Is Chiropractic Dangerous? Side Effects and Real Contraindications

The short answer

It is not dangerous, but it is not nothing either, and both halves of that sentence matter.

What you will probably notice: local soreness, tiredness or a mild headache in the hours after an adjustment. Roughly half of patients report something, and in three out of four cases it has gone within 24 hours (Senstad et al., Spine, 1997).

What you almost certainly will not get: a serious event. Published estimates range from 1 per 2 million manipulations to 13 per 10,000 patients (Swait and Finch, Chiropractic & Manual Therapies, 2017). That is an enormous spread, and I explain below why it is that wide rather than quoting whichever end suits me.

And some people should not receive a high-velocity adjustment at all. That list is short and entirely checkable at a first consultation. That is the risk actually worth watching: not the technique, but nobody checking the list before using it.

And if you moved here from the UK, Ireland, the US, Canada or Australia, there is one structural difference worth knowing about — it is not the technique. It is in the screening section below.

First: what does not belong on a treatment table

Some symptoms are not a chiropractor's business. They belong with a doctor or at A&E (urgencias):

• Loss of bladder or bowel control, or numbness in the area that would touch a saddle — perineum, inner thighs, buttocks — especially with weakness in both legs. That is the picture of cauda equina syndrome, a neurological emergency.
• Weakness that is getting worse: your foot drops, you trip, you cannot stand on tiptoe.
• Fever, chills or feeling generally unwell alongside back pain.
• Unexplained weight loss, a history of cancer, or night pain that does not ease when you change position.
• A significant fall or impact, particularly with osteoporosis or if you are over 70.
• Sudden, severe neck or head pain unlike your usual pain, with dizziness, double vision, slurred speech or clumsiness down one side.

Clinical guidance is explicit that specific causes — cancer, infection, trauma, inflammatory disease — must be excluded before back pain is treated as mechanical (NICE NG59). Any competent practitioner refers you on when these appear. If yours does not, find another one.

The common side effects, and how long they last

This is the part most clinic websites leave out, which is strange, because it is the only part almost everyone experiences.

The reference study recorded reactions from 1,058 new patients across 4,712 treatments. Fifty-five per cent reported at least one unpleasant reaction at some point. The most common were local discomfort (53% of reactions), headache (12%), tiredness (11%) and radiating discomfort (10%). Reactions were mild or moderate in 85% of patients, 64% appeared within four hours, and 74% had gone within 24. There were no serious complications (Senstad et al., 1997).

A later meta-analysis of manual therapy put mild-to-moderate transient adverse events at around 41% in cohort studies and 22% in randomised trials, and major adverse events at approximately 0.13% (Carnes et al., Manual Therapy, 2010). The BMJ review of 47 trials describes the same pattern: adverse events mostly musculoskeletal, transient, and mild to moderate in severity (Rubinstein et al., 2019).

In practice: expect something like the ache after a first gym session in the area treated, for 24 to 48 hours. Gentle movement and local heat is usually all it takes — bed is not. What is not normal is soreness clearly worsening past 48 to 72 hours, or any of the warning signs above appearing.

Serious events: why the published numbers vary so much

The most thorough review of risks from spinal manual treatment screened 250 articles and concluded two things: benign adverse events are common, serious ones are rare, with incidence estimates from 1 per 2 million manipulations to 13 per 10,000 patients (Swait and Finch, 2017).

That spread is not sloppiness. The denominators are not comparable — "per manipulation" and "per patient" are different scales, since one patient receives many manipulations. "Serious" is not defined the same way across studies. And there is no mandatory reporting registry, so most of what is known comes from published case reports. Anyone quoting a flat "less than one in a million" is picking one end of the range; the honest version is that they are rare and the precise figure is not known.

One figure that helps put it in proportion: for cauda equina syndrome, a study of 67,220 low back pain patients per group found an incidence of 0.07% after chiropractic manipulation versus 0.11% after a physiotherapy evaluation without manipulation (Trager et al., PLoS One, 2024). The risk appears to come from the underlying problem, not the adjustment.

Neck manipulation and stroke

This is the specific fear behind the question, usually because of a news story. So here is the evidence rather than reassurance.

The most cited population study identified 818 vertebrobasilar strokes in a population of more than 100 million person-years. Among under-45s, cases were about three times more likely than controls to have seen a chiropractor or a family doctor before their stroke; in over-45s there was no association with chiropractic visits. The authors found no excess risk compared with primary care, and explain the association this way: a dissection already under way causes neck pain and headache, and that person seeks help before the stroke happens (Cassidy et al., Spine, 2008). The same group repeated the analysis on 15,523 carotid stroke cases and reached the same conclusion (Cassidy et al., 2017).

An independent meta-analysis did find a small association with cervical artery dissection (OR 1.74, 95% CI 1.26–2.41), but graded the overall quality of evidence as "very low" and concluded there is no convincing evidence of causation (Church et al., Cureus, 2016).

The American Heart Association / American Stroke Association position is the most useful one for you: current biomechanical evidence is insufficient to establish that neck manipulation causes dissection, but patients should be informed of the statistical association before undergoing cervical manipulation (Biller et al., Stroke, 2014). Which is why you are reading it here and not in small print.

