Background: Systemic sepsis is a major cause of death in preterm neonates. There are compelling theoretical reasons why treatment with haemopoietic colony-stimulating factors might reduce sepsis and improve outcomes, and as a consequence these agents have entered into use in neonatal medicine without adequate evidence. We assessed whether granulocyte-macrophage colony stimulating factor (GM-CSF) administered as prophylaxis to preterm neonates at high risk of neutropenia would reduce sepsis, mortality, and morbidity.
Methods: We undertook a single-blind, multicentre, randomised controlled trial in 26 centres between June, 2000, and June, 2006. 280 neonates of below or equal to 31 weeks' gestation and below the 10th centile for birthweight were randomised within 72 h of birth to receive GM-CSF 10 microg/kg per day subcutaneously for 5 days or standard management. From recruitment to day 28 a detailed daily clinical record form was completed by the treating clinicians. Primary outcome was sepsis-free survival to 14 days from trial entry. Analysis was by intention to treat. This study is registered as an International Standard Randomised Controlled Trial, number ISRCTN42553489.
Findings: Neutrophil counts after trial entry rose significantly more rapidly in infants treated with GM-CSF than in control infants during the first 11 days (difference between neutrophil count slopes 0.34 x 10(9)/L/day; 95% CI 0.12-0.56). There was no significant difference in sepsis-free survival for all infants (93 of 139 treated infants, 105 of 141 control infants; difference -8%, 95% CI -18 to 3). A meta-analysis of this trial and previous published prophylactic trials showed no survival benefit.
Interpretation: Early postnatal prophylactic GM-CSF corrects neutropenia but does not reduce sepsis or improve survival and short-term outcomes in extremely preterm neonates.