To examine a possible mechanism which could cause arterial hypoxaemia following pulmonary embolism, we collapsed and did not ventilate one lung in each of eleven dogs, to produce hypoxic pulmonary vasoconstriction. In five dogs (Starch Group), Pap2 fell from 10 to 7.7 kPa (76.2 to 58.4 torr) as shunt fraction (Qs/Qt) rose from 19 to 31 per cent. Mean pulmonary artery pressure (P~-A), Paco2 and VD/VT remained constant. Starch emboli (63-74 ~t) were then injected. PF'A increased significantly and Pap2 dropped further to 5 kPa (37.8 torr) as Qs/Qt rose to 57 per cent. VD/VT increased and Paco2 remained constant. Microscopic examination of the lungs showed that three times as many emboli went to the ventilated lung compared to the unventilated lung. Six dogs (Blood Clot Group) received S~Cr labelled autologous blood clot. Changes after emboli in PP~, Pap2, Qs]Qt, Paco2 and VD/VT were similar to the results in the Starch group, t 2s 1 serum albumin was then injected and the dogs were sacrificed. The lungs were homogenized separately and the s i Cr and ~ 25I counted. The 5 ~Cr counts indicated 66 per cent of the blood clot emboli went to the ventilated lung. Following embolization, the ~2Sl counts suggested a shift in perfusion to the unventilated lung. We conclude from these results that emboli are preferentially distributed to ventilated lung. After embolization PP-~ increases. At least in this pulmonary embolism model the increased Pg'A may overcome hypoxic pulmonary vasoconstriction, redistribute blood hypoxaemia.VARIOUS HYPOTHESES have been advanced to explain the hypoxaemia that occurs following pulmonary embolism. These include bronchoconstriction, 1-3 opening of arterio-venous anastomoses in the embolised lung,'* impairment of diffusion ~'6 pulmonary oedema 7 and ventilation perfusion abnormalities. 8-*~ However, none of these hypotheses are universally accepted.The hypothesis we propose is that areas of the lung with reduced ventilation, hypoxic pulmonary vasoconstriction and, therefore, reduced perfusion, should not receive emboli. The emboli should go to the ",veil ventilated and perfuscd lung. The resulting pulmonary hypertension should shift blood flow to the poorly ventilated lung and, therefore, create hypoxaemia. To test this hypothesis we determined, first, the distribution of emboli in the lung when a portion of it Canad. Anaesth. Soc. J., vol. 27, no. 3, May 1980 to non-ventilated lung and create arterial is not ventilated and, secondly, the distribution of pulmonary blood flow after pulmonary embolism. We found that emboli are preferentially distributed to the vasculature of well ventilated lung and that, following embolism, pulmonary hypertension occurs, redistributing pulmonary blood flow to unventilated lung and contributing to immediate arterial hypoxaemia.
MATERIALS AND METHODSEleven mongrel dogs weighing between 15-25 kg were anaesthelised with intravenous pentobarbitone (Nembutal | 30 rag. kg-L Pentobarbitone was chosen because it does not reduce hypoxic pulmonary vasoconstriction...
