Saturday, April 17, 2010

Science, sometimes - stroke and chiropractic

Louis XIV visiting the  Académie des sciences ...Image via Wikipedia

The Connecticut Board of Chiropractic Examiners ruled on the issue of informed consent.  There was a hearing on a declaratory ruling regarding informed consent and stroke.  The board heard four days of testimony (which are available as videos at CT-N).  Although the board's ruling hasn't yet hit the internet where I can find the exact text, it appears from newspaper reports that they ruled affirming that informed consent is the law in Connecticut but refused to require specific consent about a risk of stroke.  Both the New Haven Register and the Hartford Courant have weighed in on this issue and it appears that science isn't the strong suit for either newspaper.

Both papers believe that there is a risk of stroke from cervical manipulation.  Of course the two groups that advocated before the board for the ruling to require telling prospective chiropractic patients of the risk also believed there is a risk. To me what I have learned being part of this process is that science matters to many people only when it serves their interests. 

Nature of science

Science is an epistemological theory.  Epistemology is the study of how we acquire knowledge.  Science, as we know it, is just one method to acquire knowledge through what's called the scientific method.  Wikipedia's explanation of the scientific method is well done (at least today).  Simply, the method is to observe some phenomenon, create a hypothesis (i.e. a prediction) about that phenomenon, then test that hypothesis by controlled observation, an experiment. 

There are other epistemologies besides science.  For example divine revelation where knowledge comes from a deity. Some have called my profession, chiropractic a cult because there are some in the profession whose epistemology is based upon the work of a guru, typically BJ Palmer.  At one time most of medicine was based upon some guru's dogma.  For example anatomy was controlled by the works of Galen (1st - 2nd century of the common era) who really didn't have a clue what was in the human body - he'd looked at monkeys as human dissection wasn't allowed in his time. It wasn't until a skeptic Andreas Vesalius (16th century) applied the empirical method - noting that the bodies he dissected didn't look the way Galen said that anatomy progressed beyond the dogma of Galen.

One common epistemology is the use of anecdotes.  You know, "I once knew a guy who smoked a carton of cigarettes a day for 70 years and never got lung cancer.  So cigarettes are safe."  OK that's extreme but anecdotes have a tremendous hold upon our thinking and often it is very difficult to let science triumph when anecdote sits in contradiction of the science. Anecdotes which in health care publications are called case reports are very good at hypothesis generation, remember the scientific method is based upon an observation generating a hypothesis.  Case reports can help generate a hypothesis but never can validate the hypothesis.  I've heard it said often amongst skeptics of chiropractic that the pleural of anecdote is anecdotes not evidence.  I'll come back to this quote again!

Autism and Vaccinations

I know it looks like I'm completely switching gears but there is a logic to my digression.  As I have thought a lot about stroke and manipulation and autism and vaccinations I've come to see that they are analogous situations.  Think about this, a family has what appears to be a healthy normal baby.  They are in this blissful state of early parenthood enthralled with the baby who is developing a similar love for them.  Then suddenly the developing relationship on the babies side changes and the child becomes distant.  The parents seek advice and find that their now more withdrawn child appears to have autistic spectrum disorder.  To the parent this can be a devastating event. They want desperately to find a cure and if there isn't a cure (which there doesn't seem to be) they want to blame someone, something. This is only natural.  The parents feel as if someone, something has taken their child away from them.  Then they hear about a supposed link between autism and vaccinations.  Now they have people to blame.  They can blame the vaccine manufactures.  They can blame the government for trying to tell us that vaccines are safe. They can blame their doctors for giving their baby this poison which caused the problem.  To hear the anti-vaccination folks go on about autism and vaccines is to hear the voice of anger and vindictiveness.  They are not malicious people just untrained in the ways of science and prone to letting their emotions have greater sway over how they look at the evidence.

