “most chiropractors don’t believe subluxation theory anymore anyway.” Even if they are right, it admits that the profession is still burdened with a faction that does believe. But it’s not that small a faction. In fact, it’s probably a large one, between those who vocally defend the original, untainted theory and those who avoid the term because of its baggage but still believe it quietly, or who reject the most extreme versions but are still deeply influenced by the theory. Sam Homola:
Although many chiropractors are now backing away from the chiropractic vertebral subluxation theory, many continue to use the theory to justify treating a broad scope of health problems by “adjusting” the spine. … Some who do not use the word “subluxation” simply substitute another word or words, such as “joint dysfunction,” in support of their belief that some kind of segmental spinal “lesion” can affect overall health.
Openly defending the original chiropractic subluxation theory are many chiropractors who often call themselves “straight” chiropractors, meaning “pure.” Straights subscribe to the full, traditional “subluxation theory” more or less exactly as formulated more than a hundred years ago, they practice “subluxation-based chiropractic,” and they definitely think of a chiropractic subluxation as being fundamentally different than how a doctor would define subluxation. They believe that subluxations cause not only back and neck pain, but practically any health problem, simply by impinging or irritating spinal nerve roots.
An orthopedic subluxation, a true vertebral misalignment, or a mechanical joint dysfunction that affects mobility in the spine is not the same as a mysterious asymptomatic “chiropractic subluxation” that is alleged to cause disease by interfering with nerve supply to organs.
Chiropractic Vertebral Subluxations: Science vs. Pseudoscience, Homola (ScienceBasedMedicine.org)
The idea of “adjusting” the spine refers to many different manual therapies, usually performed by chiropractors, but also sometimes performed by other professionals.22 The knuckle-like facet joints of the low back may be mobilized (stragetically wiggled, basically23), cracked (like knuckles), tapped with a chiropractic “activator,” gently stretched or pushed on, and so on and on. Different styles may be intense or gentle — so gentle in some cases that you can hardly tell that anything is being done to you. It’s difficult to address them all at once, because there are so many different ways of going about it.
Spinal manipulative therapy is an extremely controversial topic, partly because of its connection to subluxation theory, and partly in its own right — is it safe and does it work? However, SMT for the low back joints is notably much safer and therefore less controversial than SMT for the neck. In the case of the neck, the risks are almost certainly much higher — it’s potentially lethal, in fact.24 I’ll discuss this more below. Here are the main concerns about SMT:
Indeed, not as many people want their spines “adjusted” as you might think. Despite the importance of the topic, only about 5% of the US population uses any kind of alternative therapy for their back pain, and only about 75% of those go to chiropractors.25 (SMT provided by physical therapists may be more common — I don’t have a statistic for that.) Still, even a small slice of the entire population is an awful lot of people.
Sam Homola points out that “there are at least 97 named antithetical adjustive techniques, all based on a nonsensical, nonfalsifiable chiropractic vertebral subluxation theory”26 — almost a hundred variations of SMT that cannot peacefully co-exist, the truth of one invalidating the others, like religions. So it’s hardly surprising that it’s difficult to determine, scientifically, what works and what doesn’t. Proponents of a flavour of SMT have a handy, comfortable defense: “my method is better than the method that was tested.”
Every published review of the literature comments on the lack of good quality evidence, making it impossible to be confident of anything.27 Even the most promising sources of research have had serious quality control problems.28 And yet plenty of that research evidence — weak though it may be — does give some hope that certain kinds of SMT might have some benefits.2930313233
However, most of those conclusions are also qualified by various gotchas and uncertainties. Sometimes even major journals publish results that sound fabulous, but critics find flaws so major that it one wonders if it was even worth doing. There was a particularly noteworthy example early in 2011.34 Neil O’Connell:
The email from the industry was effusive. In a cock-a-hoop, caps lock-happy frenzy it bellowed “ALL MANUAL MEDICINE PROVIDERS SHOULD BE AWARE OF THIS STUDY.”
Or not? Mr. O’Connell readily identified numerous serious problems with this research in an article for Body In Mind:
It is of course possible that the results of this study are accurate and maintenance manipulations are effective, but these problems make it difficult to judge. The message from this one back pain trial might seem appealing and I can see why the email was so enthusiastic. But by focusing on one particular cherry that seems so ripe and juicy we might miss the bigger picture from the rest of the tree. And there is always the chance that the tastiest cherries contain a few artificial sweeteners. Personally I would lay off the caps lock for now.
So much for that hype.
