Endocrine Abstracts

Context: Hyponatremia is associated with an increased risk of osteoporosis and fractures, and in recent years increasing evidence is accumulating in favor of a likely causal relationship between hyponatremia and bone loss. In rat models, the induction of hyponatremia enhances osteoclast activation and bone catabolism. In humans, the correction of hyponatremia by tolvaptan or SGLT2-inhibitors has a favorable effect on bone turnover markers, possibly due to an interplay both with osteoblast and osteoclast activation. The impact of hyponatremia on non-invasive indices of bone quality, however, is unknown.

Objective: To evaluate whether trabecular bone microarchitecture, assessed non-invasively by trabecular bone score (TBS), is altered in patients with hyponatremia.

Methods: We conducted a cross-sectional analysis of the 2005–2008 cycles of the National Health and Nutrition Examination Survey (NHANES), in which TBS measurement was performed. The main outcome measures were TBS values and bone mineral density (BMD) T-scores at the lumbar spine, total hip and femoral neck.

Results: A total of 4204 subjects aged 50 years or older were included (4041 normonatremic, 163 hyponatremic). The mean serum sodium value in hyponatremic patients was 132.6±2.1 mmol/l, with sodium levels in the range of mild hyponatremia (130–134 mmol/l) in 90.8% of cases. Univariate analyses did not show any difference in TBS between patients with and without hyponatremia (1.308±0.145 vs 1.311±0.141, P=0.806). Hyponatremic subjects had lower BMD T-score at total hip (−0.70±1.46 vs −0.13±1.32, P<0.001) and femoral neck (−1.11±1.26 vs −0.72±1.14, P=0.004), while no difference was observed at lumbar spine (−0.27±1.63 vs −0.31±1.51, P=0.772). After adjustment for relevant confounders (i.e.: age, sex, ethnicity, smoking, body mass index, 25-hydroxyvitaminD, diabetes, chronic kidney disease, use of loop, thiazide or potassium-sparing diuretics, history of chronic glucocorticoid treatment), hyponatremia was confirmed as an independent predictor of lower BMD T-score at the total hip (β=−0.20, 95%CI:[−0.39,−0.02], P=0.029), while the significance was lost at the femoral neck (P=0.308). The lack of association between hyponatremia and lumbar spine BMD (P=0.236) or TBS (P=0.346) was confirmed.

Conclusions: This study confirms that hyponatremia is associated with a reduced bone density, particularly at the total hip. On the other hand, no significant impact on trabecular bone microarchitecture, evaluated by TBS, was found. The lack of prospective data does not allow to assess the prognostic role of these parameters as predictors of incident fractures, but the available evidence may suggest, differently from other forms of secondary osteoporosis, a possible limited added value of TBS in this setting.