What about the evidence?  Those parents when confronted with the scientific evidence that does not support their belief that autism is caused by the vaccine, reject the evidence.  What is the basis for the rejection, valid critique of the science?  No, usually it is what comedian Stephen Colbert called truthinessTruthiness, which was the Merriam-Webster word of the year in 2006, is defined as: "Truth that comes from the gut, not books," and "The quality of preferring concepts or facts one wishes to be true, rather than concepts or facts known to be true."

Then a whole support industry develops around this belief.  There are organizations whose purpose is to advocate on behalf of the parents and spread the word that autism is definitively caused by vaccines.  This theory is advanced in such a way that reading the web one might believe it was the only one.

I was a participant in a webcast on vaccines for the California Department of Public Health titled "Wading through the Confusion".  As one of my colleagues on the webcast, David G. Amaral, PhD, an autism researcher from the UC Davis M.I.N.D. Institute lamented is that there is inadequate research money available to find the cause of autism because the theory advocated by these groups has so dominated the discussion.
[Any one wishing to know more about this battle should read Paul Offit's book Autism’s False Prophets: Bad Science, Risky Medicine, and the Search for a Cure. The NYTimes article on Offit is enlightening about the lack of civility of public discourse on this issue.] 

Back to Stroke

Now think about ischemic strokes.  Isn't this what happens after a stroke in a younger person.  I mean we sort of think of stroke as a disorder that affects the elderly and it is.  The CDC lists death from stroke as the third most common cause of death behind heart attacks and cancer.  The rate of stroke increases dramatically as one ages. Stroke is a life changing event that affects not only the person who may be disabled, but the entire family and other caregivers as well. Utility analyses show that a major stroke is viewed by more than half of those at risk as being worse than death. (1)

So a younger person, say in their 40s - 50s has a ischemic stroke.  That person and their family are going have the same desire as the family who finds their child has developed autism, find someone to blame.  Well instead of a radicalized group of parents spreading the word that vaccines are the cause for autism there is a radicalized group of patients spreading the words that it must be a visit to a chiropractor. This theory that chiropractic cervical manipulation is the cause of strokes in the young has become so common that many MDs speak as if this is an established fact.  They don't really know what the research says about stroke but couple their distrust of chiropractic (2) which I think is a result of the AMA's illegal boycott(3) and it's easy to see how they would assume the connection between chiropractic care and stroke is valid.  Plus why search the literature to find this out, most people they know will tell them it is true.  It seems scientific evidence doesn't always matter.(4)

What is the epidemiology of these strokes?  They are rare representing 2% of all ischemic strokes but 10 to 25 percent of strokes in the young and middle aged.(5)  These facts are not well known so when someone in who is young or middle aged presents to a hospital with a stroke everyone wants to know why.   Given the bias and bad information then the question comes, "did you see a chiropractor within the past month?"  This only serves to expand the perceived value of this belief.  It is not uncommon to hear someone say that the risk is proven.  Proof is a word I tell my students that they can never use when referring to biomedical research.  There is either evidence that supports or refutes a particular hypothesis.  IF they want PROOF then their career choice should become mathematics not health care. 

Or that there is 80 years of science backing this up.  Well time to go back to that earlier quote: "the pleural of anecdote is anecdotes not evidence."    There isn't an 80 year history of science with respect to the association between cervical manipulation and stroke, there is an 80 year history of anecdotes that appears in the scientific literature.  The real science behind this issue has a much shorter history.  We get a survey of neurologists in 1995 (6) with a very poor response rate and which suffers from referral bias. (7)  Plus a survey doesn't establish cause and effect.  There are many studies of cases that have tried to discover the incidence.  For example this paper by Haldeman et al which estimated that one chiropractor in 48 would during the course of their practice career would be made aware of an arterial dissection following cervical manipulation.  But none of these papers actually evaluated the risk. 

