Correlation between radiofrequency echographic multi-spectrometry (REMS) and dual-energy X-ray absorptiometry (DXA): A systematic review and meta-analysis

Radiofrequency echographic multi-spectrometry (REMS) is an emerging ultrasound-based technology for assessing bone mineral density (BMD). However, its correlation with the current gold standard, dual-energy X-ray absorptiometry (DXA), has not been comprehensively quantified. This study aimed to systematically review and meta-analyse the correlation between REMS and DXA. We systematically searched databases for studies reporting correlations between REMS and DXA at the femoral neck and lumbar spine. Correlation coefficients were pooled via random-effects models after Fisher's z transformation. We performed subgroup analyses and meta-regression by study location, population, and sample size, along with sensitivity analyses. Seventeen studies comprising 8769 participants (90.9% female) for lumbar spine and 7464 participants (92.3% female) for femoral neck BMD were included. The pooled correlation coefficients between REMS and DXA were r = 0.819 (95% CI: 0.684–0.900) at the femoral neck and r = 0.769 (95% CI: 0.610–0.868) at the lumbar spine, indicating strong correlations. Correlations were comparable between studies from Italy and other countries but were lower in disease-specific populations and in smaller studies. Meta-regression confirmed significant effects of population type and sample size. Statistical heterogeneity was high (I 2 > 95%), and sensitivity analyses supported the robustness of the findings. REMS-derived BMD demonstrates strong correlation with DXA at both anatomic sites. Correlation strength varies by population characteristics and study size. Further longitudinal studies are needed to assess REMS's ability to predict fractures and its performance in diverse clinical and community settings. • Strong correlation found between REMS and DXA for bone density measurement • Correlation varies with population type and study sample size. • REMS may provide added value in disease-specific bone health assessment. • High heterogeneity observed, indicating context-dependent correlation strength • Further research needed for fracture prediction and diverse population validation