1106 Impact of Splanchnic Vasoconstrictors on Sodium Level in Inpatient Cirrhotics With Hyponatremia: A Retrospective Cohort Study

Abstract

INTRODUCTION: Hyponatremia (hypoNa) in cirrhosis represents another consequence of splanchnic vasodilation. Splanchnic vasoconstrictors are currently used in hepatorenal syndrome (HRS) which has similar pathogenesis, however their role in hypoNa has not been fully investigated. Aim of the study is to evaluate the effect of octreotide ± midodrine on hypoNa in cirrhotics. METHODS: We retrospectively included adult cirrhotics admitted to the Cleveland Clinic hospitals between 10/1/2011 and 10/1/2015, who had Na < 133 at or during admission. We excluded subjects with less than 5 days of follow up, those who received potent vasoconstrictors as norepinephrine or received hypertonic saline, tolvaptan or renal replacement therapy during admission. Population was divided into those who received octreotide versus those who did not. Outcomes were Na level at days 5, 7 and 10 after starting octreotide. Secondary outcome was determination of predictors of change of Na at day 5 using a multivariate model that included variables with P < 0.1 on univariate analysis. We then performed a subgroup analysis for only those with baseline Na < 125 (as this is the correction threshold for many transplant centers). RESULTS: Patients in the octreotide group were younger, had higher Model for End-Stage Liver Disease score, more patients had ascites and varices compared to controls [Table 1]. The octreotide group had lower baseline Na, but the Na levels at 5,7 and 10 days after octreotide were significantly higher [Figure 1]. The two predictors of higher Na at 5 days were having a lower baseline Na and receiving both octreotide and midodrine Table 2]. When analysis was restricted to the subgroup with baseline Na < 125, 97% of those who received octreotide had Na ≥125 by day 5 (compared to 77% of controls, P = 0.03). In this subgroup, the only predictor of reaching Na ≥125 by day 5 (multivariate logistic regression model) was receiving octreotide; OR 11.2 (1.16- 109), P = 0.04. CONCLUSION: Cirrhotics with hypoNa that received octreotide had a significantly higher rise in Na. On a multivariate analysis accounting for the potential confounders, octreotide remained a significant predictor only when combined with midodrine, suggesting a synergistic effect (similar to HRS). To our knowledge, this is the largest study evaluating the effect of splanchnic vasoconstrictors on sodium level in cirrhotics with hypoNa compared to controls. Prospective studies are needed with matched controls and longer follow up.