A randomized, placebo-controlled trial of granulocyte colony-stimulating factor administration to newborn infants with neutropenia and clinical signs of early-onset sepsis

Abstract

Objective: To determine whether recombinant human granulocyte colony-stimulating factor (G-CSF) administration: 1) accelerates production of neutrophils; 2) increases bone marrow stored and precursor neutrophils; and 3) is safe in newborn infants with neutropenia and clinical signs of early-onset sepsis.

Study design: We randomized 20 infants with neutropenia and clinical signs of early-onset sepsis in the first 3 days of life to receive G-CSF (10 microg/kg/d) or placebo for 3 days. Entry criteria included neutropenia as defined by Manroe criteria, an elevated immature to total neutrophil ratio [(I/T) >/=0.25], and a requirement for ventilatory support. Cultures were obtained and antibiotics initiated on all study infants. Circulating absolute neutrophil count (ANC), I/T ratio, bone marrow neutrophil storage pool (NSP) and neutrophil proliferative pool (NPP), and plasma G-CSF concentrations were evaluated. Also, severity of illness as determined using the Score for Neonatal Acute Physiology (SNAP), morbidity, and mortality were recorded.

Results: Circulating ANC increased in both G-CSF and placebo recipients by day 1. Also, the I/T neutrophil ratio decreased in both G-CSF and placebo recipients. There were no significant differences in the ANC or I/T ratio between the two groups during the study period. Similarly, bone marrow NSP and NPP did not differ between G-CSF and placebo recipients at study entry or day 2. No differences were observed in the secondary outcome measures including severity of illness, morbidity, and mortality.

Conclusions: Administration of recombinant G-CSF to infants with neutropenia and clinical signs of early-onset sepsis did not increase circulating ANC, or bone marrow NSP and NPP compared with placebo. No differences were observed between G-CSF and placebo recipients in severity of illness, morbidity, or mortality. No adverse effects of G-CSF administrations were noted.