Michael T. Haneline, DC; Arthur C. Croft, DC, MS, MPH; Benjamin M. Frishberg, MD
Spine Research Institute of San Diego,
University of California at San Diego School of Medicine,
San Diego, California, USA.
BACKGROUND: To determine the relationship between chiropractic manipulative therapy and internal carotid artery dissection, a MEDLINE literature search was performed for the years 1966 through 2000 using the terms internal carotid dissection. Literature that included information concerning causation of ICAD, as well as all case studies and series, was selected for review.
REVIEW SUMMARY: In reviewing the cases of internal carotid dissection potentially related to CMT, there were many confounding factors, such as connective tissue aberrations, underlying arteriopathy, or coexistent infection, that obscured any obvious cause-and-effect relationship. To date there are only 13 reported cases of ICAD temporally related to CMT. Most ICADs seem to occur spontaneously and progress from local symptoms of headache and neck pain to cortical ischemic signs. Approximately one third of the reported cases were manipulated by practitioners other than chiropractic physicians, and because of the differential risk related to major differences in training and practice between practitioners who manipulate the spine, it would be inappropriate to compare adverse outcomes between practitioner groups.
CONCLUSIONS: The medical literature does not support a clear causal relationship between CMT and ICAD. Reported cases are exceedingly scarce, and none support clear cause and effect.
From the FULL TEXT Article:
The medical literature contends that chiropractic manipulative therapy (CMT) applied to the cervical spine may be a cause of internal carotid artery dissection (ICAD). [1-4] The first report attributing ICAD to cervical CMT was by Beatty in 1977,  who described a patient who was manipulated by a chiropractic physician for the complaint of neck pain. There were no immediate ill effects reported after the manipulation, and symptomatic relief of neck pain resulted. Five days later, the patient awoke with right arm weakness and difficulty in verbal expression attributable to ischemic effects of ICAD. Beatty ascribed the ICAD to the CMT, asserting that because a tear of the arterial intima was found at surgery, there necessarily must have been trauma, and cervical CMT was the only known trauma before the dissection. More recently, investigators have observed that tearing of the intima is common in cases of spontaneous ICAD with no known associated trauma. [5-8] Some state that such tearing is a requisite condition for dissection.  The Beatty commentary is by far the most frequently cited reference by contemporary authors when connecting CMT with ICAD. [1, 9-16]
It has been established that vertebral artery dissection (VAD) can be triggered by CMT, [17-20] based on the anatomic arrangement of the vertebral artery within the cervical vertebrae, the stress within the artery that takes place in the course of neck rotation, as well as the numerous reports of VAD that are in close temporal relationship to cervical CMT. However, there is still some debate concerning the relationship between VAD and CMT, because patients with neck pain may consult a chiropractic physician with an already established dissection.  Nonetheless, the relationship of VAD to CMT is well recognized and accepted by most within the chiropractic community, whereas the connection between ICAD and cervical CMT is not as clear.
As previously mentioned, one can estimate that there are more than 7000 cases of ICAD per year in the United States, and because the primary presentation of a typical ICAD patient is headache and neck pain, it is likely that a proportion of these 7000 patients will consult chiropractic physicians for treatment in the days, weeks, or months before the ultimate ICAD diagnosis. The mere fact that an ICAD develops after CMT does not necessarily create a causal relationship. There may be a causal relationship in cases of head or neck pain, with ensuing ischemic signs that immediately follow or develop within hours after CMT, but because of the widespread uncertainty concerning ICAD etiology, [9, 23, 47] doubt remains.
While discussing cervicocephalic arteries in general, Norris et al  remarked that the hallmark of dissection was sudden and often severe neck or occipital pain. In cases absent some degree of adverse symptoms in relatively close temporal proximity to CMT, there will always be doubt about a connection with ICAD. Patients who have an associated primary arterial disease, which is often present in ICAD patients,  are at risk when subjected to virtually any sort of head or neck motion.
Lansley  wrote a letter that appeared in Lancet in response to a study concerning 4 cases of CMT-attributed VAD. He noted that there could be another explanation of their findings, that the involved patients already had spontaneous dissection, giving rise to symptoms that led them to consult a chiropractor. Mas et al  demonstrated pathologically, concerning VAD, that cervical pain that precedes and motivates chiropractic cervical manipulation may be the first symptom of a hitherto unrecognized spontaneous or traumatic dissection.
