L'ANGIOGRAPHIE CORONAIRE QUANTITATIVE. APPLICATION A L'EVALUATION DES ANGIOPLASTIES TRANSLUMINALES CORONAIRES

Quantitative coronary angiography, a technique developed at the Thorax Center in Rotterdam, is an automated analysis of the coronary contours by an analogical-digital convertor with computerised data-storage. The algorithm of detection uses a function derived from variations of brilliance between the artery and its surroundings along an axis perpendicular to the orientation of the coronary segment under analysis. The system extrapolates the theoretical contour at the site of a stenosis from the proximal and distal diameters (values retained: 90th percentile) and displays in absolute values and percentages the degrees of stenosis in diameter and surface area. The arterial contours and diagrams are displayed on a video screen allowing control and manual correction when necessary. The method was tested on experimental models and on coronary arteries fixed under physiological pressure. The choice of the image is all-important (centering, good filling, ... ), but the essential point is to analyse the coronary arteries in end diastole (determined by angiography, ECG or pressure tracings) to exclude distortion due to the rapid movement of the coronary tree during systole. Different incidences are required to analyse an eccentric stenosis; therefore the mode of expression of the results is under research: a three-dimensional reconstruction of the artery using the analysis of the 'brilliance' filmed in two orthogonal incidences is being attempted. The method has been applied to the evaluation of transluminal coronary angioplasty in the University Hospital of Nancy-Brabois. In a series of 12 patients, 24 measurements were compared before and after the procedure in the same incidence and with the same radiological constants. The following conclusions were drawn: (1) quantitative computerised analysis of coronary angiography demonstrates the positivity of the results of transluminal angioplasty (mean percentage of stenosed diameter: 64% before dilatation, 38% after dilatation, p < 0.001); any personal bias is excluded, as the measurements are made automatically by the computer; (2) the degree of stenosis measured manually appears to be increasingly overestimated with the severity of narrowing when compared to the computerised analysis; however, this difference is only slight in the measurement of residual stenosis after dilatation; (3) the results of quantitative angiography are logical, and agree with the measurements of coronary-artery diameter by other techniques and with the diameter and surface of the balloon used for angioplasty; (4) quantitative coronary angiography is a method of choice for assessing rapid variations in calibre of the coronary arteries (transluminal angioplasty, spasm, pharmacological testing, intracoronary streptokinase).