And some studies, of course, are better quality and have had negative results. For example, in 1998, Daniel Cherkin (with medical back pain expert Richard Deyo) published a paper in the New England Journal of Medicine showing that, over the long-term, “patients receiving [chiropractic manipulation] treatments had only marginally better outcomes than those receiving the minimal intervention of an educational booklet.”35 Dr. Nikolai Bogduk weighs in with a paper stating his conclusion right in the title, “Spinal manipulation for neck pain does not work.”36 Similarly, Bronfort ostensibly found a positive result treating neck pain, but concedes that just “a few instructional sessions of home exercise with advice resulted in similar outcomes at most time points.”37
Another of the more recent experiments is a good one to zoom in on, because it damns SMT with faint praise: it is one of the official “positive” studies, but so mildly positive that one can’t help feeling disappointed by it. Is this the best SMT can do?
In the fall of 2009, the Annals of the Rheumatic Diseases published a reasonably good test of SMT.38
The researchers took a hundred patients with nasty, fresh cases of acute low back pain, and delivered half of them into the care of chiropractors, and the other half into “standard care” — advice and ordinary pain medications, namely. Note that it has often been argued that SMT is best for acute low back pain, not chronic, so this is right in chiropractic’s strike zone: if there is anything special, anything even remotely impressive about SMT, it should have done rather well in this contest. It should actually pretty much pull out a can of whupass on “advice and meds.” One would hope.
There was basically no difference between the groups. They did equally well — or equally poorly, if you prefer. All the patients had the same experience that pretty much everyone with chronic back pain has: they got better slowly but surely over the course of a few weeks, roughly exactly as they would have without any therapy at all. The researchers found that “SMT is unlikely to result in relevant early pain reduction in patients with acute low back pain.”Both groups did equally well — or equally poorly, if you prefer.
Now you might say, “Well, good: chiropractic was as good as drugs, so it could replace drugs. And that’s good!” But those drugs, taken in the quantities the subjects took them, are cheap and safe as houses. And their low cost was roughly appropriate for their took-the-edge-off effectiveness. Chiropractic therapy, by contrast, would have cost an order of magnitude more and required spending time on appointments with chiropractors.
So, yes, SMT “worked” — but how well? How impressed are you by SMT’s performance in this test? On a scale of ten?
So it’s a mess. What do reviewers make of it? Better them than me! I’ll take you on a whirlwind tour of some of their older reviews, and then look more closely at the most recent and best.
Going back to 1996, Hurwitz et al wrote that “cervical spine manipulation and mobilization probably provide at least short-term benefits for some patients with neck pain and headaches.”39 Not exactly a glowing endorsement, but others have been even less enthusiastic. Jumping to 2004, Bronfort et al wrote that “There are few studies, and the evidence is currently inconclusive,” and that for chronic neck pain, mobilization/manipulation “offers at most similar pain relief to … rehabilitative exercise in the short and long term,”40 But that review also presented an incredible rats’ nest of conclusions, some of which were positive, some negative, all of which amounted to, “Um, it’s kinda complicated, and we’re not really sure.”41The journal Spine: neck adjustments are “not beneficial”
Does it get any better if we just look at the headache evidence? Worse, actually: even more inadequate and generally discouraging evidence. There were inconclusive scientific reviews of almost no good evidence in 200443 and 2007,44 and no improvement in 2011 in the most recent credible review:45
There are few rigorous RCTs testing the effectiveness of spinal manipulations for treating cervicogenic headaches. The results are mixed and the only trial accounting for placebo effects fails to be positive. Therefore, the therapeutic value of this approach remains uncertain.
For a more detailed analysis of the headache evidence, see Complete Guide to Tension Headaches.
Although the scientific bottom line is (obviously) ambiguous and debatable due to the messiness of the data, things have been getting better. There is usually an exasperating absence of evidence about most manual therapies, but just a few years later the collective data for SMT is finally getting more substantive. And it’s still discouraging.
In 2012, an (updated) Cochrane (good!) review of SMT research was published.46 Cochrane reviews are pretty much the best available summaries of the available evidence,47 and unfortunately the authors just weren’t that impressed. In their first version of the review, they came to the underwhelming conclusion that “there is no evidence that spinal manipulative therapy is superior to other standard treatments for patients with acute or chronic low back pain.” It also performed no worse, though, and was more effective than a fake intervention. A few years later, considering a bunch of new evidence,48 they still found SMT to be “no better or worse” than other therapies … which damned it again with very faint praise, of course, because no therapy has ever been shown to be all that effective for low back pain.49
A fair criticism of this review is that it lumps together too many kinds of SMT for too many different kinds of patients. Could the benefits of some SMT have been “washed out”? What if SMT works really well when done one way, with one kind of patient, but quite poorly otherwise? Real benefits in specific situations might disappear or “wash out” when mixed in with too many other scenarios where a therapy is doomed to failure.
On the other hand, any truly worthwhile benefits to SMT (in any situation) should be obvious and noteworthy. They should pull up the average. I doubt that significant benefits would ever vanish into the statistics. So while I concede that modest real benefits of SMT might get obscured and overshadowed in a big review, I doubt that significant benefits would ever vanish into the statistics.50 Thus we are still left with SMT being damned with faint praise here. As Neil O’Connell of Body in Mind described these results: “a tiny effect size for manipulation that doesn’t really tickle the undercarriage of clinical significance.”
