A randomized trial of granulocyte-macrophage colony-stimulating factor in neonates with sepsis and neutropenia

To determine whether adjunctive therapy with recombinant human granulocyte-macrophage colony-stimulating factor (rhGM-CSF) could reverse sepsis-associated neonatal neutropenia and improve neonatal survival and to assess its safety compared with conventional therapy in a control group.

This prospective, randomized, controlled trial was performed in 60 infants with neutropenia and clinical signs of sepsis. A subcutaneous injection of rhGM-CSF (5 microgram/kg/day) was administered to 30 of the patients for 7 consecutive days. Hematologic parameters (absolute neutrophil, eosinophil, monocyte, lymphocyte counts, and platelet number) and outcome were compared with 30 conventionally treated (control) patients.

Twenty-five patients from the GM-CSF-treated group and 24 from the conventionally treated group had early-onset sepsis (3 days' postnatal age). There was no difference between groups in terms of birth weight; gestational age; gender; maturity; maternal age; and incidence of prolonged rupture of membranes, maternal hypertension, or severity of sepsis. All neonates tolerated GM-CSF well with no adverse reactions. The absolute neutrophil count on day 7 was significantly increased in the GM-CSF-treated group compared with the conventionally treated group: 8088 +/- 2822/mm(3) versus 2757 +/- 823/mm(3). The mean platelet count was significantly higher on days 14 in the GM-CSF-treated group compared with conventionally treated group: 266 867 +/- 55 102/mm(3) versus 229 200 +/- 52 317/mm(3). Hematologic parameters were otherwise similar between groups before treatment and on day 28. Twenty-seven neonates in the rh-GMCSF group and 21 in the control group survived to hospital discharge. The mortality rate in the rhGM-CSF group (10%) was significantly lower than in the conventionally treated group (30%).

Treatment with rhGM-CSF is associated with an increase in absolute neutrophil, eosinophil, monocyte, lymphocyte, and platelet counts and decreased mortality in critically ill septic neutropenic neonates. These results suggest that rhGM-CSF may be effective in the treatment of neonatal sepsis with neutropenia, and further randomized trials are needed to confirm its beneficial effects.