There is longstanding controversy over the rare association between cervical manipulation and stroke related to vertebral artery dissection (VAD). In light of the evidence, there really should not be a controversy at all. This article walks through the research arc - from early case reports to population-based case-control studies, biomechanical testing, and a systematic review - and explains why the sound conclusion is reverse causation, not cause and effect.
It was once assumed that manipulation could, in rare cases, cause a dissection that led to stroke. At the time, that assumption was understandable, but the evidence accumulated over the past two decades tells a different and more coherent story. (A note on terminology: this article concerns VAD and the VAD-related stroke it can cause; the broader umbrella term used in some of the literature is “cervical artery dissection,” or CAD.)
- Vertebral artery dissection (VAD)
- A tear in the inner wall of a vertebral artery that allows blood to enter the artery wall. It commonly produces neck pain and headache first, and can lead to VAD-related stroke. In the broader literature it falls under the umbrella term "cervical artery dissection" (CAD).
The Research Arc: From Assumption to Evidence
The belief that manipulation causes VAD-related stroke did not emerge from strong evidence. It emerged from a particular kind of weak evidence (case reports), and stronger study designs have since corrected the picture.
Step 1: Case Reports
The concern began with a series of case reports - individual accounts of patients who had cervical manipulation and then experienced a VAD-related stroke. Case reports are valuable for generating hypotheses, but they are the weakest form of clinical evidence. They describe a sequence of events; they cannot establish cause and effect. A patient who has a stroke shortly after any event - a hair wash at a salon, a visit to the dentist, a trip to a practitioner - will naturally assume that event was the cause. However, sequence is not necessarily causation.
Step 2: Case-Control Studies
To move beyond anecdote, researchers turned to case-control studies, which examine large groups of people who had a VAD-related stroke alongside a control group who did not, and then compare exposures such as visits to a chiropractor versus visits to a primary care practitioner (GP).
The landmark study by Cassidy and colleagues examined a population-based dataset and found that VAD-related stroke was associated with primary care visits to the same or greater degree as with chiropractic visits.[1] If manipulation were the cause, that pattern would make no sense - primary care practitioners do not perform cervical manipulation. The pattern only makes sense if patients were already developing a dissection, which produced the neck pain and headache that sent them to a practitioner in the first place.
Kosloff and colleagues studied roughly 39 million covered lives in U.S. commercial and Medicare Advantage populations and found no significant association between chiropractic care and VAD-related stroke, while there was a significant association with primary care visits.[2] Critically, among individuals who had a VAD-related stroke after seeing a chiropractor, manipulation was not documented in roughly one-third to one-half of the cases. In other words, a substantial share of the patients who “saw a chiropractor and then had a stroke” never received the treatment that is often blamed.
Whedon and colleagues added further weight, following a cohort of 1,157,475 Medicare beneficiaries aged 66 to 99 with neck pain and concluding that manipulation is unlikely to be a cause of stroke in older adults.[3]
Step 3: Biomechanical Studies
If manipulation caused dissection, we would expect it to place unusual strain on the vertebral arteries. Biomechanical studies show the opposite. Herzog and colleagues measured vertebral artery strains during high-velocity, low-amplitude cervical manipulation and found them to be far below the strains produced by normal range-of-motion movements and diagnostic positioning, and far below the point at which the artery fails.[4] Put plainly, the ordinary movements of daily life stress the vertebral arteries more than manipulation does.
The point is reinforced by Wynd and colleagues, who applied pre-existing vascular lesions to vertebral arteries and then performed cervical manipulation. The manipulation did not further injure the already-compromised arteries.[5] Even a vulnerable artery was not damaged by the treatment.
Step 4: Systematic Review
Church and colleagues brought the body of evidence together in a systematic review and meta-analysis of chiropractic care and cervical artery dissection.[6] They found only a small association (odds ratio of 1.74) - and, tellingly, one that mirrors the association seen with ordinary primary care visits. Their conclusion was that there is no convincing evidence that manipulation causes VAD. A small association that appears equally for a visit to a chiropractor and for a visit to a primary care practitioner is the signature of reverse causation - manipulation does not cause VAD, but development of VAD leads to a visit to a chiropractor or a primary care practitioner. And sometimes stroke occurs at some point after that.
Sequence is not necessarily causation.
What Actually Happens: Reverse Causation
Put the four lines of evidence together and a consistent explanation emerges. In patients who experience a VAD-related stroke following manipulation, it is not the treatment that causes the dissection. Rather, the dissection is already underway.
Neck pain, with or without headache, is commonly the initial symptom of a vertebral artery dissection. A patient who develops this pain does the natural thing: seeks relief. Often that means seeing a spine practitioner, and it commonly also means seeing other professionals, such as a primary care practitioner. The stroke may or may not follow the visit, and when it does, it occurs independent of what the professional does or does not do.
This is what researchers mean by reverse causation - sometimes called protopathic bias. The neck pain is the first symptom of the dissection, not something caused by the treatment. The visit follows the developing dissection; it does not create it. It is the reason case-control studies find the same or greater “risk” attached to primary care visits, and the reason the association survives even when no manipulation was performed.
