Randomized controlled trial of Ringer solution versus serum for partial exchange transfusion in neonatal...

We tested whether crystalliod solutions could be used instead of colloid solutions for partial exchange transfusions (PET) in polycythaemic neonates because crystalloid solutions are cheap, carry no risk of anaphylactic reactions and can be sterilized. We randomly assigned 20 term neonates with venous haematocrit (Hct) > 0.65 l/l to PET with either a serum preparation (BISEKO) or Ringer solution. Plasma volume (PV) was measured with Evans blue dilution. Blood volume (BV) and red cell mass were calculated from PV and venous Hct. Before PET both serum and Ringer groups had the same Hct (0.69 (0.66-0.76) vs 0.69 (0.66-0.71) l/l; median (range)) and BV (108 (81-116) versus 96 (68-121) ml/kg. During PET an equivalent amount of blood was withdrawn stepwise (19 (14-26) versus 17 (13-25) ml/kg and replaced by either serum or Ringer solution. More of the Ringer solution (median 77%) than of the serum (median 36%) given left the intravascular space within 4 h after PET (P = 0.016); but there was no significant difference in Hct after Ringer-PET compared to serum-PET (median 0.58 vs 0.56 l/l). No infant required repeat PET. Ringer-PET reduced BV from high to normal values (from median 96 to 83 ml/kg; P = 0.005), whereas after serum-PET BV remained high (from median 108 to 98 ml/kg; not significant).

PET with Ringer solution resulted in a haemodilution comparable to PET with serum and a correction of hypervolaemia.