Aims:
Hyponatraemia is a common electrolyte disorder that is associated with increased falls and fractures(1,2). Hyponatraemia has been associated with reduced bone mineral density (BMD) assessed using DXA(3), and reduced trabecular and cortical bone volume in an animal model(4). We aimed to quantify the morphological basis of any reduction in volumetric BMD (vBMD) in humans using High-Resolution peripheral Quantitative Computed Tomography (HR-pQCT), a non-invasive imaging technique assessing bone microarchitecture(5). We hypothesised that hyponatraemia would be associated with reduced vBMD.
Methods:
A database of 2557 adults scanned using HR-pQCT between 2007-2021 was matched to Austin Health pathology records to identify serum sodium measurements within 6 months of HR-pQCT date (closest, and nadir if multiple results). HR-pQCT outcomes across six pre-specified clinical strata by sex and age (18-50/50-75/>75 years) were assessed using repeated‐measures mixed‐effects models. Ethics approval was granted by Austin Health Office for Research.
Results:
Serum sodium measurement within 6 months of HR-pQCT was available in 1462 individuals. Mean age was 59.5 years, 69% were female. The mean serum sodium closest to HR-pQCT date was 141mmol/L (range 127-148mmol/L). Forty-six individuals (3.1%) had serum sodium <135mmol/L, of which 5 (0.3%) had values <130mmol/L. Across the whole cohort, total vBMD at the radius and tibia was predicted by both the closest (p<0.001) and the nadir serum sodium (p<0.001), dependent on age and sex. The effect was most pronounced in women over 75 years, where correlation was then found between lower closest sodium and deteriorated bone microstructure (r=0.29, p<0.001, Fig. 1). Limitations include the observational design, reliance on single sodium measurements, and lack of information regarding potential confounders.
Conclusion:
Lower serum sodium was associated with deteriorated bone microarchitecture in older women. Whether hyponatremia is the mechanism for impaired bone microarchitecture, and whether this correlates with fracture risk, requires confirmation in prospective studies.