Evidence for cervical muscle morphometric changes on magnetic resonance images after whiplash: A systematic review and meta-analysis

Abstract

INTRODUCTION Morphometric changes to cervical musculature in whiplash associated disorder have been reported in several studies with varying results. However, the evidence is not clear because only a limited number of cohorts have been studied and one cohort has been reported in multiple publications. The aim of this study was to assess the evidence for cervical muscle morphometric changes on magnetic resonance (MR) images after whiplash using a systematic review with meta-analysis. MATERIALS AND METHODS PubMed, MEDLINE and Cochrane Library were searched without language restriction using combinations of the MeSH terms "muscles", "whiplash injuries", and "magnetic resonance imaging". Studies of acute and chronic whiplash were included if they compared whiplash and control cervical spine muscle morphometry measurements from MR images. The search identified 380 studies. After screening, eight studies describing five cohorts (one acute, three chronic, one both acute and chronic) met the inclusion criteria. Participant characteristics and outcome measures were extracted using a standard extraction format. Quality of eligible studies was assessed using the Newcastle-Ottawa Scale. Muscle cross-sectional area (CSA) and fat infiltrate (MFI) for acute and chronic whiplash cohorts were compared using mean difference and 95% confidence intervals. Meta-analysis models were created when data from more than two eligible cohorts was available, using inverse-variance random-effects models (RevMan5 version 5.3.5). RESULTS Quality assessment was uniformly good but only two studies blinded the assessor. Analysis of the acute cohorts revealed no consensus with respect to CSA. MFI was not measured in the acute cohorts. Analysis of the chronic cohorts revealed CSA is probably increased in some muscles after whiplash but there is insufficient evidence to confirm whether MFI is also increased. Because the available data were limited, meta-analyses of only multifidus were performed. In chronic whiplash multifidus CSA was significantly increased at C5 (Z = 3.51, p < 0.01) and C6 (Z = 2.66, p < 0.01); and MFI was significantly increased at C7 only (Z = 2.52, p < 0.01) but the heterogeneity was unacceptably high (I2 = 83%). CONCLUSIONS The strength of the evidence for cervical muscle morphometric changes on MR images after whiplash is inconsistent for CSA and MFI. Future study designs should be standardised with quantification of three-dimensional muscle morphometry.