Introduction: Cervical spinal manipulation therapy (SMT) has been associated with cerebrovascular events, including arterial dissection and stroke. We constructed an evidence map of the published literature and systematically reviewed epidemiological studies on the association between SMT and cerebrovascular events.
Methods: We searched MEDLINE, EMBASE, Cochrane, the ISI database, and reference lists through 7/14 to identify studies with data, evidence, or opinions on stroke or dissection occurring after SMT. Descriptive statistics were used to summarize the study design and publication date of studies. Summary ORs with 95% confidence intervals (CI) were estimated using random effects inverse variance analyses.
Results: Among 311 studies meeting eligibility criteria, most were case reports (105, 34%), editorials/letters (n=81, 26%) and case series (n=51, 16%). Six eligible analytic epidemiologic studies (in 5 publications) were found: 5 case control studies included a median of 53 cases (range 47-818; total 1546) and 100 controls (range 43-3164; total 5671) and one case-crossover study used 818 cases from one of the case control studies. SMT exposure within 30 days of the index date was not statistically significantly associated with stroke/dissection risk (OR = 1.79, 95% CI 0.88-3.64, p=0.11; I2 = 63%, p for heterogeneity = 0.03). Three studies used more reliable case definitions (imaging-confirmed dissection), and found that SMT was associated with a significantly increased risk (OR=4.31, 95% CI 1.32-14.07, p=0.015).
Conclusion: Most studies published on the relation between SMT and cerebrovascular events cannot provide robust information on causal inference, but the few available epidemiologic studies indicate possible harm associated with SMT. More well-powered studies with better case ascertainment are needed to guide inference about SMT safety.