DISCUSSION
Dissection of the cervical arteries is more common than dissection in other arteries of comparable size. This increased risk in the cervical arteries is believed to be due to their relative mobility and proximity to bony structures.4
Sudden neck movement, a feature of chiropractic treatment, is one of several known risk factors for ‘spontaneous’ cervical artery dissection.8,9 Symptom onset and stroke may be delayed after a spontaneous cervical artery dissection.10 Spontaneous dissection more commonly involves the carotid arteries;4 however, the vertebral arteries appear more prone to dissection as a consequence of chiropractic manipulation,11 likely due to their relation to the cervical spine.
The vertebral artery runs through foramina in the transverse processes of vertebral bodies C1 through C6 (FIGURE 3). On exiting the C2 transverse process, the vertebral artery has a tortuous course, making several turns over and through adjacent bony structures.12 The artery is most prone to injury between the entrance to the transverse foramen of C6 and the foramen magnum (V2 and V3 segments).13 (The area of highest vulnerability is the tortuous segment from the transverse foramen of C2 to the foramen magnum.)
Sudden movements of the cervical spine may cause arterial dissection, whether the maneuvers are performed by a physician, a chiropractor, or a physical therapist.14 Injuries reported in the literature, however, most commonly follow chiropractic manipulation. In our series of 141 dissections, we found no cases associated with manipulation by other health professionals.
A 2003 study revealed cervical spine manipulation to be an independent and strong risk factor for vertebral artery dissection. The authors believed the relationship was likely causal.5 Data from the Canadian Stroke Consortium showed a 28% incidence of chiropractic manipulation in cases of cervical artery dissection.10
A 2008 study showed an association between vertebrobasilar stroke and chiropractic visits within one month of the vascular event.15 However, the study also showed an association of similar magnitude between vertebrobasilar stroke and visits to primary care physicians within the prior month. This suggests that cervical manipulation by chiropractors poses no more risk for cervical artery dissection than visits to primary care physicians. However, it is hard to reconcile such a conclusion with other studies, including our own, in which 10 patients developed new symptoms immediately with chiropractic manipulation of their cervical spines.
Perhaps the one-month observation period of Cassidy et al was excessive. Many post-manipulation events occur within hours or at most a few days, as would be expected given the hypothesized pathogenic mechanism. Perhaps if they had shortened their interval of study to the preceding 3 days, their findings may have been different.
A recent systematic review and meta-analysis demonstrated a slight association between chiropractic neck manipulation and cervical artery dissection. It stated that the quality of the published literature was very low, and it concluded there was no convincing evidence of causation.16 The fact that 10 of the 12 patients in our case series demonstrated acute symptoms immediately upon receiving spinal manipulation suggests a possible causal link; however, we agree with the authors of the meta-analysis that the quality of the literature is low.
A recent statement from the American Heart Association/American Stroke Association (and endorsed by the American Association of Neurological Surgeons and the Congress of Neurological Surgeons) has recommended that chiropractors inform patients of the statistical association between cervical artery dissection and cervical manipulation.17 In addition, it is important for chiropractors to be aware of the signs and symptoms of cervical artery dissection and stroke and to assess for these symptoms before performing neck manipulation, as illustrated in a recent case report.18 Due to the risk of death, patients who experience symptoms consistent with cervical artery dissection after chiropractic manipulation of the cervical spine should be advised to seek medical care immediately.
Our case series has several limitations. The study was retrospective. Existing documentation of associated chiropractic care was often sparse, necessitating phone calls to supplement the information. We believe it is possible that cases may have been missed because of inaccurate medical record documentation, deficits in the interview process concerning chiropractic care at the time of hospitalization, or because information concerning chiropractic care was not recorded in the chart.
A significant portion of our information came through phone contact with several of the patients. In some cases, we relied heavily on their recollection of events that had occurred anytime from a few days to a few years earlier. The accuracy and completeness of the information supplied by patients was not verified, allowing for potential recall bias.
We do not know whether our experience is consistent with that of other areas of the United States. However, the fact that a similar-size hospital in Phoenix reported similar findings suggests the experience may be more widespread.6