Comparative longitudinal assessment of bone mineral density measured with radiofrequency echographic multi-spectrometry (REMS) and dual X-ray absorptiometry (DXA) in a cohort of peritoneal dialysis patients

In a peritoneal dialysis longitudinal cohort, REMS and DXA showed similar bone density, but only DXA detected significant cortical bone loss. REMS tended to give lower bone mineral density values and did not correlate with DXA changes. REMS may complement DXA when DXA findings are difficult to interpret or when closer longitudinal monitoring is needed, but larger studies are needed to define its role.BackgroundDual X-ray absorptiometry (DXA) is the gold standard for bone mineral density (BMD) assessment in chronic kidney disease patients, but it can have limited accuracy and radiofrequency echographic multi-spectrometry (REMS) may overcome some of its limitations. We aimed to evaluate and compare longitudinal BMD changes assessed by DXA and REMS in peritoneal dialysis (PD) patients.MethodsProspective cohort study including 20 PD patients that underwent two separate DXA and REMS evaluations. Clinical, biochemical, and imaging data, including abdominal aorta calcification (AAC) scores and fracture history, were collected. Longitudinal BMD changes were analyzed using linear mixed-effects models.ResultsOver a median follow-up period of 19.8 months, DXA revealed significant BMD declines at the femoral neck (FN) (-3.2%, p = 0.039) while REMS showed non-significant numerical declines. REMS consistently yielded lower BMD values than DXA at both lumbar spine (LS) and FN. The trajectories of BMD over time were similar between DXA and REMS for LS (p = 0.806) and FN (p = 0.229), regardless of whether patients were receiving anti-osteoporosis treatment and of the presence of fractures. However, there was no correlation of the annualized changes in BMD at the FN and LS between the two methods (r2 = 0.001, p = 0.905, and r2 = 0.001, p = 0.928, respectively).ConclusionDXA detected a significant decline in BMD at cortical sites, whereas REMS did not, and the correlation of the annualized changes between the two methods in assessing changes over time was poor. Given the small sample size, larger studies are needed to clarify the comparative and possible complementary roles of REMS and DXA in monitoring bone health in end-stage kidney disease.