FCER Response to Canadian Neurological
The Foundation for Chiropractic
Education and Research




Other News Items




Anthony L. Rosner, Ph.D.

December 22, 2000

A recent publication addressing vertebral artery dissection in The Canadian Journal of Neurological Sciences [1] is surprisingly anecdotal and sketchy in its depiction of both the possible causes and etiology of the subject it is intended to discuss. As such, it is laden with severe methodological deficiencies which severely undercut its credibility and create misleading impressions of vertebral artery [VA] dissection and raise more fundamental questions as to how retrospective studies should be conducted. There are at least five critical issues which need to be brought into consideration in order to more fully understand this particular study in a broader perspective.

1. Sampling and time frame issues:

To begin, it is peculiar that this study should implicate sports activity and chiropractic as "prominent precipitating factors" after patients with traumatic arterial dissection have been systematically excluded from consideration. This would necessitate that the presumably vigorous events which have been implicated [chiropractic and sports activity] should have occurred substantially earlier than the time the angiography was conducted. But how much earlier? With so many diverse daily events which have been identified as precipitating factors to VA dissection [see below], the time which has elapsed between the final precipitating event and the onset of symptoms becomes critical. However, there is nothing published pertaining to any intervening time span at all, other than the broad period of 1–14 days reported to elapse between the onset of symptoms and stroke. This casts a substantial cloud over the issue of correctly identifying the precipitating event.

As far as sampling is concerned, it is important to recall that this study is based upon a single site only. Out of the 26 consecutive patients reviewed over an 11–year period at the Alberta University Hospital, about half (14) had "possible links between the onset of dissection and [a] specific predisposing factor." These factors included "sports activities" (4), "chiropractic maneuvers" (3), and "possible neck injuries" (7). This would raise three immediate questions:

  1. What does "possible" refer to as a neck injury?

  2. What inferences can be drawn from almost an equal number of patients (12) not sampled for histories but who had VAs nevertheless?

  3. What was the total number of patients in the Alberta hospital against which the sample of 26 was taken?

Furthermore, the vast majority of patients must have been taken only very recently, since the authors explicitly state in their report that, regarding VA dissections, "more than 60% of the cases studied were diagnosed in the last two years compared to only one patient diagnosed before 1995." The real time frame from which associations are drawn with VA dissections, therefore, is far narrower than the 11 years over which hospital records were consulted.

2. Lack of a control population:

This study bases its conclusions only upon the association of a single observation (presence of vertebral artery dissection in an angiograph) with previous events recalled by the patient. There are no baseline (control) readings to accompany this. One could argue that without a control hospital laboratory finding (e.g., elevated blood urine creatinine or presence of an arterial artery occlusion), the frequencies of possible precipitating events prior to the primary finding (presence of arterial artery dissection) are meaningless. By the reasoning put forth in this study, we are forced to the rather absurd conclusion that patients who recall sports or chiropractic events prior to their yielding elevated urine creatinine (for example) could be used as evidence that these particular activities are associated with the aberrant blood chemistry levels obtained.

3. Incorrect identification of precipitating factors to VA dissection:

Even from the authors' own reasoning, cerebrovascular accidents appear to be a cumulative rather than a traumatic event––as attested by their excluding neck trauma patients from the study. This fact is emphatically driven home in Attachment 1, which indicates that no less than 68 everyday activities have been implicated in disrupting cerebral circulation. [2–4] Among those activities listed, 18 (childbirth, interventions by surgeon or anesthetist during surgery, calisthenics, yoga, overhead work, neck extension during radiography, neck extension for a bleeding nose, turning the head while driving a vehicle, archery, wrestling, emergency resuscitation, star gazing, sleeping position, swimming, rap dancing, fitness exercise, beauty parlor events, and Tai Chi) have actually been associated with vascular accidents but are decidedly non-manipulative. [4]

