We did two studies to see if severe neutropenia might reduce the severity or delay development of O2-induced lung microvascular injury. First, we treated 11 rabbits with nitrogen mustard until their circulating neurophil count decreased to less than 50/microliters of blood, after which the rabbits breathed pure O2 until death; nine other rabbits received no nitrogen mustard and had normal numbers of circulating neutrophils during O2 breathing. All rabbits died of respiratory failure with pulmonary edema, and although chemotherapy decreased the number of neutrophils in the lungs by greater than 90%, it did not influence survival time or extravascular lung water content. To see if severe neutropenia might slow the development of O2-induced lung microvascular injury, we assessed the effects of sustained hyperoxia on lung fluid balance in unanesthetized lambs treated with hydroxyurea, so that their absolute neutrophil count was less than 50/microliters of blood. We measured pulmonary arterial and left atrial pressures, cardiac output, lung lymph flow, and concentrations of protein in lymph and plasma during a 2- to 4-h control period and then daily for 2 to 4 h as the lambs continuously breathed pure O2. After 3 days of hyperoxia, lymph flow doubled and the concentration of protein in lymph increased from 3.3 +/- 0.5 to 4.2 +/- 0.3 g/dl. Tracer studies with 125I-albumin before and 3 days after the start of O2 breathing confirmed the development of increased lung vascular permeability to protein. All lambs died of respiratory failure with pulmonary edema after 3–5 days in O2.