There are only three studies that have evaluated the risk of stroke after manipulation.  And all three agree there is an association.  (9- 11)  The first two found a risk (9, 10) and the third (11) finding the same association provided evidence that explains the first two.  Cassidy et al (11) used the same data sources as Rothwell et al (9) but Cassidy et al study investigated one thing neither Smith (9) nor Rothwell (10) did and that is they determined the background rate of these strokes.  The issue is if the rate of strokes after chiropractic care was greater than the background rate (those that were occurring spontaneously not caused by the manipulation) then we would know what the added risk was due to manipulation. The finding, the rate of strokes in patients under 45 seeking care from their primary care physician was identical to the rate for those who had seen a chiropractor.  Thus seeking chiropractic care produced no extra risk for stroke over and above the background rate.  It appears that patients about to have a stroke from a dissection of a vertebral artery (the kind blamed on chiropractic for 80 years) seem to go to a doctor (a medical doctor or a chiropractic doctor) and the only reason the association between chiropractic and stroke appeared to be strong and not for PCPs is that no one asks the question in the emergency room, "did you see your MD within the past month?"

Some have complained that Cassidy et al was a "statistical study" (see this opinion piece or listen to Dr. Murray Katz's testimony before the Connecticut Board of Chiropractic Examiners)  That's almost comical as a critique if not that it might ring true to people who don't understand biomedical research.  Those same critics never said that Rothwell (9) and Lee (10) were "statistical studies" which they are. 

In the publication BottomLine Health there is an article about stroke risk.  The author Dr. Steven Messe essentially dismisses cervical manipulation as a risk factor based upon Cassidy et al's work.(12) 

The nature of science is that what was known one day may change the next.  This isn't often comfortable to the average person but it is the way it is.As the late Prof. Natalia Romalis-Reytblatt (she was a math professor at UB and UB’s 2006-07 Professor of the Year) wrote:

Mathematical results are either false or true (that is, proven or false) and nothing in between. Mathematics is one of the oldest fields of study and, probably, the only field whose findings are not reversed over time, only accumulated.
Math doesn't reverse it's findings but biomedical science does.  A few years ago I wrote about the risk of stroke from cervical manipulation.  But then the evidence changed.  So in the end this is an argument about looking at the science or ignoring the science and using emotion or choosing to use old science only.  To hear the science you can listen to Dr. David Cassidy testify before the board.  You can decide. 

  1. Goldstein LB, Adams R, Alberts MJ, Appel LJ, Brass LM, Bushnell CD, et al. Primary prevention of ischemic stroke: a guideline from the American Heart Association/American Stroke Association Stroke Council: cosponsored by the Atherosclerotic Peripheral Vascular Disease Interdisciplinary Working Group; Cardiovascular Nursing Council; Clinical Cardiology Council; Nutrition, Physical Activity, and Metabolism Council; and the Quality of Care and Outcomes Research Interdisciplinary Working Group: the American Academy of Neurology affirms the value of this guideline. Stroke. 2006 Jun;37(6):1583-633.  (this paper is available for free here)
  2. Busse JW, Jacobs C, Ngo T, Rodine R, Torrance D, Jim J, et al. Attitudes toward chiropractic: a survey of North American orthopedic surgeons. Spine (Phila Pa 1976). 2009 Dec 1;34(25):2818-25. (Pubmed link)
  3. Getzendanner S. Permanent injunction order against AMA. Jama. 1988;259(1):81-2. (Pubmed link)
  4. Gabbay J, le May A. Evidence based guidelines or collectively constructed "mindlines?" Ethnographic study of knowledge management in primary care. Bmj. 2004 Oct 30;329(7473):1013.(this paper is available for free here)
  5. Schievink WI. Spontaneous dissection of the carotid and vertebral arteries. N Engl J Med. 2001 Mar 22;344(12):898-906. (Pubmed link)
  6. Lee KP, Carlini WG, McCormick GF, Albers GW. Neurologic complications following chiropractic manipulation: a survey of California neurologists. Neurology. 1995;45(6):1213-5. (Pubmed link)
  7. Haldeman S, Carey P, Townsend M, Papadopoulos C. Clinical perceptions of the risk of vertebral artery dissection after cervical manipulation: the effect of referral bias. Spine J. 2002 Sep-Oct;2(5):334-42. (Pubmed link)
  8. Haldeman S, Carey P, Townsend M, Papadopoulos C. Arterial dissections following cervical manipulation: the chiropractic experience. CMAJ. 2001 Oct 2;165(7):905-6. (this paper is available for free here)
  9. Rothwell DM, Bondy SJ, Williams JI. Chiropractic manipulation and stroke: a population-based case-control study. Stroke. 2001;32(5):1054-60. (this paper is available for free here)
  10. Smith WS, Johnston SC, Skalabrin EJ, Weaver M, Azari P, Albers GW, et al. Spinal manipulative therapy is an independent risk factor for vertebral artery dissection. Neurology. 2003 May 13;60(9):1424-8. (Pubmed link)
  11. Cassidy JD, Boyle E, Cote P, He Y, Hogg-Johnson S, Silver FL, et al. Risk of vertebrobasilar stroke and chiropractic care: results of a population-based case-control and case-crossover study. Spine. 2008 Feb 15;33(4 Suppl):S176-83. (Pubmed link)
  12. Messe SR. Stroke Risk. BottomLine Health. 2010;24(3):1-2. 


