We prospectively studied newborn infants with sepsis and neutropenia who were randomly selected to receive standard supportive care and either adjuvant granulocyte transfusions or intravenous immune globulin (IVIG) infusions; 21 infants received granulocyte transfusions and 14 received IVIG infusions. Half of the patients were premature (gestational age less than or equal to 32 weeks); the average postnatal age was 5
days (range 3 to 8 days). All infants had neutropenia by the criteria of Manroe et al., and the mean average bone marrow neutrophil storage pool ranged between 35% and 37%. There were no significant differences with respect to serum IgG, IgA, IgM, and total hemolytic complement values between treatment groups or between survivors and nonsurvivors. Clinical severity as defined by hypoxia, acidosis, and hypotension was similar between treatment groups. Group B streptococcus was the most common organism identified and accounted for almost 33% of all bacterial isolates. There was a significantly different survival rate in the group receiving polymorphonuclear leukocyte transfusions (100%, 21/21) compared with the group receiving IVIG infusions (64%, 9/14; p = less than 0.03). There were no significant complications in either treatment group with respect to fluid overload, secondary infection, blood group sensitization, pulmonary complications, or graft-versus-host disease. This pilot study suggests a possible benefit of granulocyte transfusions compared with 'IVIG therapy in the adjuvant treatment of neonatal neutropenia and overwhelming bacterial sepsis.