Effectiveness of manual therapy in the treatment of non-specific neck pain: a review

Abstract

The aim of the current review was to assess the evidence for the effectiveness of cervical spine manipulation and mobilisation in the treatment of non-specific neck pain. A review of the literature was performed in a systematic format, based upon a structured search of six computerised bibliographic databases, together with manual searching of selected journals and reference lists to identify randomised clinical trials (RCTs) researching the effectiveness of manual therapy in the treatment of non-specific neck pain. The methodological quality of the studies was assessed using a set of pre-defined criteria adapted from Koes et al. (Koes BW, Assendelft WJJ, van der Heijden GJMG, Bouter LM, Knipschild PG. Spinal manipulation and mobilisation for back and neck pain: a blinded review. BMJ 1991; 303 :1298–303) and updated in the light of recent literature.A total of 12 RCTs met the selection criteria. The scores achieved for methodological quality of the studies ranged from 25 points to 67 points out of a possible 100 points. Only four RCTs scored above 50 points: two of these reached a positive conclusion about the effectiveness of spinal manual therapy in the treatment of non-specific neck pain, and two a negative conclusion. Eight RCTs scored below 50 points; six of these reached a positive conclusion and two a negative conclusion.RCTs researching the effectiveness of manual therapy for non-specific neck pain demonstrate methodological shortcomings in both design and reporting. The effectiveness of spinal manual therapy on non-specific neck pain remains inconclusive. The available evidence favoured spinal manual therapy when it was used in conjunction with exercise, particularly in the treatment of patients with chronic non-specific neck pain. Further, well-designed RCTs are needed to draw valid conclusions about the effectiveness of spinal manual therapy on nonspecific neck pain. In such future RCTs, placebo treatments should be designed to be indistinguishable (as far as possible) from manual therapy, and without the specific effects of active therapy. Furthermore, psychosocial factors should be considered prior to randomisation of patients by using appropriate measures.