The risk of fatal stroke following cervical manipulation has been assessed in an exhaustive systematic literature review of many sources to be 3 per 10 million manipulations, [5] or about 0.00025%. [6] The mortality rate from stroke in the general population in 1992–93 was 0.00057%, which raises the possibility that the death rate from stroke in the general population could conceivably be higher than that amongst chiropractic patients. [7]

Given the frequency of significant consequences from cervical manipulations (6 per 10 million manipulations, or 0.0006%), [5] and given the many lifestyle activities shown above to trigger cerebrovascular accidents, it would seem nearly impossible as this study has done to attribute the VA dissections reported at indefinite time periods following chiropractic manipulation to the latter. This association, based on a vague recollection of the patient of events in the past, cannot be counted upon to have definitively identified spinal manipulation as a causative event. Identifying the chiropractor in this association is even more problematical, as will be shown immediately below.

4. Undetermined identification of caregiver:

Did the three cases of VA dissection attributed to chiropractic in the study actually follow manipulation by a licensed chiropractor? There is no validation of this fact in the study as reported. The actual number of iatrogenic complications specifically ascribed to chiropractic has been shown to be significantly overestimated due to the fact that the practitioner actually involved is in many cases a nonchiropractor. Rather, a major portion of these accidents have occurred at the hands of an individual with inadequate professional training but incorrectly represented in the medical literature as a chiropractor. This particular review is alarming in that it suggests that for many years chiropractors have been overrepresented (possibly in a systematic manner) in the literature as having brought on VAs. [8]

Risks are inherent in every medical procedure or lifestyle activity that we encounter. In terms of interventions of the spine, chiropractic has been shown to be many orders of magnitude safer than medication or surgery. Assuming that each patient receives an average of 10 manipulations in treatment, death rates following cervical manipulation calculate to anywhere between 1/100–1/400 the rates seen in the use of NSAIDs for the same condition. [6, 9] Death rates from lumbar spine operations have been reported to be 300 times higher than the rate produced by cerebrovascular accidents in spinal manipulation; [10, 11] for cervical surgeries, recent death rates have been estimated to be 700–fold greater. [10] As Rome has pointed out, [2] risks for "virtually all" medical procedures ranging from the taking of blood samples, [12] use of vitamins, [13] drugs, [13] "natural" medications, [14] and vaccination [15] are routinely accepted by the public as a matter of course.

How risks are interpreted is another matter. The VA rate for chiropractic as described above, while extremely low, does represent a challenge to be improved upon. On the other hand, as Rome points out, [2] such entities mena as (i) patient informed consent, (ii) "low and acceptable rates of complications" stated in a policy by the Australian College of Ophthalmalogists, [16] or (iii) "trading off" risks of surgeries and stroke as stated in a recent study of endarterectomies [17] all attest to the fact that certain levels of risk have been habitually accepted in our society until improvements can be made. Why should chiropractic be singled out as having an unacceptable risk?

In his distinction of specific provider types associated with cerebrovascular accidents, Terrett has identified 34 deaths associated with manipulation over 61 years worldwide. [7] For the sake of comparison, 12,000 deaths per year from unnecessary surgery, 7,000 deaths per year from medication errors in hospitals, about 80,000 deaths per year from nosocomial infections in hospitals, and 106,000 deaths per year from nonerror, adverse effects of medications have been recently reported with regard to conventional medicine. [18–20] These data are presented simply to prevent our losing perspective on the entire issue of risk/benefit ratios raised by the study published in the Canadian journal.

This discussion would not be complete without considering "acceptable" lifestyle risks, which should be common knowledge if we are to evaluate the safety of any healthcare intervention––chiropractic or otherwise. Attachment 2* from the study of Dinman [21] clearly indicates that the risk of death per person per year in many of the activities that we accept as normal and engage in are for the most part many orders of magnitude greater than those seen in serious VA complications following chiropractic manipulation. Once again, we must be skeptical if cervical chiropractic manipulation seems to have been singled out as a particularly conspicuous and noxious threat to our livelihood.

