Management of Hyponatremia in Preterm Infants on Diuretics

Study Question: Among very low birth weight infants, 400-1500 g, with bronchopulmonary dysplasia who develop hyponatremia while receiving hydrochlorothiazide diuretics: does oral sodium supplementation compared to fluid restriction affect FiO2 requirements (change in Respiratory Index Score \[RIS\] in Ventilated or CPAP babies, or change in FiO2 in spontaneously breathing babies) after four weeks? Randomization Method: Enrolled patients will be randomly assigned to either the sodium supplementation group or the fluid restriction group once they become hyponatremic (serum sodium \< 130) while taking hydrochlorothiazide. Interventions: Sodium Supplementation: Patients randomized to the sodium supplementation group will receive oral NaCl added to their feeds. When the serum sodium is 125-130, they will have 2 meq/kg/day of NaCl added to their feeds. If the serum sodium is 120-124, they will have 4 meq/kg/day of NaCl added to their feeds. Na supplementation will continue until the serum Na is \>135. Fluid Restriction: Patients randomized to the fluid restriction group will have fluid intake decreased by 20cc/kg/day. In order to maintain approximately the same caloric intake, 0.5 cc/kg/dose of corn oil (8.4kcal/cc), will be administered as a bolus every 6 hours. If this fluid restriction doesn't increase the serum sodium to above 130 within one week or if the serum Na is 120-124 and if the infant is receiving \> 140 cc/kg/d, the fluid intake will be decreased by an additional 10cc/kg/day for one additional week. Outcome Assessments Primary Outcome - the change in FiO2/RIS between baseline and outcome at 4 weeks after enrollment. Secondary Outcomes 1. 24-hour urine sodium, calcium, and creatinine at 4 weeks. 2. The mean serum Na+ nadir for each group. 3. The mean serum K+ nadir for each group. 4. Time to extubation for infants ventilated at enrollment. 5. Time on CPAP or mechanical ventilation for infants on CPAP at enrollment. Sample Size: The estimated sample size for the study will be 58, based on an effect size of 10% if the mean FiO2 is 40% (0.1 x 40= 4% absolute difference), expected standard deviation of 5% for FiO2, alpha (two-sided) = 0.05; Beta = 1 - 0.80 = 0.20. Analysis: The following analysis plan has been designed to allow every randomized infant to be included in the analysis (intention-to-treat analysis) regardless of whether they are intubated, extubated, taken off or put onto CPAP, or if they die during the course of the 4-week study period. All infants (both study groups combined) will be assigned a rank at baseline and at outcome (4 weeks) within each of the following subgroups: infants on O2 by oxyhood, infants on nasal cannula, infants on CPAP, infants on the ventilator, infants who die during the study. For each infant, a change in rank (outcome minus baseline) will be calculated. The change in rank will be compared between the two study groups using a non-parametric test.