Is manipulative therapy clinically necessary for relief of neck pain? A systematic review and meta-analysis.

OBJECTIVE: To summarize and critically assess the effificacy of Eastern and Western manipulative therapies for the treatment of neck pain in adults.

METHODS: A search of PubMed/MEDLINE, the Cochrane Central Register of Controlled Trials, ClinicalTrials.gov, EMBASE, etc. from their inception date to January 2014 with Chinese, Japanese, and Korean databases. Two reviewers independently selected randomized controlled trials (RCTs) with negative control or blank control, extracted data and assessed methodological quality. Meta-analysis and levels of evidence were performed by Revman5.1 and Grades of Recommendations Assessment, Development and Evaluation (GRADE) approach.

RESULTS: Nineteen clinical trials with adequate randomization were included in this review, 11 of them had a low risk of bias. The primary outcome for short-term pain had no significant differences, however, the secondary outcome, only the Numerical Pain Rating Scale (NPRS) score of intermediate-term [n=916, pooled mean differences (MD) =-0.29, P=0.02], the Neck Disability Index (NDI) score of short-term (n=1,145, pooled MD=-2.10, P<0.01), and intermediate-term (n=987, pooled MD=-1.45, P=0.01) were signifificantly reduced with moderate quality evidence. However, it supported the minimally clinically important difference (MCID) of the Visual Analogue Scale and NPRS pain score to be 13 mm, while NDI was 3.5 points. The meta-analysis only suggested a trend in favor of manipulative therapy rather than clinical signifificance.

CONCLUSIONS: The results do not support the existing evidences for the clinical value of Eastern or Western manipulative therapy for neck pain of short-term follow-up according to MCIDs. The limitations of our review related to blinding, allocation concealment and small sample size.