Research Review by Dr. Michael Haneline©


Sept. 2007

Study Title:

Mild mechanical traumas are possible risk factors for cervical artery dissection


Dittrich R et al.

Publication Information:

Cerebrovascular Diseases, 2007; 23(4): 275-81.


Cervical artery dissection (CAD) is an uncommon condition that involves a separation of the intimal layer from the medial layer in one or more of the vertebral or internal carotid arteries. CAD is more likely to affect younger patients and is one of the most common underlying causes of ischemic stroke in patients under 45 years of age.

Several factors have been suggested as being associated with CAD, including extrinsic factors, such as cervical manipulative therapy (CMT) and recent infections, as well as intrinsic factors, such as subclinical connective tissue disorders and generalized arteriopathy. All of these associations, however, have been determined by retrospective studies, which are subject to a number of biases. Furthermore, the association between CMT and CAD is an ongoing debate in the literature, based primarily on a pair of case-control studies in which the definition of CAD incidents was based on retrospective review of patient charts.

The current study utilized a case-control design, wherein the patients (both cases and controls) were interviewed within 24 hours of hospital admission. The authors thought this methodology, along with the blinding of the investigator as to disease status, were important improvements over previous CAD studies.

Consecutive patients under 60 years of age who presented to the neurology department of a university hospital with an established CAD diagnosis during a 2-year period (2003-2005) were prospectively included as cases. Controls were consecutive stroke patients with an etiology other than CAD. The controls were matched with the CAD patients so the groups would be of similar age and gender.

Interviews were done face-to-face and were carried out by a trained investigator. Basic information was acquired on demographics, clinical characteristics, and coexisting conditions. In addition, the patients were asked about signs, symptoms, and activities that immediately preceded the onset of CAD or stroke.

A standardized questionnaire was designed that queried patients about 7 potential mechanical CAD triggers that might have been present for < 24 hours and < 7 days before symptom onset, as follows:
  1. heavy lifting of > 25 kg or using maximal strength
  2. sexual intercourse
  3. mild direct neck trauma caused by external forces striking directly onto the neck
  4. indirect neck trauma caused by external forces striking other parts of the body and affecting the neck (e.g., mild whiplash injury)
  5. jerky and abnormal movement of the head with abrupt head deviation of >30°
  6. sports activity involving more powerful effort than usual (e.g. > 45 minutes of jogging, weight training, or aerobic exercise)
  7. CMT (an additional interval of < 30 days prior to symptom onset was applied for this factor)
It should be noted that the definition of CMT included manipulation by licensed chiropractors, as well as by unlicensed “chiropractitioners.”

Pertinent results of this study include:
  • 47 CAD patients and 47 matched non-CAD stroke patients were included
  • hypertension, diabetes, smoking, atrial fibrillation, history of stroke, and family history of vascular events were analyzed as risk factors
  • atrial fibrillation and history of stroke were found to be significantly more frequent in the control group
  • neck pain < 7 days before the onset of symptoms was significantly more frequent in CAD patients (p = 0.01)
  • none of the potential mechanical CAD trigger factors were present at a rate that was significantly different between the groups
  • all of the mechanical trigger factors were analyzed cumulatively, scoring the presence of each of the 7 trigger factors as 1 point, which showed a significant difference between the two groups for the < 24 hours time period (p = 0.01), but not for the < 7 days time period (p = 0.35)
  • CMT < 30 days before symptom onset was more frequent in CAD patients (n = 10) than non-CAD patients (n = 5), but the resulting odds ratio of 2.3 (95% CI 0.7-7.2) was not statistically significant (p = 0.16)
  • infection < 7 days prior to symptom onset was more common in CAD patients (n = 18) than non-CAD patients (n = 10) – this association was almost significant (p = 0.07)
  • multivariate analysis (adjusting for sex, diabetes, smoking, previous stroke, and positive family history) revealed that recent infection produced a 3.5-fold higher risk of CAD
  • multivariate analysis actually decreased the strength of the association between CMT and CAD

Conclusions & Practical Application:

The findings of this study point to a very weak association of CMT with CAD, which is in stark contrast with 2 previous case-control studies (Smith et al. and Rothwell et al. – listed below) that reported strong associations. Both of them reported high odds ratios, indicating that CAD patients were much more likely to have had prior CMT.

CMT was found to be much more common in this study’s control group than what was reported by Smith et al. and Rothwell et al. The authors thought this may be because their patients were interviewed within 24 hours of hospital admission, as opposed to the retrospective chart review methodology used in the former studies. In those studies, it was more likely that CAD patients were asked about previous CMT during hospitalization than patients with non-CAD stroke and, to a large extent, it was because of this bias that CMT was reported more often in CAD patients than in non-CAD stroke patients.

Even though the association of CMT and CAD was weak and not statistically significant, the authors still thought CMT might be important in the pathogenesis of CAD. They suggested that another study with a larger sample size could possibly detect this association.

The authors also contrasted their findings that pointed to a link between recent infection and CAD with the work of Grau et al. and Guillon et al., all of which showed a trend toward an association. They thought the consistency of findings between studies supported the influence of a recent infection on the development of CAD.

In conclusion, a 2-factor pathogenesis of CAD was suggested:
  1. an underlying mechanical “vulnerability” of the brain-supplying vessels, together with…
  2. the presence of certain trigger factors inducing CAD
It appears that mechanical stress may be one of these risk factors, possibly in conjunction with a recent infection. Further research is needed to better clarify this issue, but clinicians should keep these factors in mind when assessing neck pain patients.

Additional References:

  1. Grau, A.J., et al., Association of cervical artery dissection with recent infection. Archives of Neurology 1999; 56(7): 851-6.
  2. Guillon, B., et al., Infection and the risk of spontaneous cervical artery dissection: a case-control study. Stroke 2003; 34(7): e79-81.
  3. Rothwell, D.M., S.J. Bondy, and J.I. Williams, Chiropractic manipulation and stroke: a population-based case-control study. Stroke 2001; 32(5): 1054-60.
  4. Smith, W.S., et al., Spinal manipulative therapy is an independent risk factor for vertebral artery dissection. Neurology 2003; 60(9): 1424-8.