Effects of hypo and iso - tonic intravenous maintenance fluids on plasma sodium level in hospitalized children. a double blind randomized clinical trial

Background: Hypotonic fluids are widely used in pediatrics and are the standard of care when giving maintenance fluids. However, there are several reports of risk of iatrogenic hyponatremia attributed to this practice in the literature. There is therefore uncertainty as to whether isotonic fluids would be the more appropriate fluid. Objectives: The primary objective was to compare effects of hypotonic and isotonic maintenance fluids on plasma sodium levels in children between the ages of 2 months and 15 years admitted at Aga Khan University Hospital (AKUH). The secondary objective was to compare effects of hypotonic and isotonic maintenance fluids on other plasma electrolytes and need for additional fluid boluses in the same population of patients. Study design: Double blind randomized controlled trial of isotonic vs hypotonic maintenance intravenous fluids in children. Methods: One hundred and fifty two children with serum sodium levels between 130-150 mEq/L who required intravenous maintenance fluids were randomized to receive either 0.9% dextrose normal saline or hypotonic maintenance fluids as per normal practice. Hypotonic solution which was used in the control group, had sodium concentration between 20 and 100 mEq/L corresponding to 4mEq/Kg/24hr. Children aged between 2 months and 15 years requiring hospitalization at AKUH were eligible only when their physician prescribed intravenous maintenance fluid therapy. Patients with chronic or acute kidney failure/disease, at risk of cerebral edema (diabetic ketoacidosis or cranio- encephalic trauma), neonates (age \u3c2month old), sickle cell, with plasma sodium levels at hospital admission \u3c130mEq/L or \u3e150mEq/L, and/or cerebral malaria or severe malnutrition ( Z score \u3c-3) and who refused to consent were excluded. Children were clinically monitored as per standard protocol. Blood electrolytes were obtained before commencement of v infusions to determine those eligible for inclusion and repeated after 24 hours or when fluids were stopped whichever was earlier. Any undesirable side effects were documented and managed as per standard practice. Primary outcome: Proportion of children with hyponatremia at 24hours after administration of maintenance fluids. Results: A total of 152 subjects were enrolled after obtaining informed consent. Of these, 78 and 74 were randomized to isotonic and hypotonic groups respectively. Overall, 25 (16.4%) patients dropped out of the study for various reasons. Hyponatremia occurred in 16.4% of the patients at 24 hours and a higher proportion of patients (23%) in the hypotonic group experienced hyponatremia compared to the isotonic group (10.3%) that was statistically significant (p = 0.03). The relative risk of hyponatremia in isotonic group was 0.45 (95% CI = 0.21 – 0.97). The overall mean serum potassium between the arms was 4.11 mEq/L (0.58 ±SD) (p = 0.01, 95% CI = 4.01 to 4.20). The overall mean serum chloride between the arms at 24 hours was 105.6 mEq/L (4.67±SD) (p=0.76 95% CI 104.83 - 106.34). In the isotonic group there was a significant difference in the mean change of chloride level (mean = 104.31 mEq/L ±5.17SD, p = 0.005, 95 CI% = 103.5 - 105.13) between admission chloride level and 24 hours chloride level but not in the hypotonic group (mean = 105.04 mEq/L ±5.46SD, p = 0.15, 95% CI 104.15 to 105.93). There was significant difference in the mean change of sodium in the hypotonic group ( mean = 137.37 mEq/L ±3.9SD, p = 0.05, 95% CI 136.74 – 138.0) but not in the isotonic group. Potassium levels between admission and 24 hours in both the isotonic and hypotonic groups were not significant. Adverse events occurred in 8 (5.6%) patients comprising of 3 (4.23%) in hypotonic group and 5(7%) in isotonic group (fisher’s exact = 0.72). One patient in both groups developed hypokalemia. Number needed to treat was 8 with isotonic maintenance fluids. In multivariate analysis, admission hyponatremia (p = 0.02, IRR 2.76, 95% CI 1.22 – 6.23), gastroenteritis (p = 0.03, IRR 2.68, 95% CI 1.11 – 6.48) and otitis media (p = 0.04, v i IRR 3.84, 95% CI 1.05 – 14.0) were independently associated with increased risk of hyponatremia at 24 hours. Isotonic maintenance and use of isotonic saline bolus fluids for resuscitation at admission both independently reduced the risk of hyponatremia at 24 hours by 61% (p = 0.03, IRR 0.39, 95% CI 0.16 - 0.91) and 74% (p = 0.04, IRR 0.26, 95% CI 0.07 - 0.93) respectively. Conclusion: An unacceptably high proportion (16.4%) of hospitalized children on maintenance intravenous fluids at Aga Khan University Hospital experience hyponatremia. Use of isotonic maintenance fluids significantly reduced the risk of hyponatremia at 24 hour by 55% and for every eight patients treated with isotonic maintenance fluids, one was protected from developing hospital acquired hyponatremia hence is preferred to hypotonic maintenance solutions currently recommended in the pediatric textbooks and Aga Khan University Hospital. Children presenting with hyponatremia, gastroenteritis and otitis media at admission and who received bolus resuscitation fluids are at a lower risk of developing hyponatremia at 24hr irrespective of the maintenance fluids used. A study to specifically determine the effectiveness of bolus saline fluids administration in prevention of hyponatremia in this subgroup of patients is recommended