So: I cannot tell you the risk is zero, because nobody knows that. I can tell you it is very small, and that if the idea of a neck adjustment unsettles you, low-force and instrument-assisted cervical techniques exist. Asking for them is entirely reasonable.

The real contraindications

This is worth being literal about. The World Health Organization's guidelines on training and safety in chiropractic carry an explicit list of absolute contraindications (WHO, 2005). The ones that come up in a real clinic:

• Active bone or joint infection: osteomyelitis, septic discitis, spinal tuberculosis.
• Malignancy of the spine, spinal cord or meningeal tumour, including bone metastases.
• Acute fracture or vertebral dislocation, or a healed fracture with signs of instability.
• Frank disc herniation with progressive neurological deficit.
• Cauda equina syndrome.
• Instability: atlantoaxial instability, unstable os odontoideum, congenital generalised hypermobility.
• Internal fixation hardware: no bony manipulation of that region, though soft tissue can be worked safely.
• Active inflammatory arthritis with ligamentous laxity, in the region involved.

Note that an absolute contraindication is usually regional. A lumbar metastasis does not rule out working on a stiff neck; it rules out that area and its immediate vicinity.

Then the relative ones, which is where most real people sit. Osteoporosis and demineralised bone are relative contraindications because of the risk of pathological fracture, and the WHO singles out the spine and ribs as particularly vulnerable, with greatest susceptibility after long-term steroid therapy and in post-menopausal women. With severe osteoporosis, high-velocity manual thrust to that area is off the table in practice. The same category covers anticoagulants and bleeding disorders, vertebrobasilar insufficiency or previous stroke (relative-to-absolute for the neck), a diagnosed aneurysm, recent spinal surgery, and spondylolisthesis with progressive slippage.

"Relative" does not mean "never mind". It means the technique changes — mobilisation without thrust, low-force, instrument-assisted — or that region is left alone. Pregnancy is not a contraindication, but it does require adapted positioning and technique; it is covered with other common questions in our chiropractic FAQs.

What proper screening looks like before a first adjustment

The WHO puts it plainly: incidents arising from manipulative therapy are prevented by careful appraisal of the patient's history and examination findings, actively asking about coexisting disease and medication, including long-term steroid use and anticoagulant therapy.

A properly run first visit involves all of this, and you should notice it happening:

1. A full history: medication, surgeries, cancer history, osteoporosis or bone density scans, inflammatory disease, pregnancy, recent falls.
2. Explicit red-flag screening — the list in the second section. Specific questions, not small talk.
3. Physical and neurological examination: mobility, palpation, strength, reflexes and sensation where pain radiates.
4. Referral when warranted: to a doctor or for imaging if something does not add up, before anything is treated.
5. Real informed consent: what will be done, what you will feel over the next 48 hours, and what the risks are — including the statistical association with cervical dissection if the neck is to be treated.
6. A plan with a review point: when it gets reassessed and what happens if you do not improve.

Here is the structural difference mentioned at the top, and it is the most useful thing on this page for you. Chiropractic is not a statutorily regulated healthcare profession in Spain. In the UK, the General Chiropractic Council registers practitioners and the title is legally protected; the US and Australia license comparably. Spain has none of that: no official degree, no professional college, no register to check. So the screening quality you could previously take for granted is not guaranteed by any institution here — only by the person in front of you. Ask where they trained and how long the programme was. The benchmark is a five-year university degree accredited by the European Council on Chiropractic Education. We go into the discipline's limits in chiropractic benefits and considerations.

Conversely: being adjusted with no history and no examination, nobody mentioning any risk at all, or a multi-session package sold before any assessment — those are the warning signs. If you would rather start with an assessment and no commitment to treatment, you can book a first appointment and say so when you book.

Frequently asked questions

How long does soreness after an adjustment last? Between a few hours and 48 hours. In the Senstad study, 64% of reactions appeared within the first four hours and 74% had resolved within 24. If it is clearly worse at 72 hours, get it looked at rather than waiting.

Is neck manipulation dangerous? It is the most debated technique in the field. Population studies find no excess stroke risk compared with seeing a family doctor for the same symptom, and meta-analyses grade the causal evidence as very low quality. Even so, the AHA/ASA recommends that patients be told about the statistical association beforehand, and low-force cervical techniques exist if you would rather avoid a thrust.

I have osteoporosis — can I have chiropractic treatment? It depends on the degree, and your bone density scan needs to be seen first. With severe osteoporosis, high-velocity manual adjustment of the affected area is not done: the work shifts to gentle mobilisation, instrument-assisted techniques and soft tissue, or that region is left alone.

I take anticoagulants. Do I need to mention it? Always — along with steroids, immunosuppressants and any previous surgery. Anticoagulation changes which techniques are appropriate. It is not paperwork; it is the information that decides what is safe for you.

Can an adjustment give me a disc herniation? There is no evidence that manipulation causes herniation in a healthy spine. The reverse is what matters: a herniation with progressive neurological deficit is an absolute contraindication, which is exactly why the examination beforehand is not optional.

When should I go to A&E after an adjustment? If you develop sudden severe head or neck pain, intense dizziness, double vision, difficulty speaking, weakness or clumsiness down one side, or loss of bladder or bowel control. Do not ring the clinic first — go to urgencias and tell them you have had a cervical manipulation.