Indeed — and this is the clincher — there is no individual study that I know of that shows any significant benefit. Not one. If any specific flavour of SMT works especially well for any particular type of patient, that combination was not studied in any of the 26 experiments covered by this Cochrane review.
And, just to bang that coffin nail one more time, all this is echoed by another major review from just a couple years before.51
What the mess of science amounts to for patients is that SMT is pretty unimpressive on average, and a bit of a crapshoot … but there are enough flickers of hope here and there that it might be worth a shot for the particularly desperate patient. Chiropractor Sam Homola sums the situation up nicely:
There is no definitive evidence that spinal manipulative therapy is more effective than other forms of treatment for patients with acute or chronic low-back pain. However, manual therapists know from experience that spinal manipulation is often more effective for providing immediate short-term relief for some types of back pain.
This echoes his opinion in his 1999 book, Inside Chiropractic: A patient’s guide, that “many patients have told me that manipulation was more effective [than physiotherapeutic management] in providing immediate relief.”52 My own clients have often reported the same thing. The benefits are almost certainly there, for some people, some of the time.
But, just as clearly, SMT is not working any miracles, and thus it fails the “impress me” test completely. Some evidence may support SMT, but it doesn’t support it strongly. No one — or almost no one — is getting “cured.” After decades of study, the effectiveness of a therapy should be quite clear and significant. If it’s still hopelessly mired in controversy, how good can it possibly be? That’s an awkward question for a lot of alternative therapies.
In a word, no: SMT is definitely not a risk-free treatment. Few treatments are completely safe, of course. Other manual therapies are also unsafe. Half of all manual therapy treatments of any kind will result in some kind of unpleasant side effect.53 Unsurprisingly, neck manipulation can cause trouble — it’s a vulnerable structure. There is a risk of life-threatening injury to the brain stem or vertebral arteries. This is unlikely but extremely serious, and I’ll cover that separately in the next section. This section covers the lesser risks.
In 2010, Carlesso et al analysed the scientific literature looking for evidence of harm from SMT for the neck.54 They found a statistically insignificant trend towards the negative, an uncertain number that leaned in the direction of bad news: increased neck pain might be 25% more likely with SMT than if you did nothing, or if you just stuck to safe and neutral treatments. The same murky data could also suggest basically the opposite: the absence of a clear signal constitutes “strong evidence that neck manipulation or mobilization does not result in an increase in neck pain,” according to the authors. Debatable,55 but noted. What about non-pain symptoms?
Much more certainly and strikingly, the same data does show that SMT patients are 100% more likely to have “transient neurological symptoms” — anything from feeling a bit woozy all the way up to serious unpleasantness, such as severe dizziness, nausea and vomiting. A typical case is described in What Happened To My Barber?.
When performed intensely and/or carelessly — in the presence of undiagnosed vulnerabilities, say — SMT almost can cause direct injury. Even a completely healthy spinal joint could be mildly injured by a strong enough spinal manipulation. Most such injuries would be both rare and minor. Only extremely aggressive SMT could injure a healthy spine, and even if it did, it would probably just be a minor strain — hardly the result you want, but not that big a deal and not that common.
…the chances of injury are multiplied when chiropractic neck manipulation is routinely used many times on every patient in a misguided attempt to improve health, as opposed to occasional use of neck manipulation by physical therapists and other manual therapists who are concerned only about restoring mobility in a stiff spine.
Chiropractic Vertebral Subluxations: Science vs. Pseudoscience, Homola (ScienceBasedMedicine.org)
The risks to an unhealthy spine are also quite rare, but somewhat more serious. Most practitioners are unlikely to use aggressive SMT on a spine exhibiting symptoms of any trauma, disease or other vulnerability. However, mistakes happen — it’s unlikely, but possible.
The main common risk of SMT, however, is psychological: SMT can badly frighten patients, which is a serious risk factor for chronic neck and back pain. This is certainly not a life-threatening risk, but it’s not trivial either. The nervous patient has much to lose with SMT. Patients may be nervous because:
Combining these factors — acute pain, minor potential therapeutic benefits, a high rate of harm, fear of your spine’s fragility, fear of the treatment — is all just a recipe for disaster. And so, although many clients have reported being “cured” by chiropractic treatment (sometimes even as they limp into my office still suffering from significant symptoms), many also have reported being “ruined” by a chiropractic treatment they found to be intense and terrifying and painful.
This unfortunate situation may then be more deeply aggravated by the therapist’s response: using the bad reaction as evidence of profound structural instability in the spine, and as a justification for (much) more treatment. People tend to emerge from such experiences hopelessly tangled up in the idea that they are fragile, broken and need regular SMT for the rest of their lives. It is quite tragic how one bad day can convince people that they have a really “bad back” that needs regular therapy almost forever.
