Intracarotid pressure measurements in the evaluation of a computer model of the cerebral circulation

It is difficult to predict which patients will tolerate occlusion of the internal carotid artery. This difficulty arises primarily because of uncertainties in the prediction of the adequacy of collateral circulation. Because of these uncertainties, balloon test occlusion and other methods have been developed to determine a priori the safety of carotid occlusion. However, all the methods are associated with significant false-positive and false-negative rates, as well as other neurologic complications. Because of these problems, more accurate and less invasive methods for predicting tolerance of carotid occlusion are needed.

In this report, we present the initial clinical evaluation of a new method for assessing the collateral circulation aided by a mathematical model of the cerebral vasculature. Data from the angiograms of 14 patients who underwent carotid endarterectomy were used to create individualized simulations of their cerebral circulations. As a test of the accuracy of the simulations, we compared values of the intracarotid stump pressures predicted by the model to those measured at surgery during the period of carotid occlusion.

The pressure predictions of the model correlated well with those measured at surgery. Linear regression analysis of measured versus predicted values yielded a line with slope 1.05. The line with slope 1.00, which denotes perfect agreement between predictions and measurements, is within the 95% confidence interval of the slope determined from the regression analysis.

Mathematical models of the cerebrovascular circulation can provide good predictions of intravascular pressure in the collateral circulation, and may provide accurate predictions of the flow as well. The present study reveals several areas that need further development, such as the models of the microvasculature, measurement of the arterial dimensions from angiograms, and consideration of other collateral sources such as the leptomeningeal and retrograde ophthalmic sources of flow. Incorporation of these improvements may lead to a clinically useful, noninvasive assessment of the state of the cerebrovascular collateral circulation in the individual patient.