Analysis of the relationship between serum sodium concentration and laboratory parameters in diabetic patients

Abstract

Introduction: Diabetes is a chronic civilization disease that is associated with numerous complications and disturbances in laboratory parameters. One of the observed changes in laboratory studies of individuals suffering from diabetes is a significantly low level of sodium. Aim of study: The aim of study was to compare natremia with selected blood parameters determined on admission to hospital, and then to find correlation, if any, between those parameters. Materials and methods: We retrospectively reviewed the medical records of 169 patients with diagnosed type 2 diabetes and the average age of 72.43±11.62 years. We tested the correlation between natremia and: total, LDL and HDL cholesterol; creatinine; potassium; calcium; eGFRCKD-EPI and alanine aminotransferase levels in the patients. Statistical analysis was run on STATISTICA 13. The nonparametric Mann-Whitney, Spearmans’ rank correlation and the Shapiro-Wilk tests were used to analyse factors affecting laboratory test results for the patients, perform correlation analyses and check the conformity with normal distribution, respectively. Research results: A positive correlation was revealed between LDL level and natremia (p = 0.023), as well as between natremia and chloride ion concentration (p < 0.001) in the patients’ blood. No correlation was found between natremia and sex, age, eGFRCKD-EPI, alanine aminotransferase, HDL cholesterol, triglycerides, creatinine, as well as calcium and potassium concentrations. Conclusions: Statistically significant correlations were found exclusively between LDL level and natremia, and between chloride and sodium concentrations. High LDL level was correlated with high natremia, which requires further research due to the scant literature. The sodium, potassium and chloride concentrations are known to be interconnected by the need to ensure the electrolyte balance, as seen in the Gamble diagram. The remaining parameters were not statistically significant.

Keywords

• diabetes mellitus
• hyponatremia
• LDL
• cardiovascular risk