ObjectiveTo determine the safety and efficacy of filgrastim (r-metHuG-CSF) in combination with intravenous antibiotics to reduce the rate of mortality in patients with pneumonia and sepsis.DesignThis study was multicenter, double-blind, and randomized.SettingIntensive care unitsPatientsAdult patients with bacterial pneumonia, either acquired or nosocomial, as confirmed by chest radiograph and positive culture or Gram-negative stain, and severe sepsis, defined as sepsis-induced hypotension or organ dysfunction.InterventionsStandard antibiotic therapy with or without filgrastim (300 &mgr;g/day) or placebo administered as a 30-min intravenous infusion. The study drug was started within 24 hrs of enrollment and was continued for 5 days or until the white blood cell count reached >75.0 × 109cells/L.Measurements and Main ResultsThe primary end point was the occurrence of mortality through day 29; secondary end points included occurrence of subsequent organ dysfunction, time to discharge from intensive care unit, number of days on mechanical ventilatory support, and time to death. Study-related observations were recorded through day 10 and included vital signs, onset of organ dysfunction, clinical laboratory variables, and adverse events. Filgrastim increased the white blood cell count to a median peak of 31.7 × 109cells/L from a baseline of 12.3 × 109cells/L. The two groups were well matched and did not differ significantly with regard to severe adverse events, time to death, occurrence of end-organ dysfunction, days of intensive care unit hospitalization, or days on mechanical ventilatory support. Mortality was low in both treatment groups; the mortality rate in patients with adult respiratory distress syndrome was similar between the two groups.ConclusionsThe addition of filgrastim to the antibiotic and supportive care treatment of patients with pneumonia complicated by severe sepsis appeared to be safe, but not efficacious in reducing mortality rates or complications from this infection.