Does cervical manipulative therapy cause vertebral artery dissection and stroke?

OBJECTIVE Does cervical manipulative therapy (CMT) cause vertebral arterial dissection (VAD) and subsequent ischemic stroke? What is the best estimate of the incidence of CMT associated with VAD and ischemic stroke? METHODS The questions were addressed with a structured evidence-based clinical neurologic practice review. Participants included neuroscience students, consultant neurologists, clinical epidemiologists, medical librarians, and clinical content experts. A critically appraised topic format was employed, starting with a clinical scenario and structured question. The participant group devised search strategies, located and compiled the best evidence, performed critical appraisals, synthesized the results, summarized the evidence, provided commentary, and declared bottom-line conclusions. RESULTS The search yielded 169 citations, of which 55 were deemed most relevant. From this return, we selected 26 publications of the highest evidence available: 3 case-control studies, 8 prospective and retrospective case series studies, 4 illustrative case reports, 1 survey, 1 systematic review of observational research, 5 reviews, and 4 opinion and expert commentary pieces. Five of the applicable 7 criteria for causation were satisfactorily met and supported weak to moderate strength of evidence for causation between CMT and VAD and associated stroke, especially in young adults. Young vertebrobasilar artery territory stroke patients were 5 times more likely than controls to have had CMT within 1 week of the event date (OR 5.03, 95% CI, 1.32-43.87). No significant associations were found for those > or =45 years of age. The best available estimate of incidence is approximately 1.3 cases of VAD or occlusion attributable to CMT for every 100,000 persons <45 years of age receiving CMT within 1 week of manipulative therapy. CONCLUSIONS Weak to moderately strong evidence exists to support causation between CMT and VAD and associated stroke. Ultimately, the acceptable level of risk associated with a therapeutic intervention like CMT must be balanced against evidence of therapeutic efficacy. Further research, employing prospective cohort study designs, is indicated to uncover both the benefits and the harms associated with CMT.