  1. Excellent article. Thank you for bringing a rational point of view to this debate.

  2. Dr. Michael P. KaneApril 18, 2010 at 9:30 AM

    Stephen- What a great post. Very fair and balanced! Thanks again for all your help.

  3. Dr. Michelle WidmeierApril 18, 2010 at 10:56 AM

    Excellent post! Thank you for all your work.

  4. Stephen,

    Excellent summary! I like the contrast between autism and stroke, it's a clever way to bring home the point. Well done.

    I also smiled at your comment: "IF they want PROOF then their career choice should be mathematics" hehehe

    When I was still at Palmer (90-93) our profession agonized over the possibility that spinal adjusting could "cause" damage to the vertebral artery. In fact, that was a primary reason why I chose to use a technique that did not involve rotation as an component of the adjustment.

    Since then, studies have demonstrated that the force it would take to cause an arterial dissection was an order of magnitude higher than the force generated by diversified (rotational) care. What a relief!

    And now Cassidy has shown that there is no increased risk in seeking chiropractic care versus medical care. But, the "background" risk IS still there.

    So, we still need to maintain vigilance for the signs of possible stroke:

    Did the patient experience the sudden onset of headache/neck/face pain that's different than the patient has had before?

    If so, can the patient: smile, raise both arms, stand steady on both feet with their eyes closed, speak a simple sentence with several vowels that run together, such as “Simple Simon Says”, or stick out their tongue?

    There's a more in-depth review of this topic at the Stroke and Chiropractic Page.

    Thanks again for an excellent post!

  5. Well written Stephen - thank you for balancing the debate.

  6. Frank,

    I think the issue of stroke and manipulation goes through phases. There was no research just anecdotes when I was a student at TCC ('79 - '83). We learned George's Test but it wasn't a must do thing. Then in the late '80s - early '90s it became the standard of care despite a lack of evidence of cause and effect between manipulation and stroke or of the validity of George's Test. The 90s were a period of some fear about cervical manipulation, mostly due to press not science. As science took hold we found that George's Test wasn't valuable (1) and all the chiropractic colleges finally stopped teaching it in 2004.

    So now the issue for both chiropractors and medical doctors alike is to properly evaluate the patient for signs vertebral artery dissection which could lead to a stroke, as you have noted.

    Thanks for the comment.


    1. Thiel H, Rix G. Is it time to stop functional pre-manipulation testing of the cervical spine? Man Ther. 2005 May;10(2):154-8.

  7. Stephen,
    Excellent and succinct review! I also like the analogy with autism. Thank you for reviewing Schievink's study - I did not know that the incidence of ischemic strokes is higher (10-25%) in the young and middle aged.


  8. 'The most scholarly and educational bottom-line discourse I've ever read on the stroke issue! Thanks, Stephen!


  9. Stephen, a well written and interesting connexion with autism. I think it adds to the understanding of why stroke victims and their families can be mislead.

  10. This article should be mandatory reading for every chiropractor in practice and every chiropractic student in training. Thank you Dr. Perle for writing this scholarly manuscript.

