When defining the mechanism of hypoxic pulmonary vasoconstriction (HPV), investigators have employed ex vivo preparations because of the belief that accurate, quantitative assessment of pulmonary microvessels could not be obtained in vivo. We hypothesize that accurate, quantitative assessment of pulmonary microvascular reactivity can be performed using a simple, in vivo preparation. Our aim was to provide this quantitative assessment in a defined animal model, and to confirm that the chosen preparation could discriminate changes in microvascular reactivity as influenced by endogenous mediators. New Zealand rabbits were instrumented for in vivo microscopy and direct measurement of subpleural arterioles. Rabbits were first randomized to either control (n = 7) or endotoxin (n = 5), infusion of Escherichia coli lipopolysaccharide (200 Fg/kg). All rabbits were then exposed to a repeated protocol of normoxia (21% O2) for 20 min and then hypoxia (15% O2) for 10 min over 2 h. The changes in arteriole diameter were measured at the end of each interval. Normal pulmonary arterioles repeatedly constrict 15+/-3.5% during hypoxia. Altering endogenous vasoactive mediators, as with infusion of endotoxin, caused a loss of hypoxia-induced vasoconstriction. The results of our study validate this experimental preparation for the reliable quantification of pulmonary microvascular reactivity and investigation of hypoxic pulmonary vasoconstriction under both normal and pathologic conditions.