• Incidence of hyponatremia (defined as serum Na+ less than 130mmol/L) [ Time Frame: 72 hrs ] [ Designated as safety issue: Yes ]
  

• Incidence of hypernatremia (defined as serum Na+ more than 150 mmol/L) [ Time Frame: 72 hrs ] [ Designated as safety issue: Yes ]
  
• Plasma vasopressin levels at 24 hr and 48 hours [ Time Frame: 48 hrs ] [ Designated as safety issue: No ]
  
• Incidence of symptoms of hyponatremia such as nausea, malaise, headache, lethargy, confusion, obtundation, stupor, seizures and coma [ Time Frame: 72 hrs ] [ Designated as safety issue: Yes ]
  
• Incidence of symptoms of hypernatremia as altered sensorium, thirst, weakness, irritability, focal neurological deficits, seizures and coma [ Time Frame: 72 hrs ] [ Designated as safety issue: Yes ]
  

Hyponatraemia arises in between 20% and 45% of sick hospitalized children. An important reason for this high incidence could be use of hypotonic fluids in sick children for maintenance fluid therapy. There are no randomized controlled trials to evaluate the effect of various types of intravenous fluids on the incidence of hyponatremia in sick hospitalized children. We therefore plan to conduct a randomized controlled trial to evaluate the effect of normal saline in 5% dextrose at standard maintenance rate, reduced volume (2/3 maintenance rate) of N/5 saline in 5% dextrose and N/5 saline in 5% dextrose at standard maintenance rate on the incidence of hyponatremia in hospitalized children, aged 3 months- 12 years. & (2.) To determine serial plasma vasopressin levels in hospitalized children at baseline, 24 hours and 48 hours of intravenous fluid therapy and compare the values in the three fluid regimens. Study design: Randomized controlled trial. Hospitalized children who fulfill inclusion criteria and not having any of the exclusion criteria will be considered for the enrolment after written informed consent. Venous blood samples will be taken at enrollment for estimation of serum sodium, potassium, chloride, bicarbonate, blood gas, blood sugar, blood urea, serum creatinine, and plasma osmolality. A sample for estimation of plasma vasopressin will be collected at baseline. After randomization into three groups, one group of children will receive N/5 saline in 5% dextrose at standard maintenance rate (100 ml/kg for the first 10 kg of body weight, 50 ml/kg for the next 10 kg, and 20 ml/kg for body weight exceeding 20 kg).The second group of children will receive N/5 saline in 5% dextrose at 2/3 maintenance rate. The third group will receive dextrose normal saline at standard maintenance rate. Serum Na+, K+ and urine Na+, K+ will be estimated every 12 hourly till the patient is on intravenous fluid therapy and 12 hrs after stopping exclusive intravenous maintenance fluids. Serum and urine osmolality will be estimated every 24 hrs by an osmometer. Plasma vasopressin will be estimated in children in the 3 groups at 24, and 48 hours of intravenous fluid therapy. Children will be weighed every 24 hours. The fluid balance, sodium balance, free water clearance will be calculated in a subset of children.

The study measurements will be carried out only till the time the child is on exclusive intravenous maintenance fluid therapy or 72 hrs of starting the intravenous fluid therapy. The decision to decrease/ stop intravenous fluid therapy will be left to the treating unit.

The primary outcome measure will be incidence of hyponatremia (defined as serum Na+ less than 130 mmol/L).

The secondary outcomes studied will be Plasma vasopressin levels at 24 hr and 48 hours and incidence of hypernatremia.

Sample size: Based on literature review, the incidence of hyponatremia with standard intravenous fluid therapy is approximately 30%. Sample of 72 patients will be needed in each group to demonstrate the decrease in incidence of hyponatremia to 10%, with a beta of 0.2 (Power 80%) and alpha error of 0.05. Analysis: The data will be analyzed using STATA software. The outcomes (primary and secondary) in the 3 groups will be compared. For continuous variables, t test or Wilcoxon rank-sum test will be used to determine statistical significance. For categorical variables, chi square test will be used.