Another surprising risk is that, even if all you have is back pain, chiropractors may recommend SMT for the neck even when your only symptoms are in your low back.57 And, as mentioned above, there may be substantial risks to neck adjustment. And most chiropractors don’t inform their patients about the risks.58 Let’s look at that topic now.People tend to emerge from bad SMT experiences hopelessly tangled up in the idea that they are fragile.
SMT can probably cause stroke in one of two ways, either by tearing delicate blood vessels in the neck, or by causing brain stem injury in persons with unsuspected instability of the upper spine (atlantoaxial instability, AAI). The evidence for this will be discussed as we proceed. These hazards are certainly more likely for patients with severe and chronic symptoms. These are also therefore the patients most likely to desperately try a wide variety of treatment options, including more aggressive and frequent neck adjustments.
Disturbingly, AAI is such a complex condition that even substantial dislocation of the upper cervical spine can be clinically silent, and virtually impossible to diagnose without a specific type of X-ray (a type that chiropractors do not generally do).59 Thus, some people with AAI are walking around without so much as a headache as a warning sign … yet they are vulnerable to severe brain stem injury if they receive cervical SMT.
Less serious but extremely unpleasant consequences are also possible for patients with AAI. A light poke in the brain stem won’t kill you, but it can sure ruin your day! I knew a woman with a confirmed case of post-traumatic AAI who was treated too roughly by a student massage therapist, and she suffered about a day of severe disorientation, vertigo, and vomiting.
That’s what a poke in brain stem can do.
Incredibly, some therapists may dismiss such severe reactions with the absurdly unsafe theory that the body is just processing toxins released during the treatment. It’s difficult to imagine a more daft and irresponsible rationalization for a serious treatment mistake. For an emotionally compelling example of such dangerous incompetence, see the article What Happened To My Barber?
These dangers are, like everything else about SMT, controversial. Proof of the danger is hard to come by. Above we discussed a study by Carlesso et al showing clear evidence that minor to moderate neurological side effects are more than twice as likely with SMT than with other neck treatments. But they did not find “smoking gun” proof that neck adjustment can kill. In fact, they didn’t even try to calculate the relative risk — the data just wasn’t up to the task. That doesn’t mean that the phenomenon doesn’t exist.60
Considering the relatively minor benefits and the high stakes, is any risk acceptable? This has been a subject of fierce debate for many years, with the chiropractic profession consistently on the defensive,61 but even key sources like the infamous Cassidy paper are weak sauce.62 Credible experts continue to publish evidence that concern about this issue is justified. For instance, in recent years the journal Physical Therapy concluded that “the literature does not demonstrate that the benefits of manipulation of the cervical spine outweigh the risks,”63 and the journal Spine concluded that “adverse reactions to chiropractic care for neck pain are common.”64 The value of a treatment must be weighed against the potential harm. As a group of Canadian neurologists ask, “Is a headache worth dying for?” Their concerns are outlined in a detailed document, “Statement of Concern to the Canadian Public from Canadian Neurologists Regarding the Debilitating and Fatal Damage Manipulation of the Neck May Cause to the Nervous System.”65
Even the most dire risks to treatment are not necessarily unacceptable, however. The chances of dying in an accident on the way to the dry cleaner is probably much greater. For the chronic neck pain patient, the risks may be acceptable if there’s a reason to believe the treatment might actually work. Some cases of neck pain and crick are extremely severe — and such patients may literally feel they have little to lose.
Unfortunately, the rationale and scientific evidence to support the use of SMT for neck pain is poor, and some methods of SMT are almost certainly useless. In chiropractic offices with a traditional view (the majority), most of the point of SMT is to treat “subluxation” -- which, as we've discussed, is an idea so under siege that many chiropractors themselves have denounced it. If SMT is largely supposed to treat a problem that is poorly defined and highly controversial at best, and maybe doesn’t exist at all, how good can it possibly be?
Finally, there is the problem of SMT methods that are probably less effective. According to Dr. Homola, responsible chiropractic therapy for back pain should mostly be limited to mobilizing (wiggling) the lumbar intervertebral joints. However, there are numerous other methods.
A good example is the common use of an “activator” — a little tool that gently thumps the skin and muscle over the joint. The rationale for the use of the activator is extremely thin, in my opinion, and there is no credible evidence whatsoever that it actually does anything — it’s virtually unstudied, nor is there much reason to study it — it’s just not a strong idea.
Considering that even standard SMT isn’t working any miracles, you need to be skeptical of obscure variations on it, especially the “subtle” ones that hardly seem to be doing anything.
Joint popping and cracking can feel great. I know. I’m a crack addict, a junkie. Many of my joints pop easily and pleasantly, even if I don’t make the effort to do so. I find it hard to imagine life without joint popping!
