Objectives. To determine whether adjunctive therapy with recombinant human granulocyte colony-stimulating factor (rhG-CSF) could reverse the neutropenia and reduce the incidence of sepsis (less than or equal to 28 days postnatal age) in neonates with prolonged preeclampsia-associated neutropenia compared with conventional therapy. Study Design. An intravenous infusion of rhG-CSF (10 mu g/kg/day x 3 days for 10 neonates or 5 mu g/kg/day x 3 days for 5 neonates) was administered to ventilated patients with prolonged (greater than or equal to 3 consecutive days in the first postnatal week) preeclampsia-associated neutropenia (absolute neutrophil count [ANC] <1500/mm(3)). Neutrophilic responses and the incidence of neonatal sepsis in the next 28 postnatal days were compared with 13 case-matched control neonates who also had prolonged preeclampsia-associated neutropenia. Sepsis was defined as at least one positive blood culture in a newly symptomatic neonate treated with antibiotics for greater than or equal to 7 days. Results. No significant differences existed among the three groups in the birth weight, gestational age, sex, growth retardation, method of delivery, magnitude of respiratory support, use of surfactant, usage of intravascular catheters, or in the initial (pretreatment) ANC. The average baseline ANC (pretreatment) in the 10-mu g rhG-CSF group was 815 +/- 169/mm(3) (mean +/- SEM), in the 5 mu g group it was 786 +/- 165/mm(3), and in the conventional group it was 965 +/- 283. Eighteen of 28 (64%) neonates with preeclampsia-associated neutropenia were neutropenic at birth, the other 10 (36%) had normal neutrophil counts at birth but subsequently developed greater than or equal to 3 days of neutropenia between 24 and 120 hours after birth. The ANC increased by 2-fold at 24 hours, by 4-fold at 72 hours, and 14-fold by the 7th day in the 10-mu g group. In the 5-mu g group, a 2-fold and 5-fold increase occurred at 72 hours and 7 days, respectively. In the conventionally-treated group, only a 4-fold increase was seen as late as 7 days after achieving entry criteria. Sepsis was observed in 13% (2/15) of the rhG-CSF-treated neonates compared with arm incidence of 54% (7/13) in the conventionally-treated neonates. Conclusions. rhG-CSF increases the ANC significantly (at 10 mu g/kg/day x 3 days) and reduces the incidence of neonatal sepsis in critically ill ventilated neonates with prolonged preeclampsia-associated neutropenia when compared with conventional therapy. A future prospective, randomized, and blinded trial is needed to validate the beneficial effects of prophylactic rhG-CSF therapy in neonates with prolonged preeclampsia-associated neutropenia.
