Skelton T N, Kisslo K B, Bashore T M
Department of Medicine, Duke University Medical Center, Durham, North Carolina 27710.
Am J Cardiol. 1988 Sep 1;62(7):381-6. doi: 10.1016/0002-9149(88)90963-0.
To examine the effects of digital image acquisition mode and subtraction techniques on the results of coronary stenosis quantitation, 100 discrete lesions from 45 patients undergoing routine diagnostic angiography were analyzed in each of 3 image types: direct on-line digital, electrocardiogram-gated digital subtraction and digitized cine film images. For the geometric measurements (minimal lumen diameter and percent diameter stenosis) correlation coefficients for 2-way comparisons among the image types ranged from 0.90 to 0.96. Linear regression slopes ranged from 0.93 to 1.00, with intercepts from 0.03 to 0.07 mm for minimal diameter and -0.5 to 4.4% for percent diameter stenosis. For the videodensitometric percent area stenosis data, the correlation coefficients ranged from 0.80 to 0.89, with linear regression slopes from 0.84 to 0.89 and intercepts from 8.3 to 12.8%. Thus, the results of quantitative geometric measurements of coronary stenosis severity were not strongly affected by image acquisition mode (on-line versus cine film digitization) or by electrocardiogram-gated digital subtraction, while densitometric data correlated less well when on-line digital and digitized cine film acquisition methodology were compared.