I’m hardly alone. I have also observed many clients expressing relief and pleasure in response to incidental spinal “adjustments” — joint pops that occur in the course of doing massage therapy, little explosions as I glide up the spine. Many people seem to feel that a successful spinal adjustment feels like “scratching an itch you can’t reach.” The same feeling can result even without a pop. This feeling needs explaining!
Why does joint popping feel good? If SMT works, why does it work, and does it have anything to do with this relieving sensation?
The pretty obvious honest answer is that no one knows. If we knew how SMT worked, we would also know if it works! Obviously, if there were any clear science on this, SMT wouldn’t be controversial.
Whatever you have been told before, and despite the availability of many explanations on the internet, the nature of joint popping is not well understood.66 It is firmly in that category of trivial mysteries for which there is simply no research funding, and as such it will probably remain unexplained for some time to come. We simply do not know.
What is reasonably certain is that joint pops (and mobilizations) probably constitutes novel sensory input: a little blast of proprioceptive stimulation.67 Since all living systems seem to thrive on sensory input, and generally suffer without it, I speculate that a joint crack or mobilization feels something like getting “unstuck,” roughly the same feeling as finally getting to stretch your legs after getting off a long flight — and that analogy is not intended to trivialize it. The sensation of relief may be extremely strong! (In both the case of a joint pop and the stretching example.) The details of such a physiological effect are definitely not well understood, but the effect probably exists in some form.
The strength of this idea is that it has the potential to explain a lot, but without claiming too much. It’s reasonable and plausible. It can account for the satisfied feelings many people report, but it doesn’t promise the moon, either. Indeed, it might explain both why SMT works, and why it doesn’t work particularly well. For instance, it’s consistent with a widely reported problem with spinal adjustments: the benefits often don’t last long! Soon the “itch” needs to be “scratched” again.68
It also probably explains another common problem with SMT: why treatment results seem to be all over the map. What the nervous system does with novel sensory input certainly depends on a lot of factors. Some people, some of the time, seem to be able to enjoy and “exploit” such sensations and get a therapeutic effect. Other patients seem to largely ignore it — it just doesn’t make much of an impression on them. Still others are afraid of it! For that not-so-minor minority who find joint popping kind of alarming, SMT won’t seem refreshing and relieving, but more like a bitter medicine (at best) or just a bad idea. Such an unpleasant psychological context is probably all the more severe in the case of low back pain, where structural problems are so often feared by patients.
In the best case scenario, could such a simple sensory experience help to actually “cure” severe or chronic neck and back pain? Probably in a few cases, yes.
For instance, under the right conditions, the novel sensory input of SMT might have a significant therapeutic effect on muscle knots (trigger points). Bear in mind what gives trigger points their clinical importance: they often both cause and complicate other problems. Trigger points form in response to a joint that is dysfunctional for some other reason. Joint dysfunction and trigger points probably reinforce each other. Joint manipulation may help to break this vicious cycle, either relieving some of the reason the trigger points are persisting, or relieving the trigger points themselves.
Since trigger points are clinically significant in general, and the likely cause of most back pain, and can often be lessened or relieved by virtually any physical stimulation, it’s not surprising that there is a strong resemblance between the satisfaction of a good rub and the satisfaction of a good crack. A crack is effectively just a kind of “massage” of the joint and the tissues around it, and thus SMT may be about as relieving as anything else (i.e. heat, massage, stretching) that tends to reduce the discomfort of trigger points and stagnant, “grouchy” tissues.
What you almost certainly want to avoid is the worst-of-both-worlds scenario in which you take the risks for the sake of a treating mild neck pain or — yeeks — only as a preventative treatment (as chiropractors often recommend).69 Don’t do that! There is simply no justification for SMT for the neck unless you have a problem that is worth the risk.
Here are some guidelines that may improve your chances of getting benefit from this approach, and minimizing risk. Evidence published by Manual Therapy shows that, if neck manipulation works at all, it may be more likely to work for people with these signs and symptoms.70 Perhaps to some extent this is also true of low back pain. Check all that apply to you:
And here are few more guidelines for safety of my own:
If you discover, or already know, that cervical spine adjustment is helpful for you, make an effort to find a chiropractor you trust, and establish a long term relationship, confident that most of the risks discussed here will be eliminated. “Good science-based chiropractors who do not subscribe to the vertebral subluxation theory and who use manipulation appropriately can offer a service of value,” Sam Homola writes. “Unfortunately, such chiropractors are not easy to find.”
I am a science writer, former massage therapist, and I was the assistant editor at ScienceBasedMedicine.org for several years. I have had my share of injuries and pain challenges as a runner and ultimate player. My wife and I live in downtown Vancouver, Canada. See my full bio and qualifications, or my blog, Writerly. You might run into me on Facebook or Twitter.
