The association between vertebrobasilar artery system (VBA) stroke and cervical spinal manipulative therapy (C-SMT) is a controversial topic and it evokes strong emotions in some.
Damage to the VBA system usually leads to major disability or death. Vertebrobasilar stroke carries a mortality rate of more than 85%. Because it involves the brainstem and cerebellum, most survivors have multisystem dysfunction, such as quadriplegia or hemiplegia, ataxia, dysphagia, dysarthria, gaze abnormalities, and cranial neuropathies.
VBA cases are rare, which means that more is unknown than is known. The knowledge vacuum magnifies the value attached to anecdotes, which are prone to the logical fallacy of post hoc ergo propter hoc (after this, therefore because of this).
The literature has many case studies documenting VBA stroke following violent neck movements that apply unusual forces to the neck. So the biological plausibility for VBA stroke following forceful neck manipulation is reasonable.
However, from a research point of view the condition’s small incidence means that inevitably the most commonly used method to establish causation is the case control study. Case control studies are usually retrospective and are known for their bias including recall bias.
In a study published today, Kosloff and colleagues have analyzed the largest health insurance data set (both commercial insurance and Medicare Advantage (MA) plans) used to investigate the association between chiropractic visits and stroke.
Data from approximately 5% of the US population, over 39 million persons from 49 of the 50 US states (only North Dakota was excluded) were used. Three years’ worth of data were searched to find cases, which were all patients admitted to an acute care hospital with VBA occlusion and stenosis strokes (chosen by ICD-9 codes).
Four age and gender matched controls were randomly selected. Exposures were encounters with either a chiropractor or a primary care physician (in the US a medical doctor who is typically an internist or family practitioner) prior to the VBA stroke.
The cases included 1,159 VBA strokes in the commercial health plan and 670 in the MA plan. Consistent with previous research, no significant association was found between chiropractic visits and VBA strokes in the older population. However, contrary to other case control studies Kosloff and colleagues also found no association between chiropractic visits and VBA strokes.
The authors acknowledge certain limitations of their study due to the nature of insurance claims data. These data do not code for what specific treatment was rendered or immediate responses to treatment. Thus it is not known if chiropractic manipulation was performed during any office visit and if there was any immediate adverse response. Further, the accuracy of the VBA stroke diagnoses is unknown. Finally as the authors note there is a loss of “contextual information surrounding clinical encounters between chiropractors and Primary Care Physicians and their patients.” This limits the knowledge of other known risk factors.
The authors’ conclusions are correctly reserved given the limitations. The dataset does not find a significant association between chiropractic manipulation and VBA stroke, thus adding weight to the view that chiropractic care is an unlikely cause. But this finding does not exclude the possibility that chiropractic manipulation might have some role in causation.