The research published in the Canadian Journal of Neurological Sciences therefore needs to be interpreted with extreme caution. It rightfully begins the process of attempting to clarify the sequence of events leading to stroke and vertebral artery dissections. Until the failure limits of vertebral arteries following various motions and activities are more directly measured (currently a promising area of research being pursued at the University of Calgary [22]), however, efforts to single out chiropractic manipulation as a significant source of vertebral artery dissections and stroke will most likely be conjectural at best, futile at worst.


Attachment 1


[Age not a factor]

Postural head changes

A bleeding nose

Radiographic procedure (vertebral 


  artery angiography)

Archery (bow hunter)

Rap dancing


Reversing a vehicle (see ‘backing up’)

Axial traction

Roller coaster

Backing up a car

Self manipulation ‘clicked on turning’

Beauty parlour

Self manipulation (rapid)

Birth trauma (see also ‘childbirth’)

Sitting in a barber’s chair

Bread dancing (see also rap dancing)

Sit-up exercises


Sliding head-first down a water slide

Childbirth ‘doubtful relationship’

Sleeping positions

Contraceptive pill

Spontaneous rupture of aneurisms


Spontaneous turning of head

Dental procedure

Spontaneous vertebral artery dissection

Diving into shallow water (see ‘falls’)

Star gazing

During surgery

Stooping to pick up a bucket

During x-ray examination

Surgery, neck positioning during 

Emergency resuscitation


Falls (minor)


Falls causing hyperextension

Tai chi

Fitness exercise

Telephone call (cordless)


Traction of cervical spine

‘Golden Gate Bridge’ syndrome

Traction and short wave diathermy

  (sightseeing, San Francisco Bay Bridge)




Hair dressing

Turning one’s head

Hanging out washing

Turning one’s head while driving

Head banging

Under anaesthesia

Motor vehicle accidents

Voluntary movement

Neck callisthenics (Tai chi)

Watching aircraft

Ophthalmological perimetric visual field 



Yawning & vigorous stretching 

Overhead work

  (anterior spinal artery)

Painting ceiling

Yoga (‘Bridge’ or ‘Back push-up’)

Post-operative complications of 

Yoga (rotating head)


Rome PL. Perspective: An overview of comparative considerations of cerebrovascular accidents. Chiropractic Journal of Australia 1999; 23(3): 87-102.


Attachment 2

Voluntary Risk Risk of Death 
per Person
per Year
Smoking: 20 cigarettes/day 1 in 200
Drinking: 1 bottle of wine per day 1 in 13,300
Soccer, football 1 in 25,500
Automobile racing 1 in 1,000
Automobile driving (United Kingdom) 1 in 5,900
Motorcycling 1 in 50
Rock climbing 1 in 7,150
Taking contraceptive pills 1 in 5,000
Power boating 1 in 5,900
Canoeing 1 in 100,000
Horse racing 1 in 740
Amateur boxing 1 in 2 million
Professional boxing 1 in 14,300
Skiing 1 in 430,000
Pregnancy (United Kingdom) 1 in 4,350
Abortion: Legal: <12 wk 1 in 50,000
Abortion: Legal: >14 wk 1 in 5,900

Dinman BD. The reality and acceptance of risk. Journal of the American Medical Association 1980; 244(11): 1226-1228.



1   Bin Saeed A, Shuaib A, Al Sulaiti G, Emery D.
Vertebral Artery Dissection: Warning Symptoms, Clinical Features and Prognosis in 26 Patients
Canadian Journal of Neurological Sciences 2000 (Nov); 27 (4): 292–296

2   Rome PL.
Perspectives: An Overview of Comparative Considerations of Cerebrovascular Accidents
Chiropractic Journal of Australia 1999 (Mar); 29 (3): 87–102

3   Terrett AGL.
Vascular accidents from cervical spine manipulation.
Journal of the Australian Chiropractic Association 1987; 17: 15-24.

