Inhibition of hypoxic pulmonary vasoconstriction by carbon monoxide in dogs.

Objective: We tested the hypothesis that carbon monoxide might participate in the modulation of hypoxic pulmonary vasoconstriction (HPV) by prostacyclin (PGI 2 ) and nitric oxide. Design: Prospective, interventional study. Setting: University laboratory. Subjects: Nineteen intact anesthetized mongrel dogs. Interventions: Right heart catheterization for the measurements of mean pulmonary artery pressure (Ppa), left atrial pressure estimated from occluded Ppa (Ppao), pulmonary capillary pressure (Pcp) calculated from the Ppa decay curve after balloon occlusion, and cardiac output (Q); inferior vena cava balloon for the control of Q by manipulation of venous return; ventilation in hyperoxia (fraction of inspired O 2 , 0.4) or in hypoxia (Fio 2 , 0.1); inhibition of cyclooxygenase by indomethacin (Indo); inhibition of nitric oxide synthase by N G -nitro-L-arginine (L-NA); inhibition of heme oxygenase by mesoporphyrin IX (SnMP); inhalation of nitric oxide (20 ppm); and inhalation of carbon monoxide (100 ppm). Measurements and Main Results: The first seven dogs were weak responders to hypoxia as assessed by a hypoxia-induced increase in the gradient between Ppa and Ppao, measured at one level of Q kept constant, by an average of only 2 mm Hg (p = NS). This HPV was markedly increased by the combined administration of Indo and L-NA. A further enhancement of HPV was observed after the addition of SnMP, leading to severe pulmonary hypertension with an average increase in Ppa to 39 mm Hg. Inhaled nitric oxide inhibited HPV only after the combined administration of Indo, L-NA, and SnMP. Inhaled carbon monoxide had no effect. The next 12 dogs were stronger responders to hypoxia, as assessed by a hypoxia-induced increase in the gradient between Ppa and Ppao, measured at several levels of Q, by an average of 3 mm Hg (p <.05). This HPV was of the same magnitude after administration of placebo (n = 6) or SnMP (n = 6). Addition of Indo enhanced HPV to the same extent in the placebo and in the SnMP groups. Addition of L-NA induced a further enhancement of HPV, which was, however, greater in the SnMP group. There was a slight increase in the capillary-venous segment relative to the arterial segment in hypoxic conditions, but the partitioning of pulmonary vascular resistance was otherwise unaffected by nitric oxide, carbon monoxide, or PGI 2 . Conclusions: Endogenous carbon monoxide modulates canine HPV only in the absence of nitric oxide. The vasodilation mediated by nitric oxide, PGI 2 , or carbon monoxide is essentially distributed between proximal and distal sites proportionally to the degree of constriction produced during hypoxia.