    James J. Lehman, DC, MBA, DABCO

  11. Stephen,

    Very well written! Please keep them coming!

    Norman Eng

  12. Thanks Stephen.It is so true that in situations such with the stroke issue or a disk herniation that society dictates that we need to find "blame" even though situations arise such as in a stroke or disc herniation where there is in fact no provoking incident.

    Brian Lecker D.C., C.A.F.C.I.

  13. Paul Hooper, DC, MPH, MSApril 19, 2010 at 10:08 AM

    Stephen; Great article. I'm also of the opinion that science doesn't really matter unless you agree with what it states. Paul Hooper, DC, MPH, MS

  14. Thanks so much Stephen!

    Great article. I think now the responsibility is upon chiropractors and other physicians that treat patients with head and neck pain to be cognizant of the signs of a stroke in progress and properly refer. It is regretful that sometimes those signs aren't always clear.

    Blum CL. Chiropractic & Stroke - What Are Our Responsibilities. Journal of Vertebral Subluxation Research. July 2008: 1-4.

    Charles L. Blum, DC

  15. This article reminded me of the famous review done on anecdotal evidence on the effectiveness of parachutes.

    Their conclusions: As with many interventions intended to prevent ill health, the effectiveness of parachutes has not been subjected to rigorous evaluation by using randomised controlled trials. Advocates of evidence based medicine have criticised the adoption of interventions evaluated by using only observational data. We think that everyone might benefit if the most radical protagonists of evidence based medicine organised and participated in a double blind, randomised, placebo controlled, crossover trial of the parachute.

    Sometimes the science is not all there yet.

    Andre A. Broussard, D.C.

  16. Dr. Broussard,

    Thanks for reminding us that the limits of science are real but when the science is there we can't use the excuse that science has limits to ignore the science.


  17. Dr. Perle,

    Thank you for standing up for our profession.

    Marina Gadaborshev

  18. Stephen,

    This is an extremely well written article. You have throughly expressed yourself in a scientific manner, one that the world of science will understand. This article deserves to be published in all newspapers across the land. What we need is more in the chiropractic world to write like you do and join the remainder of the world in the 21st Century and discharge themselves from the past. Thank you once again, you have made me and other like thinkers in this profession very happy,


  19. Great work Stephen, I needed that!

  20. Thank you for not just writing about chiropractic issues but the issue of truth and science in general.

    For many people in our litigious society it is, in fact, easy to be emotional and not analytical when looking for blame when something goes wrong.

    Your article is even-handed and fair and allows the reader to understand and draw their own conclusions on this issue. Every death by arterial dissection is a person that someone cared about and it is easy to be drawn into a desire to find something (anything) to point a finger at for causation.

    Thank you for caring about science in general and chiropractic in particular.

  21. Great well balanced article. I appreciate the comprehensive reporting on such a controversial topic.

  22. Wayne Hoskins said:
    Thanks for the article Stephen. Very well written, good use of analogies and review of the literature. Keep up the good work.

  23. Dr. Perle,

    Excellent article. I think this article explains it best!


  24. Great Article.

  25. A very well balanced post. I'd like to see what biased skeptics like Hall and Barret would say about this.

    Buddy Touchinsky, D.C.

  26. Dr. Perle,

    Thank you for taking the time to write a fair and balanced article on the risk of a CVA with manual manipulative therapy.

    David Radford

  27. said...
    [quote]A very well balanced post. I'd like to see what biased skeptics like Hall and Barret would say about this.[/quote]

    David Gorski ('Orac' of quackwatch/Hall/Barrett affiliation) opines:

    "We do not have enough evidence to be able to say that neck manipulation is futile and unsafe endeavor under all circumstances."
    A statement of similar accuracy:

    "We do not have enough evidence to be able to say that vaccines do not cause autism under all circumstances."

