IMPACT OF LESION CHARACTERISTICS ON THE Q UALITY OF MEASUREMENTS WITHDIFFERENT QCA SYSTEMS

Reproducibility and accuracy of in vitro measurements are very high us ing recently developed QCA systems. We analyzed the impact of lesion c haracteristics and the image quality on the quality of measurements un der clinical conditions. For the study we selected 57 coronary artery lesions which had a clinically relevant distribution for stenosis seve rity, lesion characteristics, and image quality. Every effort was made to eliminate procedural sources of error. Three investigators measure d each lesion five times with each of three QCA systems (AWOS, Cardio and CMS). Only the measurements of the minimal stenosis diameter were analyzed. The precision of all the measurements was high with the AWOS (0,04 mm), the Cardio (0,05 mm), and the CMS systems (0,06 mm). Varia bility of measurements increased for the following criteria: Ambrose-I II morphology (CMS 0,082 mm), surface irregularities (Cardio 0,069 mm, CMS 0,073 mm), TIMI I (Cardio 0,084 mm, CMS 0,121 mm), and moderate i mage quality (CMS 0,07 mm). There were no differences in the precision of the measurements in the other groups of lesion characteristics. Th ere were no relevant differences in any of the measurements between th e systems (AWOS-Cardio -0,07 mm, AWOS-CMS -0,11 mm, Cardio-CMS -0,04 m m). Smaller diameters were measured with the AWOS system than with the CMS and the Cardio systems when the lesion was calcified (AWOS-Cardio -0,109 mm, AWOS-CMS -0,161 mm). This was only a trend, however, and d id not reach statistical significance, which was also true for the oth er differences found between the systems according to various lesion c haracteristics. In summary, we found that the measurement quality of t he QCA systems used in this study is not altered by the underlying les ion characteristics or the image quality.