Botulinum toxin intramuscular injections for neck pain: a systematic review and metaanalysis

To assess the effect of intramuscular botulinum toxin type A (BoNT-A) injections on pain, function/disability, global perceived effect, and quality of life (QOL) in adults with neck pain (NP).

We searched Central, Medline, and Embase databases up to June 2010. A minimum of 2 authors independently selected articles, abstracted data, and assessed risk of bias and clinical applicability. We estimated standard mean differences (SMD) with 95% CI, relative risks (RR), and performed metaanalyses (SMD(p)) using a random-effects model for nonheterogeneous data. The approach of the Grading of Recommendations Assessment, Development, and Evaluation working group summarizes the quality of evidence.

We selected 14 trials. High-quality evidence suggested BoNT-A was no better than saline at 4 weeks [4 trials/183 participants; SMD(p) -0.21 (95% CI -0.50 to 0.07)] and 6 months for chronic NP. Moderate-quality evidence showed a similar effect for subacute/chronic whiplash-associated disorder (WAD) on pain [4 trials/122 participants; SMD(p) -0.21 (95% CI -0.57 to 0.15)], disability, and QOL. Very low-quality evidence indicated BoNT-A combined with exercise and analgesics was not significant for chronic NP reduction at 4 weeks [3 trials/114 participants; SMD(p) -0.08 (95% CI -0.45 to 0.29)] but was at 6 months [2 trials/43 participants; SMD(p) -0.66 (95% CI -1.29 to -0.04)].

Current evidence does not confirm a clinically or statistically significant benefit of BoNT-A used alone on chronic NP in the short term or on subacute/chronic WAD pain, disability, and QOL. Larger trials, subgroups, and predictors of responses defined a priori (to facilitate selection of patients most likely to benefit) and factorial designs to explore BoNT as an adjunct treatment to physiotherapeutic exercise and analgesics are needed.