Get access

Botulinum toxin for subacute/chronic neck pain

  • Review
  • Intervention




Neck disorders are common, disabling and costly. Botulinum toxin (BoNT) intramuscular injections are often used with the intention of treating neck pain.


To systematically evaluate the literature on the treatment effectiveness of BoNT for neck pain, disability, global perceived effect and quality of life in adults with neck pain with or without associated cervicogenic headache, but excluding cervical radiculopathy and whiplash associated disorder.

Search methods

We searched CENTRAL, MEDLINE, AMED, Index to Chiropractic Literature, CINAHL, LILACS, and EMBASE from their origin to 20 September 2010.

Selection criteria

We included randomised and quasi-randomised controlled trials in which BoNT injections were used to treat subacute or chronic neck pain.

Data collection and analysis

A minimum of two review authors independently selected articles, abstracted data, and assessed risk of bias, using the Cochrane Back Review Group criteria. In the absence of clinical heterogeneity, we calculated standardized mean differences (SMD) and relative risks, and performed meta-analyses using a random-effects model. The quality of the evidence and the strength of recommendations were assigned an overall grade for each outcome.

Main results

We included nine trials (503 participants). Only BoNT type A (BoNT-A) was used in these studies.

High quality evidence suggests there was little or no difference in pain between BoNT-A and saline injections at four weeks (five trials; 252 participants; SMD pooled -0.07 (95% confidence intervals (CI) -0.36 to 0.21)) and six months for chronic neck pain. Very low quality evidence indicated little or no difference in pain between BoNT-A combined with physiotherapeutic exercise and analgesics and saline injection with physiotherapeutic exercise and analgesics for patients with chronic neck pain at four weeks (two trials; 95 participants; SMD pooled 0.09 (95% CI -0.55 to 0.73)) and six months (one trial; 24 participants; SMD -0.56 (95% CI -1.39 to 0.27)). Very low quality evidence from one trial (32 participants) showed little or no difference between BoNT-A and placebo at four weeks (SMD 0.16 (95% CI -0.53 to 0.86)) and six months (SMD 0.00 (95% CI -0.69 to 0.69)) for chronic cervicogenic headache. Very low quality evidence from one trial (31 participants), showed a difference in global perceived effect favouring BoNT-A in chronic neck pain at four weeks (SMD -1.12 (95% CI: -1.89 to -0.36)).

Authors' conclusions

Current evidence fails to confirm either a clinically important or a statistically significant benefit of BoNT-A injection for chronic neck pain associated with or without associated cervicogenic headache. Likewise, there was no benefit seen for disability and quality of life at four week and six months.

Plain language summary

Botulinum toxin for subacute and chronic neck pain

Neck disorders are common, disabling and costly. They are often accompanied by increased muscle tension.

Botulinum toxin (BoNT) is a drug that can temporarily stop muscles from contracting, thereby reducing muscle tightness or spasm. It is given as one or a series of injections. Botulinum toxin is commercially distributed as 'Botox', 'Dysport', 'Xeomin', 'Myobloc', or 'Neurobloc'. Following an injection of a very small volume of BoNT, patients may experience muscle tenderness or pain, weakness, or a general sense of being unwell. Effects begin within three to eight days and may last up to four months; treatment can be re-administered. Treatments using BoNT can be highly variable, and it is not certain that it is superior to placebo.  

This review included nine trials (503 participants) that examined the effects of Botulinum toxin Type-A (BoNT-A) for patients with neck pain that had lasted for at least four to six weeks. In some instances, they also had pain that extended into the head, called a 'cervicogenic headache'.

High quality evidence was available from five trials (252 participants) that found no significant difference in pain at four weeks, between those who received BoNT-A for subacute and chronic neck pain and those who received placebo injections. Low quality evidence showed that BoNT-A injections were no better than placebo for neck pain at six months (one trial; 24 participants) or for disability (one trial; 45 participants). Very low quality evidence from one study (31 participants) suggested that BoNT-A was better than placebo for patients with chronic neck pain when patients reported on their overall sense of benefit at four weeks.

Evidence from a single small study in patients with neck pain and accompanying headache (32 participants) showed that BoNT-A was no better than placebo at four weeks and six months after treatment.

There was very low quality evidence (two trials, 95 participants) showing that BoNT-A combined with exercise for chronic neck pain patients, was no better at four weeks or six months, than a variety of other treatments (exercise, or exercise combined with dry needling, or lidocaine, or ultrasound), for several different types of outcomes, including pain, disability and quality of life.

Side effects were reported to be mild and transient, with about 27% of subjects reporting side effects after receiving BoNT-A treatment. They included injection site soreness, shoulder or arm weakness, fatigue, heaviness, numbness, flu-like symptoms, systemic fever, shivering, generalized muscle soreness, vertigo (dizziness) and headache. Rare, but serious side effects, such as the relaxing of respiratory muscles or allergic reactions have been reported.

At this time, there is no evidence to support the use of BoNT-A for treating individuals with chronic neck pain or associated headache. These conclusions are limited by the overall quality and quantity of the available data.