Here are some books I recommend:
And some more immediately accessible resources:
Seven updates have been logged for this article since publication (2009). All PainScience.com updates are logged to show a long term commitment to quality, accuracy, and currency. more
When’s the last time you read a blog post and found a list of many changes made to that page since publication? Like good footnotes, this sets PainScience.com apart from other health websites and blogs. Although footnotes are more useful, the update logs are important. They are “fine print,” but more meaningful than most of the comments that most Internet pages waste pixels on.
I log any change to articles that might be of interest to a keen reader. Complete update logging of all noteworthy improvements to all articles 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.
— Minor update to references related to SMT for headache.
— Added reference to body perception, Moseley 2008.
— Added scientific cases studies, examples, pictures and video of true dislocation and abnormal anatomy to help drive home the point that even significant spinal joint dysfunction can be surprisingly harmless … never mind subtle joint problems.
— Added footnote to “Minimizing the risks of SMT”, explaining how much SMT is too much.
— Added reference to article in The Guardian about chiropractor disclosure of risk to patients.
— A good thorough editing and cleanup. There was a bit of a mess in the aftermath of the upgrades a few days ago. 😜
— Major upgrades inspired by a new scientific paper about the dangers of spinal manipulative therapy, and by Dr. Sam Homola’s new article at ScienceBasedMedicine.org, Chiropractic Vertebral Subluxations. I revised and improved content and referencing throughout the article in this update.
In this small, unusual study, six patients with low back pain were asked to draw their perceptions of their back and spine. For comparison, a group of ten patients with no recent back pain were asked to do the same exercises. They were encouraged to “draw what it feels like” rather than how it actually looked.
“All the patients, and none of the controls, showed disrupted body image of the back” on the same side and level as the pain. For instance, they did not have a clear sense of the outline of their trunk.
Most intriguingly, patients with back pain on just one side illustrated vertebrae deviated towards the painful side (without any obvious actual deviation).BACK TO TEXT
A case report (and disturbing X-ray) of a traumatic cervical spine dislocation, notable for being mostly asymptomatic: just torticollis and limited motion, but no pain, weakness or altered sensation. That such a serious injury can have so little impact on a person is quite interesting!
BACK TO TEXT
A 22-year-old man presented with a 10-day history of torticollis. Two months before presentation, he had fallen from standing height but did not report subsequent cervical pain, weakness, or paresthesia. On examination, there was a reduced range of movement of the cervical spine without other neurologic deficits. Radiography of the cervical spine showed a marked reversal of the cervical lordosis, anterior displacement of 60% of C4 on C5, anterior callus formation, bilateral facet dislocation, and increased posterior interspinous distance (double arrow) (Panel A). Magnetic resonance imaging was performed. Sagittal T2-weighted imaging revealed a spinal cord angulation with mild compression but without intramedullary edema or hemorrhage (Panel B). The patient was admitted for surgery. With the use of an anterior approach, diskectomy, tricortical bone grafting, and insertion of a cervical plate were carried out, with a good outcome.
This paper describes the case of a man who developed severe neck and shoulder pain after a fall. It turned out that a part of one of his neck vertebrae was entirely missing, since birth (“a relatively uncommon developmental anomaly”). The fall subluxed the joint substantially. Although painful, the lack of neurological symptoms is surprising — and more evidence that nerve roots are not easily pinched.
BACK TO TEXT
STUDY DESIGN: Case description.
OBJECTIVES: To describe a case of traumatic subluxation in association with a congenitally absent cervical pedicle, and review the pertinent medical literature.
SUMMARY OF BACKGROUND DATA: The congenital absence of a cervical pedicle is a relatively uncommon developmental anomaly that is frequently mistaken for a unilateral facet fracture-dislocation in the context of acute trauma. Because there is little evidence of recovery after surgery, and the symptoms are usually not disabling, surgery is not recommended for most cases.
METHODS: A 62-year-old man presented with severe neck and right shoulder pain after falling. Plain radiographs and computed tomography of the cervical spine showed the typical features of a congenitally absent pedicle at C6 with anterolisthesis of C6 on C7. We performed anterior interbody fusion at C6-C7 because of persistent neck pain and progressive instability.
RESULTS: Complete restoration of the C6-C7 subluxation was achieved with resolution of the presenting symptoms. At 18 months follow-up, flexion and extension dynamic radiographs demonstrated good alignment with solid fusion at C6-C7.
CONCLUSION: Although conservative treatment is the primary treatment for this clinical entity, surgery is an alternative option for those patients who fail to achieve recovery after conservative treatment or exhibit instability.
In a series of informal but devious and persuasive tests, physician Stephen Barrett contrived to challenge the diagnostic skills of a number of chiropractors. The results were inconsistent, and make for some fascinating and disturbing reading.BACK TO TEXT
Dr. Edzard Ernst is a highly qualified critic of sloppy researchers in alternative medicine. In this review of The National Center for Complementary and Alternative Medicine (NCCAM) studies of chiropractic therapy, he finds that “their quality was frequently questionable. Several randomized controlled trials failed to report adverse effects and the majority was not described in sufficient detail to allow replication.” But if NCCAM cannot produce the best quality studies of alternative medicine, who can? No organization has ever been better funded (or motivated) to validate alternative therapies.
