Chronic hyponatremia is linked to cognitive deficits, but the biological mechanisms connecting low serum sodium to impaired cognition are not well defined. Brain-derived neurotrophic factor (BDNF) is a central growth factor for neuronal survival, synaptic plasticity, and cognition; its relationship to hyponatremia had not previously been investigated. This report presents a secondary post hoc analysis from a randomized, double-blind, placebo-controlled, crossover trial that explored whether correction of hyponatremia influences serum BDNF levels.
The analysis included nine patients diagnosed with syndrome of inappropriate antidiuresis (SIAD) who completed both treatment periods of the parent trial. The trial employed a randomized, double-blind, placebo-controlled, crossover design. Patient selection and flow for this post hoc analysis are described in the study figures. The small sample size reflects the pilot, exploratory nature of this secondary analysis.
Participants underwent two 4-week treatment cycles: oral empagliflozin 25 mg/day and matched placebo. Serum sodium concentration, serum BDNF levels, and cognitive performance using the Montreal Cognitive Assessment (MoCA) were measured at baseline and after each treatment cycle. Analyses adjusted for potential covariates including treatment arm, sex, age, baseline sodium, and concurrent antidepressant therapy where applicable.
Median baseline sodium in the analyzed cohort was 131 mmol/L (interquartile range 130–133). After the empagliflozin treatment phase, median sodium rose to 134 mmol/L (131–136), a change that reached statistical significance (p = .04). During the placebo phase median sodium remained essentially unchanged at 130 mmol/L (126–132; p = .9). The study reports an observed increase in serum BDNF that correlated with increases in serum sodium.
The authors report a statistically significant association between the magnitude of sodium increase and the magnitude of BDNF increase. After adjustment for treatment arm, sex, age, baseline sodium, and antidepressant treatment, the estimated change in BDNF was 0.70 ng/mL per 1 mmol/L increase in serum sodium (p = .048). An interaction analysis indicated that this association was stronger during the empagliflozin treatment phase (interaction p = .03). In a subgroup analysis, patients who achieved normonatremia after empagliflozin showed a slightly greater BDNF increase compared with those who remained hyponatremic; detailed subgroup sample sizes and absolute BDNF values were not provided in the abstract.
MoCA scores were measured at baseline and after each treatment phase. In this cohort, MoCA scores did not differ between the empagliflozin and placebo phases. The report does not indicate a detectable short-term change in global cognitive screening performance associated with the observed biochemical and biomarker changes in this small sample.
In this pilot secondary analysis, correction of hyponatremia—most notably during treatment with empagliflozin—was associated with an increase in serum BDNF, a neurotrophic factor linked to cognition and memory. The findings suggest a potential biological pathway by which sodium correction could influence neuronal-supportive processes. However, cognitive screening using MoCA did not show parallel improvements in this small sample over the treatment intervals studied.
These results are hypothesis-generating: they raise the possibility that restoration of normal sodium levels may alter neurotrophic signaling measured peripherally, but causality and clinical significance for cognition remain unproven.
Key limitations are intrinsic to the study design and reporting: the analysis is a secondary post hoc exploration with a very small sample size (nine patients), limiting statistical power and generalizability. The abstract does not provide detailed absolute BDNF concentrations, timing of sample collection beyond baseline and post-treatment, or granular subgroup counts and effect sizes. Peripheral serum BDNF measurements may not directly reflect central nervous system BDNF activity. The lack of change in MoCA scores in this short-term study does not exclude cognitive effects over longer intervals or on more sensitive cognitive measures.
The authors call for additional research to clarify whether BDNF mediates cognitive impairment in hyponatremia, whether different hyponatremia treatments have distinct effects on BDNF, and whether changes in peripheral BDNF predict clinical cognitive outcomes.
The parent trial is registered on ClinicalTrials.gov (NCT03202667). The authors declared no conflicts of interest in the published report.