SUMMARY Ten normotensive premature infants with idiopathic respiratory distress syndrome, and albumin concentrations of less than 30 g/l were given 5 ml/kg of 20% salt poor albumin by infusion. Concentrations measured six hours after infusion had increased significantly and these were associated with significant reduction in weight and improvement in urine output.Sick premature infants are often oedematous in spite of being hypovolaemic. ' The present study was designed to find out whether infusions of albumin were effective in treating hypoalbuminaemia in normotensive sick infants.
Patients and methodsInfants less than 2 weeks of age, except those receiving peritoneal dialysis and those with chest drains in situ (both of which interfere with weighing) were eligible for the study. Babies given albumin infusions for the emergency treatment of hypotension were also excluded.Ten infants entered the study. Their mean birth weight was 1390 g (range 560-3315 g) and mean gestational age 29 weeks (range 24-36). The postnatal age at the time of infusion ranged from 24 hours to 5 days, mean age 1*7 days. All had a clinical diagnosis of respiratory distress syndrome and were receiving artificial ventilation by a Sechrist ventilator. They were ventilated at peak inspiratory pressures ranging from 14-32 cm H20. Five of the infants were paralysed before and during the study. All 10 infants had albumin concentrations of <30 g/l, with peripheral oedema. The content of fluid to be given to each baby was decided daily. On the first day 40 ml/kg were given and this amount was not increased if the infant gained weight during the first three days.Urea, electrolytes, and albumin concentrations were measured daily in all babies, and infants who were hypoalbuminaemic (<30 g/l) were entered into the study. A solution of 5 ml/kg/20% salt poor albumin was then infused, and the volume subtracted from the total daily fluid requirement. Before the transfusion and at the end of the six hours after it the infant was weighed, and urea, electrolytes, and albumin concentrations measured again. For six hours before and after the infusion urine output was measured. Manual expression was not used and urine was collected that had been voided spontaneously in the two six hour periods.' During a similar period blood pressure was recorded hourly as the mean of the systolic and diastolic pressures using an intra-arterial transducer, and reported as the average for each six hour period.During the 12 hours of the study period no other alteration was made in the infant's fluid input, nor was there any change made in the type of incubator or the use of overhead heaters. Twelve hours was chosen as the study period as it was the longest over which the fluid intake of sick infants on ventilators could be guaranteed to be constant. Treatment with phototherapy was recorded.To assess any effects of the albumin infusion, albumin and electrolyte concentrations, blood pressure, weight, and urine output were compared before and after the infusion.Ethical permission ...