Ernst concludes: “It seems questionable whether such research is worthwhile.”BACK TO TEXT
From the abstract, “In the first week following [spinal] manipulation, these patients improved to a greater degree ... and more rapidly ....”BACK TO TEXT
This study supports the usefulness of spinal adjustment for acute pain only. From the abstract: “Patients who received [spinal] manipulative treatment were much more likely to report immediate relief after the first treatment.” However, “at discharge there was no significant difference between the two groups because both showed substantial improvement.”BACK TO TEXT
“There is moderate-to-high quality evidence that immediate clinically important improvements are obtained from a single session of spinal manipulation. The evidence for mobilization is less substantial, with fewer studies reporting smaller immediate changes.”BACK TO TEXT
From the abstract: “Most reviews of these trials indicate that spinal manipulative therapy provides some short-term benefit to patients, especially with acute low back pain.”BACK TO TEXT
This seems to a straightforward “thumbs up” study showing that “traditional bone setting” (chiropractic adjustment, spinal manipulative therapy) has a good effect on chronic neck pain. I admit to being skeptical for no clear reasons (meaning, I see no obvious flaws with the paper). The conclusions seem too strong, too much at odds with a lot of other very mixed evidence on this topic. I can’t reject it outright, but I suspect the risk of researcher bias is high, and a careful reading of the whole paper would probably turn up concerns.BACK TO TEXT
This reasonably well-designed, big, 12-week NCCAM trial of spinal manipulative therapy (SMT) for neck pain concludes with an important disclaimer: although SMT “won” and chiropractors cite this study as evidence that adjustment works, the authors acknowledge that just “a few instructional sessions of home exercise with advice resulted in similar outcomes at most time points.” And so SMT is damned, damned, damned with (extremely) faint praise yet again, as it always is, every time it gets studied: it costs vastly more and performs barely better than sending someone home to do a few simple exercises! Now that hurts.BACK TO TEXT
The complexity of this experiment’s results must be seen to be believed! Here’s a sampling:
There is limited to moderate evidence that SMT is better than physical therapy and home back exercise in both the short and long term. There is limited evidence that SMT is superior to sham SMT in the short term and superior to chemonucleolysis for disc herniation in the short term. However, there is also limited evidence that mobilization is inferior to back exercise after disc herniation surgery. Mix of acute and chronic low back pain: SMT/mobilization provides either similar or better pain outcomes in the short and long term when compared with placebo and with other treatments, such as McKenzie therapy, medical care, management by physical therapists, soft tissue treatment and back school. Acute neck pain: There are few studies, and the evidence is currently inconclusive.
Uh, yeah. That really clears things up! I’m not saying this is a bad review — I’m just saying it’s mired in the poor quality of the body of literature, and the complexity of the problem.BACK TO TEXT
The Cochrane Collaboration publishes comprehensive reviews of what the science can tell us so far about medical treatments, and they promote and facilitate evidence-based medicine. Although not perfect — in recent years their quality control has slipped a little, in my opinion — they still generally produce the most authoritative reviews available of pain and injury science.
Now, if only they weren’t so infamous for never drawing any conclusion other than “more research is needed.”BACK TO TEXT
This is a meticulous, sensible, and readable analysis of the very best studies of back pain treatments that have ever been done: the greatest hits of back pain science. There is a great deal of back pain science to review, but authors Machado, Kamper, Herbert, Maher and McCauley found that shockingly little of it was worth their while: just 34 acceptable studies out of a 1031 candidates, and even among those “trial quality was highly variable.” Their conclusions are derived from only the best sort of scientific experiments: not just the gold-standard of randomized and placebo-controlled tests, but carefully choosing only the “right” kind of placebos (several kinds of placebos were grounds for disqualification, because of their known potential to skew the results). They do a good job of explaining exactly how and why they picked the studies they did, and pre-emptively defending it from a couple common concerns. The results were sad and predictable, robust evidence of absence: “The average effects of treatments … are not much greater those of placebos.”BACK TO TEXT
This review of manual therapies focusses on spinal manipulative therapy and massage therapy for low back and neck pain, with predictably underwhelming results: both are “effective” in some circumstances but certainly not impressively so, and generally no different from other therapies that help a little but haven’t exactly put a dent in the epidemic. For instance, the authors write that SMT is effective but “similar in effect to other commonly used efficacious therapies such as usual care, exercise, or back school.” Um, that’s nice. I’m thrilled for SMT that it can hold it’s own against “back school” — which, of course, is so legendary for curing low back pain!
