EP1029 - ECE_1026 - Serum neurofilament light chain: a candidate biomarker of neuroaxonal injury in patients with dysnatremia, associated with functional outcomes and mortality

Abstract

Both hypo- and hypernatremia are associated with neurocognitive and neuromuscular dysfunction. However, specific biomarkers for these complications are lacking. In the present study, we aimed to explore the association between serum sodium levels and serum neurofilament light chain (sNfL), an emerging circulating biomarker of neuroaxonal injury. sNfL levels were evaluated in serum obtained from stored specimens from subjects that participated in the population-based 2013-2014 cycles of the National Health and Nutrition Examination Survey (NHANES). Cognitive function was evaluated using the Consortium to Establish a Registry for Alzheimer's Disease-Word Learning, the Animal Fluency Test, and the Digit Symbol Substitution Test. Muscle strength was measured through a handgrip dynamometer. All-cause mortality was evaluated by linking NHANES data with data from the National Death Index. All analyses were adjusted for relevant confounders. The study population consisted of 2071 subjects. Serum sodium levels were low (≤135 mmol/L), normal (136-144 mmol/L), or high (≥145 mmol/L) in 47, 1999, and 25 participants, respectively. Patients with hyponatremia had higher sNfL levels compared to normonatremic subjects, independent of potential confounders (mean difference in log-transformed sNfL values: +0.44, 95%CI:[+0.06,+0.81], P = .03). Conversely, no significant alteration of sNfL levels was observed in hypernatremic patients. Evaluating sodium as a continuous predictor, a significant non-linear association was observed between sodium levels and sNfL (P < .01 for association; P < .01 for non-linearity), with an inflection at 138 mmol/L; below this threshold, an increase of +0.09 (95%CI:[+0.01,+0.16], P = .02) in log-sNfL was observed for each 1 mmol/L serum sodium decrease. Higher sNfL levels were associated with worse cognitive performance and muscle strength, independent of sodium levels and other potential confounders (P < .05 for all functional tests). Hyponatremia (HR = 2.18 compared to normonatremia, 95%CI:[1.13, 4.21], P = .02) and sNfL (HR = 2.63 per 1-unit increase in log-transformed sNfL, 95%CI:[1.91, 3.64], P < .01) were independent predictors of all-cause mortality. No significant differences in the predictive value of sNfL were observed across sNa categories (P = .13 for interaction). The predictive value of sNfL levels was also confirmed in a subgroup analysis limited to hyponatremic patients (HR = 2.89 per 1 unit increase in log-transformed sNfL, 95%CI:[1.76, 4.76], P < .01). We provide first-time evidence of an association between hyponatremia and increased sNfL levels. Conversely, albeit with the limitations imposed by the small number of hypernatremic participants, no significant increase in sNfL was observed in hypernatremia. sNfL could serve as a novel biomarker of neuroaxonal injury in patients with hyponatremia, correlated with functional outcomes and mortality.