Theoretical investigations into methods of detecting pulmonary arteriovenous blood shunts

Abstract

The existence of a pulmonary-arteriovenous blood shunt, resulting in severe hypoxaemia (8) requires rapid and accurate detection. In order to evaluate the experimental techniques available for routine clinical use,a detailed understanding is required of the mechanism of shunt and its effect on pulmonary gas exchange. Initially, therefore, a mathematical model of gas exchange in the lung is formulated to give a continuous simulation of the non-steady state, as opposed to a ‘breath by breath’analysis (45, 47). The model illustrates clearly the effects, upon gas exchange, of parameters such as the solubility of a gas, abnormal regional distributions of ventilation, perfusion and lung volume,anatomical dead space, different levels of minute volume, in addition to pulmonary arteriovenous blood shunts. The model is utilized to simulate two experimental indicator diluation techniques which purport to measure intra-pulmonary blood shunts (15, 25), and tests the degree of discrimination of each method in determining true blood shunts in the presence of various defects. The accuracies of three tracer gases, xenon, krypton and sulphur hexafluoride in evaluating the proportion of cardiac output traversing the shunt pathways are compared. In light of these theoretical findings, the most suitable clinical arrangements for the evaluation of true shunt are proposed. Experiments performed on dogs with Xe!33 are discussed, and compared with theoretical predictions. Further development of the model to include recirculation of blood, and hence tissue absorption of gases, is described briefly since models of inert gas exchange have very important applications in the field of radiation dosimetry, gas washout studies and other medical problems (31).