Diabetes Atherosclerosis Intervention Study (DAIS): quantitative coronary angiographic analysis of coronary artery atherosclerosis.

The primary objective of the Diabetes Atherosclerosis Intervention Study (DAIS) is to determine by quantitative coronary angiography whether long-term correction of the dyslipoproteinemia of diabetes with micronized fenofibrate results in evidence of decreased progression or regression of angiographically measured obstructive coronary atherosclerosis. The purpose of this communication is to describe the angiographic methodology for the DAIS project, and to present data documenting the reproducibility of measurements that will determine the primary outcome of DAIS. Four hundred eighteen subjects between the ages of 40 and 65 were entered from 11 centers in Canada, France, Finland, and Sweden, with 305 males and 113 females. Thirty-two percent of subjects had undergone a previous coronary artery intervention, either PTCA or bypass grafting. Subjects underwent coronary arteriography at baseline according to a strictly defined protocol. The coronary tree was divided into AHA segments and quantitative analysis of segments was performed using the cardiovascular measurement system described by Reiber. Geometric parameters including mean lumen diameter, minimum lumen diameter, maximum lumen diameter, and segment length were determined. In 15 studies, measurements were carried out on the same frame by two observers, and at least 1 week apart by the same observer. In 13 studies, measurements were performed by the same observer on two separate injections of the same coronary artery. The mean of the standard deviation of the differences of measurements of all segments for the primary study analyst was 0.029 mm, with a mean of correlation coefficients of 1.00. Between two observers, the mean of the standard deviations of segmental mean lumen diameters was 0.347 mm with a mean of coefficients of variation of 0.78. The mean of standard deviations for measurements of segmental mean lumen diameter from two separate coronary injections was 0.122, with a mean of correlation coefficients of 0.94. The mean of correlation coefficients for minimum lumen diameter were 0.98 for intraobserver variability, 0.77 for inter-observer variability, and 0.96 for inter-angiogram variability. For segment length the corresponding values were 0.99, 0.79, and 0.94. These data demonstrate that a high level of reproducibility and precision may be achieved in a multicenter study in assessment of the coronary tree in carefully performed studies using this methodology. We anticipate the results will provide a statistically powerful conclusion with new and unique data to answer the question of the effect of long-term correction of dyslipoproteinemia on coronary atherosclerosis in type II diabetic patients with dyslipoproteinemia.