The conclusions about SMT in particular are silly. The data is consistent with what other major reviews have concluded, most notably the 2012 Cochrane review (Rubinstein et al).BACK TO TEXT
The sound bite in this study is that 20-40% of all manual therapy treatments — massage, chiropractic, physiotherapy — will cause some kind of unpleasantness, side effect or “adverse event” in medicalspeak.
In a word: yikes!
Perspective cuts both ways here. On the one hand, it’s not as bad as it sounds: these “events” are minor and moderate in severity; only 1 or 2 per thousand visits causes a serious problem; and drugs are actually relatively worse. That is, you are modestly more likely to have an “adverse event” if you are given a pill. This just refers to typical side effects, such as ibuprofen’s tendency to cause indigestion.
But when you take a pill, the side effect is usually unrelated to the problem (i.e. it doesn’t make the problem you’re treating worse), you are generally trading those side effects for some pretty clear benefits, and it’s usually cheap. In manual therapy, most adverse events are backfires — that is, you go for a neck adjustment at the chiropractor, and you come out with more neck pain instead of less. Other data shows this is 25% more likely than if you did nothing at all (see Carlesso). And you pay through the nose for this! Manual therapy is much more expensive than most drug therapy.
Manual therapists routinely claim that their services are much safer and more effective than drug therapies. Yet this data pretty clearly shows that the difference is really not great. Depending on how you look at it, drugs are only a little worse in some ways, or maybe a little better in other ways. But no matter how you slice it, 20-40% is a pretty unpleasant rate of harm — especially at $60–120/hour!BACK TO TEXT
This 1996 paper found that “there is no correlation between the measure of hypermobility and the presence of clinical symptoms. Also, the validity of the upper-cervical stability tests is questionable.”BACK TO TEXT
In the end, only 32 studies could be analyzed. So it remains entirely possible that the phenomenon of severe neck injury from SMT is real but rare, and simply didn’t occur in any of those 32 studies, or wasn’t observed and tracked. Similarly, you could analyze dozens of studies of the health effects of hiking, but probably none of them would have data about bear attacks — yet bears do attack hikers! BACK TO TEXT
Like the infamous Cassidy et al paper, this document is often cited by chiropractors when they are trying to allay fears about serious complications of cervical adjustment. Although the athors make an effort to be scientifically sound, they obviously have a conflict of interest, stating outright that they wish to make the case that cervical manipulation is not dangerous. Consider this excerpt from the executive summary: “In addition, as part of our ongoing commitment to giving NCMIC doctors the best defense possible should the need arise, we are providing this information to our network of chiropractic defense attorneys. We expect this latest research will be an important tool for our defense attorneys to use in presenting the most contemporary findings from recent research and to help overcome common biases held by judges and juries.” So, whatever else this document might be, it’s not objective. Note: Allan Terrett originally wrote a monograph for the National Chiropractic Mutual Insurance Company in 2001. The 2005 version is not written by him, but by Triano and Kawchuk “with grateful appreciation” to him.BACK TO TEXT
Since its publication, “the Cassidy paper” has been the defensive citation used by chiropractors to respond to accusations that neck adjustments involve a risk of stroke, and therefore should not be conducted without proven benefit and informed consent. The abstract seems to strongly exonerate chiropractors: “We found no evidence of excess risk of VBA stroke associated chiropractic care compared to primary care.”
“But abstracts are like movie trailers,” as Dr. Mark Crislip wrote. “They give a flavor of the movie, but often leave out many important plot devices and characters. … If you were to read this article in its entirety, you would not be so sanguine about the safety of chiropractic.” He goes on to explain exactly why in one of the earliest popular posts on ScienceBasedMedicine.org, Chiropractic and Stroke: Evaluation of One Paper.BACK TO TEXT
We Canadian neurologists hereby express our strong concern and thereby issue this warning to Canadians. The public must be made aware that the neurological damage that can result subsequent to upper neck manipulation can be debilitating and fatal.
We make the following recommendations for the attention of the Canadian public, the practitioners of manipulation, the medical community, the provincial Ministries of Health and the health care professional regulatory bodies.
Our concerns are significant. Stroke and death due to neck manipulation has been reported in the scientific literature for over 50 years. New deaths, in the past few years, have been reported to the Canadian Stroke Consortium. The Canadian Stroke Consortium recently published a major prospective study. The latest data from the Stroke Consortium indicates that “more than 100 cases of dissection per year are associated with neck manipulation.” The resulting stroke and debilitation from such a large number is very significant.
A recent study by the Institute of Clinical Evaluative Sciences (ICES Ontario) indicates that patients with posterior circulation strokes under the age of 45 are 5 times more likely than controls to have visited a chiropractor within one week of the event.BACK TO TEXT
From the abstract: “The presence of four or more of these predictors increased the probability of success with manipulation to 89%. We concluded that using favourable predictors to identify treatment responders before administering cervical manipulations could significantly increase the probabilities of a successful treatment.”BACK TO TEXT