An algorithm for noninvasive identification of angiographic three-vessel and/or left main coronary artery disease in symptomatic patients on the basis of cardiac risk and electron-beam computed tomographic calcium scores
Axel Schmermund, MD*,
Kent R. Bailey, PhD ,
John A. Rumberger, PhD, MD, FACC*,
Judd E. Reed, BS ,
Patrick F. Sheedy, II, MD and
Robert S. Schwartz, MD, FACC*
* Division of Cardiovascular Diseases and Internal Medicine, Rochester, Minnesota, USA
Section of Biostatistics, Mayo Clinic and Foundation, Rochester, Minnesota, USA
Department of Information Services, Mayo Clinic and Foundation, Rochester, Minnesota, USA
Department of Diagnostic Radiology, Mayo Clinic and Foundation, Rochester, Minnesota, USA
View larger version (13K):
[in a new window]
|
Figure 1 Boxplot of the range of loge-transformed total calcium scores and calcium scores per major coronary artery in patients with ("1") or without ("0") angiographic three-vessel and/or left main CAD. The boxes represent 25th through 75th percentiles, and the lines in the boxes represent the median values. The whiskers show the smallest and largest observed values within 1.5 box-lengths of the 25th and 75th percentiles, respectively. All calcium scores shown in this figure, i.e., total and major coronary artery scores, discriminated between the presence versus the absence of three-vessel and/or left main disease (p < 0.0001 for all comparisons, except LM; p = 0.01). LAD = left anterior descending coronary artery, LCx = left circumflex coronary artery, LM = left main coronary artery, RCA = right coronary artery, total = complete coronary tree.
|
|
View larger version (28K):
[in a new window]
|
Figure 2 Receiver operating characteristic (ROC) curve analysis for separation of the presence or absence of angiographic three-vessel and/or left main disease. Comparison between noninvasive index (solid thick line), EBCT-derived parameters alone, i.e., separate calcium scores for the left anterior descending and left circumflex arteries (solid fine line), and risk factors alone (dotted fine line). The area under the ROC curves (mean ± SEM) were 0.88 ± 0.03 for the NI, 0.86 ± 0.03 for the EBCT parameters, and 0.77 ± 0.04 for the risk factors. Each was significantly different from a random distribution (dashed line).
|
|
View larger version (17K):
[in a new window]
|
Figure 3 Illustration of the power of the NI to rule in (a) or rule out (b) angiographic three-vessel and/or left main disease. Posttest probabilities of disease could not be calculated at the extreme ends of the noninvasive index, because sensitivities and specificities regarding the detection of angiographic three-vessel and/or left main disease were 100%. (a) The noninvasive index is represented on the X-axis and the posttest probability of severe angiographic disease after a positive test result on the Y-axis ((+)PTPD). The solid horizontal line represents the pretest probability of severe angiographic disease, i.e., 23%. Increasing values of the NI enhance the probability of severe angiographic disease (posttest) to levels above 65–70% (arrow). (b) The NI is represented on the X-axis and the posttest probability of no severe angiographic disease after a negative test result on the Y-axis ((–)PTPN). The solid horizontal line represents the pretest probability of no severe angiographic disease, i.e., 77%. Decreasing values of the NI enhance the probability of no severe angiographic disease (posttest) to levels above 95% (arrow).
|
|
|