What This Means for Medicolegal Practice
The implications are direct.
Sequence is being mistaken for causation. The common assumption, when manipulation comes first and VAD-related stroke comes second, is that the manipulation caused the dissection and stroke. That is a case report in disguise, and case reports cannot establish cause and effect. The stronger study designs that can address causation indicate the opposite.
The comparison group matters. The most powerful single fact is that VAD-related stroke is just as likely, or more likely, after a primary care visit than after a chiropractic visit. Since primary care practitioners rarely, if ever, apply manipulation, the association cannot be the manipulation. It is the dissection that brought the patient to care.
Missing manipulation undermines the theory. In roughly one-third to one-half of the chiropractic cases in the Kosloff data, no manipulation was documented at all. A theory that blames a treatment the patient never received is not a sound theory.
Biomechanics closes the loop. Manipulation strains the vertebral arteries less than everyday movement does, and does not injure even a pre-lesioned artery. There is no plausible mechanism by which manipulation applied by a trained professional ruptures a healthy artery.
The Primary Spine Practitioner’s Role
For clinicians, the practical lesson is not fear of manipulation but skill in recognizing the patient who is already dissecting, which is often impossible but at times there are clues. Neck pain and headache of sudden or unusual character deserve careful evaluation. This kind of clinical reasoning is the foundation of the Clinical Reasoning in Spine Pain framework and of the Primary Spine Practitioner model.
Dr. Murphy’s medicolegal services apply this evidence directly to case review and expert testimony, helping attorneys and health systems separate what the sequence of events suggests from what the evidence actually supports.
Frequently Asked Questions
Does cervical manipulation cause vertebral artery dissection or stroke?
No. In light of the evidence there really should not be a controversy. Large case-control studies, biomechanical research, and a systematic review consistently show that any association between manipulation and vertebral artery dissection (VAD) related stroke is explained by reverse causation - the dissection was already underway before the patient sought care - rather than by the treatment itself.
Why do some patients have a stroke after seeing a chiropractor?
Neck pain, often with headache, is commonly the initial symptom of a vertebral artery dissection that is already in progress. A patient with these symptoms seeks relief from a spine practitioner, and commonly also sees a primary care practitioner. The stroke may then occur regardless of what any of those professionals does or does not do. This is why case-control studies find VAD-related stroke is just as likely, or more likely, after seeing a primary care practitioner as after seeing a manipulation practitioner.
What do biomechanical studies show about manipulation and the vertebral arteries?
Biomechanical studies show that the strain placed on the vertebral arteries during cervical manipulation is less than the strain produced by common everyday neck movements and diagnostic positioning, and well below the point at which the artery fails. In one study, vertebral arteries with lesions applied prior to manipulation were not further injured by the treatment.
What is reverse causation in the context of manipulation and stroke?
Reverse causation - sometimes called protopathic bias - means the neck pain is the first symptom of the dissection, not something caused by the treatment. Because dissection produces neck and head pain before it produces a stroke, patients seek care for that pain. The visit to the practitioner therefore follows the developing dissection; it does not cause it.
How should this evidence be handled in a medicolegal case?
The evidence does not support the position that cervical manipulation caused a vertebral artery dissection or stroke. Case-control studies show the same or greater association with primary care visits, biomechanical studies show manipulation strains are far below injury thresholds, and a systematic review found no evidence for causation.
References
-
Cassidy JD, Boyle E, Côté P, He Y, Hogg-Johnson S, Silver FL, Bondy SJ. Risk of vertebrobasilar stroke and chiropractic care: results of a population-based case-control and case-crossover study. Spine (Phila Pa 1976). 2008;33(4 Suppl):S176-S183. PMC ↩
-
Kosloff TM, Elton D, Tao J, Bannister WM. Chiropractic care and the risk of vertebrobasilar stroke: results of a case-control study in U.S. commercial and Medicare Advantage populations. Chiropr Man Therap. 2015;23:19. PMC ↩
-
Whedon JM, Song Y, Mackenzie TA, Phillips RB, Lukovits TG, Lurie JD. Risk of stroke after chiropractic spinal manipulation in Medicare B beneficiaries aged 66 to 99 years with neck pain. J Manipulative Physiol Ther. 2015;38(2):93-101. View source ↩
-
Herzog W, Leonard TR, Symons B, Tang C, Wuest S. Vertebral artery strains during high-speed, low amplitude cervical spinal manipulation. J Electromyogr Kinesiol. 2012;22(5):740-746. PubMed ↩
-
Wynd S, Anderson T, Kawchuk G. Effect of cervical spine manipulation on a pre-existing vascular lesion within the canine vertebral artery. Cerebrovasc Dis. 2008;26(3):304-309. View source ↩
-
Church EW, Sieg EP, Zalatimo O, Hussain NS, Glantz M, Harbaugh RE. Systematic review and meta-analysis of chiropractic care and cervical artery dissection: no evidence for causation. Cureus. 2016;8(2):e498. PMC ↩