Hypoxemia following pulmonary embolism: a dog model of altering regional perfusion.

We studied the role played by a shift in perfusion to hypoxic lung areas after pulmonary embolism in pst embolic hypoxemia. A tracheal divider was used to separate hypoxic (N2 ventilated) from oxygenated (100 per cent O2 ventilated) lung in anaesthetized dogs. Relative perfusion was assessed from total 133Xenon (133Xe) exhaled from each lung area after intravenous infusions. When one lung area was ventilated with N2 and the other with O2 at a normal PaCO2 to allow hypoxic pulmonary vasoconstriction (HPV), there was a significant (P less than 0.001) shift away from the hypoxic side. Starch or blood clots were then infused to produce pulmonary emboli. Starch emboli were distributed predominantly to the oxygenated lung. After blood clot embolization in normocapnic dogs, pulmonary artery pressure increased 15 torr, perfusion to the hypoxic lung increased from 14 +/- 2 to 23 +/- 1 per cent, and PaCO2 fell from 278 to 186 torr. When the degree of HPV was reduced in another group of dogs by hypocapnea, a similar increase in pulmonary artery pressure (14 torr) created by blood clot embolism did not shift perfusion or create hypoxemia. In all dogs the perfusion shift to hypoxic lung was sufficient to account for all the post embolic hypoxemia. In this dog model, post embolic hypoxemia is explained by preferential distribution of emboli to oxygenated lung followed by perfusion shift to hypoxic lung as the effect of HPV is overcome by pulmonary hypertension.