Serum sodium as a risk factor for hepatic encephalopathy in patients with cirrhosis and ascites.

Background and aim Hyponatremia is associated with development of hepatic encephalopathy (HE), but the nature of the relationship between serum sodium and HE incidence is unknown. We examined the association between serum sodium, changes in serum sodium, and HE incidence using data from three randomized trials of satavaptan in cirrhosis patients with ascites. Methods During follow-up, patients were examined for HE, and serum sodium was measured regularly. We used fractional polynomials to estimate the nature of the association between current serum sodium and hazard rate of HE (e.g. with a linear, logarithmic, or exponential slope) and Cox regression to adjust for confounders. Moreover, we examined the association between serum sodium at and 30-day and 1-year cumulative risk of HE. Finally, we examined the effect of change in serum sodium since inclusion on the hazard rate of HE. Results We included 1116 patients of whom 302 developed HE. Median serum sodium at was 137 (interquartile range, 134-139). The lower the current serum sodium, the higher the rate of HE. Specifically, the confounder-adjusted HE hazard rate increased linearly by 8% (adjusted hazard ratio = 1.08, 95% confidence interval: 1.06-1.10) for every mmol/L decrease in serum sodium over the range of measured values. Current serum sodium had a stronger effect on the HE rate than the changes in serum sodium since inclusion. Conclusion The hazard rate of HE development increased by 8% for every mmol/L decrease in serum sodium. Further, current serum sodium had a stronger effect on the HE rate than changes in serum sodium.