    I fear that some antivaxer will quote mine me when they see this, but the above is, strictly speaking, a true statement. Does that mean we should accept the claim that vaccines cause autism? Of course not. We have considerable evidence from large, well-designed trials, all of which have failed to find a link between vaccines and autism. If there is a risk, it is incredibly tiny.

    In terms of chiropractic, we have lots of evidence that cracking people’s necks does them no good. We also have pretty good evidence that in some people it carries a not inconsequential risk of stroke. However, as you say, we cannot prove that “under all circumstances” chiropractic manipulation of the neck is of no benefit. However, we can say that, if there are circumstances in which it might benefit, they are pretty darned limited–rather like how we can say that if vaccines do contribute to autism it must be sufficiently rare as not to be detectable in all the large studies done. The difference, of course, is that, even if that tiny risk of autism from vaccines existed, the benefits of vaccines are real and tangible. They are quantifiable. Not so chiropractic manipulation of the neck.


    Perhaps he/they could be invited to participate here.

    It would be ineresting to engage him/them in a "real" evidence-based opposed to his/their "preaching to the choir".

  28. This comment has been removed by a blog administrator.

  29. Great article! I have always been jealous of mathematics ability to remain concrete and just build upon its foundation of knowledge whereas science all too often takes seemingly endless baby steps toward a conclusion only to reverse direction suddenly and embrace the opposite notion. Is coffee good for us? Oh, it's not? How about now? Wait, how about now? Oh, it's good again? Thank goodness.

    Brett L. Kinsler, DC

  30. Chiropractic care has proven beneficial and effective to those who undergo chiropractic therapy.

    Chiropractor Adelaide

  31. Fabulous article, this is the kind of work that needs public attention to help combat the scaremongering that goes on.

  32. "Mathematical results are either false or true (that is, proven or false) and nothing in between."

    Um, the Professor of the Year is unfamiliar with Gödel? The set of provable, let alone proven, statements and the set of true statements are not coextensive.

    As for strokes and chiropractic -- that the chance of suffering a stroke from expending time and money on something pointless is low to nonexistent is no reason to do so.

  33. "I have always been jealous of mathematics ability to remain concrete and just build upon its foundation of knowledge whereas science all too often takes seemingly endless baby steps toward a conclusion only to reverse direction suddenly and embrace the opposite notion."

    This is necessarily true because mathematics is based on logical deduction, whereas science is based on the best inference from the available evidence.

  34. ianam,

    You are correct IF a treatment is pointless no level of risk or expenditure of money would be worthwhile. Your implication, I suspect, is that cervical manipulation is pointless. Unfortunately that belief is not supported by the scientific evidence and to ignore that evidence is just another example of what my blog posting is about. High quality scientific evidence shows that cervical manipulation is amongst a group of effective treatments. Is it a perfect treatment. No but neither is anything else and the alternatives offered by non-DCs don't appear to be any more effective and in fact pose greater known risks.


  35. As long as rumors about the safety of chiropractic exist, we must work to educate the general public. The Foundation for Chiropractic Progress is doing a great job with campaigns by Jerry Rice and others. Every doctor DC needs to post articles proving efficacy of chiropractic and the fallacy of the stroke connection onto their website and work to educate the public, one patient at a time. In time, the public will come to know that chiropractic is the safest and most effective health care system available.

  36. Chiropractic is the most effective and least utilized method of headache neck and back pain treatment available.


  37. Thank you for your comment.

    While it is true that the services provided by chiropractic physicians are underutilized when it comes to the treatment of headache, neck and back pain I disagree with the your remainder of your comments in two important ways.

    1. chiropractic is not a method it is a profession.

    2. I assume that when you say chiropractic is effective that you mean that the services of a doctor of chiropractic. The current state of the evidence is such that one cannot support with good quality evidence the statement that our interventions are "the most effective". Unfortunately, at best, we can say that the evidence is very strong that some of the services offered by chiropractic physicians, most particularly spinal manipulation, is an effective treatment, not the most effective treatment for low back pain, neck pain and headaches.

    I know that this will seem like a semantic argument but when talking about scientific evidence these differences effective vs. most effective are important distinctions.