4   Terrett AGL.
Vertebral stroke following manipulation.
West Des Moines, IA: National Chiropractic Mutual Insurance Company, 1996.

5   Hurwitz EL, Aker PO, Adams AH, Meeker WC, Shekelle PG.
Manipulation and Mobilization of the Cervical Spine: A Systematic Review of the Literature
Spine (Phila Pa 1976) 1996 (Aug 1); 21 (15): 1746–1760

6   Dabbs V, Lauretti W.
A Risk Assessment of Cervical Manipulation vs. NSAIDs for the Treatment of Neck Pain
Journal of Manipulative and Physiological Therapeutics 1995 (Oct); 18 (8): 530–536.

7   Myler L.
A risk assessment of cervical manipulation vs. NSAIDs for the treatment of neck pain.
Journal of Manipulative and Physiological Therapeutics 1996; 19(5): 357.

8   Terrett AGJ.
Misuse of the Literature by Medical Authors in Discussing Spinal Manipulative Therapy Injury
J Manipulative Physiol Ther 1995 (May); 18 (4): 203–210

9   Gabriel SE, Jaakkimainen L, Bombardier C.
Risk of serious gastrointestinal complications related to the use of nonsteroidal anti-inflammatory drugs: A meta-analysis.
Annals of Internal Medicine 1991; 115: 787-796.

10   Deyo RA, Cherkin DC, Loesser JD, Bigos SJ, Ciol MA.
Morbidity and mortality in association with operations on the lumbar spine.
Journal of Bone and Joint Surgery 1992; 74A: 536-543.

11   Boullet R.
Treatment of sciatica: A comparative survey of the complications of surgical treatment and nucleolysis with chymopapain.
Clinical Orthopedics 1990; 251: 144-152.

12   Horowitz SH.
Peripheral nerve injury and causalgia secondary to routine venipuncture.
Neurology 1994; 44: 962-964.

13   Caswell A [ed].
MIMS Annual, Australian edition, 22nd edition.
St. Leonards, New South Wales: MediMedia Publishing, 1998.

14   Anonymous.
Readers' Q & A.
Australian Medicine 1998; October 5:18.

15   Burgess MA, McIntyre PB, Heath TC.
Rethinking contraindications to vaccination.
Medical Journal of Australia 1998; 168: 476-477.

16   Toy M.-A.
Vision for laser surgery loses its shine--Seeing is believing.
The Age, Melbourne 1998; Nov 7:15.

17   European Carotid Surgery Trialists' Collaborative Group.
Randomized trial of endarterectomy for recently symptomatic carotid stenosis: Final results of the MRC European Carotid Surgery Trial [ECST].
Lancet 1998; 351: 1379-1387.

18   Leape L.
Unnecessary surgery.
Annual Review of Public Health 1992; 13: 363-383.

19   Phillips D, Christenfeld N, Glynn L.
Increase in US medication-error deaths between 1983 and 1993.
Lancet 351: 643-644.

20   Lazarou J, Pomeranz B, Corey P.
Incidence of Adverse Drug Reactions in Hospitalized Patients –
A Meta-analysis of Prospective Studies

Journal of the American Medical Association 1998 (Apr 15); 279 (15): 1200–1205

21   Dinman BD.
The reality and acceptance of risk.
Journal of the American Medical Association 1980; 244 (11): 1226-1228.

22   Symonds B.
Research in progress, described by Herzog W.
Segmental biomechanics. Presentation at the Canadian Consortium for Chiropractic Research Centers Research Agenda Workshop,
Toronto, Ontario, CANADA, October 18, 2000.



[  Home ] [  FCER Store ] [  MemberWeb ] [  Subscribe ] [ News ] [  ResearchCentral ] [  About FCER ] [  Links ]

Copyright © 2001 The Foundation for Chiropractic Education and Research

Privacy Statement