None of the cases reviewed for the present study suggested a clear causal relationship between CMT and ICAD. There were other contributing factors in most cases, such as connective tissue aberrations, arteriopathy, multiple artery postmanipulation symptoms that would point to the occurrence of a dissection.
Terrett  addressed the issue of CMT-related ICAD in his 1996 monograph on the subject of vertebrobasilar stroke. After performing an extensive literature search, he was only able to find four such cases in the English and foreign biomedical literature. Only one of the practitioners involved was a chiropractic physician; the others included a medical practitioner, an osteopath, and a barber. With respect to the relationship between cervical manipulation and the occurrence of carotid artery injury, he noted that “the relationship between manual therapy and the occurrence of reported carotid injury is extremely tenuous at best” (p. 67). Dabbs and Lauretti  were unable to locate any literature that provided a plausible anatomic or pathophysiological mechanism for carotid stroke being related to CMT. They concluded that the association between manipulation and carotid stroke was poorly substantiated and speculative.
After investigating the issue of ICAD resulting from CMT, Stapf  commented, “A relationship between chiropractic manipulation and carotid dissection is plausible but unproven. Although we have seen cases of carotid dissection in individuals who have undergone cervical manipulation, it is difficult to prove that the manipulation caused the dissection in any individual case” (p. 332).
The relationship between ICAD and CMT remains tenuous, and a causal relationship is not supported by the literature. Furthermore, the notion that trivial trauma may cause ICAD is conjectural and is based entirely on a few case reports (class III evidence) (Table 3). There is no class I or II evidence available, and class III evidence is not sufficient in assigning causation. The existing studies are limited to case reports and small series that, for the most part, suffer from methodological flaws. It is plausible that persons with underlying arteriopathy, or those who are already in the process of dissection before CMT, may have a triggering effect from manipulation that proceeds to ICAD, but similar triggering could just as likely be the result of normal day-to-day movements.
The mechanism of traumatic ICA injury hypothesized by Stringer,  where the ICA is stretched across the upper cervical vertebrae in response to hyperextension with lateral flexion, involves severe trauma and should be reserved for such cases. Violent long-lever manipulation, commonly used by untrained manipulators, is not taught in any of chiropractic institutions in the United States. This may be the reason nonchiropractic providers have been responsible for a disproportionate share of reported cases of manipulation following ICAD. CMT is a highly specialized procedure that is the focal point of the educational process of chiropractic physicians.
Some would argue that osteopathic physicians are comparably trained in manipulation, yet, according to the Policy Statement on Spinal Manipulation published by the American Chiropractic Association,  most osteopathic schools only offer spinal manipulation on an elective basis. This compares with chiropractic education, which devotes approximately 52% of its 4-year education to diagnosis and manipulation.
CMT encompasses short-lever, low-amplitude thrusts that are directed segmentally, and simultaneous neck extension with rotation are minimized, whereas less-formal cervical manipulation may consist of long-lever thrusts, with contact only applied to the patient’s head. This positioning may cause strain to the associated cervical tissues. CMT cannot be learned from a book or through weekend courses, which may be the reason the per person incidence of manipulation related injuries is higher outside the chiropractic community.
The issue of ICAD following cervical CMT certainly warrants additional investigation, possibly in an attempt to identify an underlying arteriopathy that is suspected or to better define contributing factors. Because the question of causation is so speculative at this time, researchers, as well as clinicians, should use caution in the assignment of causation in these cases.
The exceedingly low risk of a patient developing ICAD following CMT may be ascertained by calculating the approximate number of cervical manipulations provided by chiropractic physicians in America during the time period covered by the reported cases, then dividing that figure into the number of reported ICADs attributed to CMT that were actually manipulated by chiropractic physicians and occurred in the United States. According to the 2000 American Chiropractic Association Statistical Survey,  the average chiropractic physician attends to approximately 115 patient visits per week. Cervical manipulation is performed in about 75% of these cases. The yearly total visits for each practitioner can then be projected to be 4300, based on a 50-week year. The population of practicing chiropractic physicians in America was roughly 20,000 in 1977 and 50,000 in 2000, with estimated even incremental growth in the years between. Using these figures, in conjunction with the previously reported incidence of ICAD in America, an approximation of the chance of developing ICAD following CMT would be 0.0000000017 or 1:601,145,000 (Table 4).
This estimate does not take into account unreported cases of CMT-associated ICAD. In fact, the literature may represent only a fraction of the actual number of cases. Therefore, there may be greater risk than discussed herein.