New Large Study Finds No Increased Rick of Cervical Artery Dissection Following Cervical Manipulation
A 2026 study of more than 860,000 matched patients provides important new evidence on one of the most persistent safety questions surrounding cervical spinal manipulation.
For decades, cervical spinal manipulative therapy (SMT) has been discussed as a possible cause of cervical artery dissection (CeAD), a rare condition involving a tear in a carotid or vertebral artery that can lead to stroke. One of the central problems in studying that relationship is that neck pain and headache are not only common reasons patients seek chiropractic care, they can also be early symptoms of an artery dissection that is already occurring.
A large new study published in Chiropractic & Manual Therapies was specifically designed to address that problem.
What did the researchers study?
Researchers used linked U.S. electronic health record and insurance claims data from 2010 through 2025. They identified adults seeking care for a new episode of neck pain or headache and compared two groups:
Patients receiving chiropractor-administered cervical SMT.
Patients receiving outpatient medical care that included a prescription for ibuprofen.
The researchers used propensity-score matching to make the groups as comparable as possible across measured characteristics associated with CeAD. After matching, each group contained 431,492 patients, for a total of 862,984 patients.
That comparison is particularly important. Rather than comparing chiropractic patients with the general population or hospitalized stroke patients, the study compared patients who were seeking care for the same general reason: a new episode of neck pain or headache.
What did they find?
During the 30 days following the initial visit:
| Outcome | Cervical SMT | Ibuprofen/medical care |
| Any cervical artery dissection | 7.2 per 100,000 | 13.4 per 100,000 |
| Carotid artery dissection | 2.8 per 100,000 | 7.0 per 100,000 |
| Vertebral artery dissection | 4.6 per 100,000 | 7.0 per 100,000 |
Overall, there were 31 cervical artery dissections among the 431,492 patients in the SMT group and 58 among the 431,492 patients in the medical-care group. The relative risk for any CeAD was 0.53 (95% CI 0.35–0.83). The difference in vertebral artery dissection alone was not statistically significant.
The study’s central conclusion is straightforward: the researchers found no evidence of an increased risk of cervical artery dissection following chiropractor-administered cervical SMT compared with outpatient medical care for patients presenting with new neck pain or headache.
That is an important finding.
Does this mean manipulation prevents artery dissection?
No.
Although the incidence of CeAD was statistically lower in the SMT group overall, the authors specifically caution against interpreting that difference as evidence that manipulation is protective.
They describe the findings as more consistent with a ”non-inferiority framework than a superiority framework.” In other words, the important finding is the absence of evidence of increased risk, not evidence that manipulation reduces the risk of dissection. The authors believe residual, unmeasured differences between the groups likely contributed to the lower observed incidence.
That distinction matters when discussing this research with patients, other healthcare professionals or the public.
The study supports saying:
This large, matched study found no increased incidence of cervical artery dissection following chiropractic cervical manipulation compared with medical care for patients with new neck pain or headache.
It does not establish that cervical manipulation can never cause a dissection, nor does it prove that manipulation protects against one.
Could clinical decision-making explain some of the lower rate?
This is an especially interesting limitation identified by the researchers.
The study could identify patients who received cervical SMT, but it could not identify patients whom a chiropractor evaluated and then decided not to manipulate because of concerning findings.
The authors specifically acknowledge this possibility, writing that chiropractors may selectively avoid cervical SMT in patients they perceive to be at higher cerebrovascular risk based on clinical findings or family histories that are not adequately captured by diagnostic codes.
The database also did not contain several pieces of information that could influence a clinician’s decision. Researchers could not match patients based on exact symptom duration or sudden onset, and they did not have neurological examination findings or patient-reported measures of symptom severity, disability or headache burden from the initial visit.
Therefore, clinical selection is one possible source of residual confounding recognized by the study itself.
What the study cannot tell us is whether chiropractors actually identified more concerning presentations, referred more patients for medical evaluation, or exercised greater caution than the medical providers in the comparison group. Those questions were not measured. Claims data cannot establish what a clinician considered during an examination or why a particular treatment decision was made.
That would require a different type of study.
The study also helps explain an older controversy
An association between chiropractic visits and subsequent CeAD does not necessarily establish that manipulation caused the dissection.
A dissection can produce neck pain or headache before it is diagnosed. A patient experiencing those symptoms may seek chiropractic care, primary care, urgent care or another form of treatment. If the dissection is diagnosed afterward, the earlier treatment can appear to have preceded the condition when the condition may actually have preceded the treatment.
Researchers refer to this problem as protopathic bias, confounding by indication or reverse causation.
This study attempted to address that problem by requiring both groups to have a new episode of neck pain or headache. The comparison was therefore not simply “manipulation versus no manipulation.” It was essentially chiropractic care versus first-line medical care among patients presenting with similar complaints.
That makes this study particularly useful in evaluating the safety question.
The timing of the dissections is also noteworthy
Across both groups, CeAD diagnoses accumulated most rapidly during approximately the first two weeks after patients sought care and then began to plateau.
The researchers concluded that this pattern is consistent with the hypothesis that adults presenting with new neck pain or headache may have a transiently elevated baseline risk of CeAD regardless of the treatment they receive.They appropriately note that additional research is necessary to clarify that relationship.
For chiropractors, this reinforces an important clinical point: new neck pain or headache deserves an appropriate history and examination before treatment.
The authors identify features that warrant further research for distinguishing CeAD from benign neck pain and headache, including acute or sudden onset, unusual or unfamiliar headache or neck pain, recent trauma or infection, and neurological signs.
Important limitations
This was a retrospective observational study, not a randomized trial, so it cannot establish causation or completely exclude an association between SMT and CeAD. The researchers also found evidence that some unmeasured differences remained between the two patient populations despite extensive matching.
There are additional limitations particularly relevant to chiropractic care. Billing codes cannot tell researchers exactly what manipulation technique was performed, the force or amplitude used, patient positioning, or whether manipulation was manual or instrument-assisted. The researchers also could not independently verify from patient charts that every patient classified as receiving cervical SMT actually received manipulation of the cervical region.
Finally, the study examined cervical artery dissection, not stroke. The authors specifically call for additional research examining stroke and higher-risk patient subgroups.
What should Kansas chiropractors take from this study?
This is reassuring evidence, but it should not change the fundamental responsibility of chiropractors evaluating patients with new neck pain or headache.
The study included nearly 1.55 million SMT visits during the 30-day follow-up period among patients in the manipulation cohort, and researchers found no evidence of a meaningful increase in CeAD risk compared with matched patients receiving first-line medical care.
At the same time, CeAD is rare but potentially serious, and its early symptoms can overlap with the complaints chiropractors evaluate every day.
Those two ideas are not contradictory. In fact, they belong together.
The evidence from this large study does not show an increased risk of CeAD following chiropractor-administered cervical SMT. It also reinforces the importance of careful evaluation of patients presenting with new or unusual neck pain and headache and appropriate referral when the clinical presentation raises concern for vascular pathology.