Assessing Intermediate Coronary Lesions: Angiographic Prediction of Lesion Severity on Intravascular Ultrasound

Intrascular ultrasound data grouped by minimum luminal area.
QCA data grouped by minimum luminal area.
Comparison of QCA and IVUS data.
Receiver operating characteristic curve for distal reference
segment diameter as a predictor of minimum luminal area ≤ 4 mm2.

*,§,£Marlos R. Fernandes, MD, MSc, §Guilherme V. Silva, MD, *Adriano Caixeta, MD, PhD,
*Miguel Rati, MD, £Nelson A. de Sousa e Silva, MD, PhD, §Emerson C. Perin, MD, PhD

Conventional angiography is the gold standard in clinical practice for diagnosing atherosclerotic compromise of the coronary artery tree. However, coronary angiography has several known limitations, including a lack of correlation between the percentage of stenosis and the lesion’s physiologic importance1 and considerable interobserver variability in classifying the lesion’s severity.2,3

Intravascular ultrasound (IVUS) has been shown to detect atherosclerotic compromise that cannot be detected by conventional angiography.4 Using tomographic slices to analyze the coronary artery, IVUS provides absolute luminal measurements. An IVUS-measured minimum luminal cross-sectional area (MLA) of ≤ 4.0 mm2 has been used as a marker of severe coronary artery stenosis, which correlates well with the findings of other methods for diagnosing myocardial ischemia, including single-photon emission computed tomography,5 Doppler wire studies6 and pressure wire measurement.7 The clinical importance of this criterion has been confirmed by a study8 showing that deferral of revascularization is safe for patients with an MLA of > 4.0 mm2.8

In this study, we investigated the rate of severe coronary artery stenosis (MLA ≤ 4.0 mm2) detected by IVUS in patients whose angiograms showed intermediate stenosis and who had evidence of myocardial ischemia. Because IVUS is more invasive, expensive and laborious than angiography, we also sought to identify angiographic factors that would predict an IVUS-measured MLA of ≤ 4.0 mm2. This could potentially minimize undue interventions and reduce cost, time and further coronary manipulation related to fractional or coronary flow reserve assessment.


Patients. This prospective, observational study was conducted at Rede D’or Hospitais (Rio de Janeiro, Brazil) and was approved by the institutional review board. Fifty-six patients, with 63 angiographically-intermediate coronary artery stenoses, underwent IVUS to further define the severity of their lesions. According to the American College of Cardiology and the American Heart Association, intermediate stenosis was defined as having a diameter of ≥ 30% and < 70%, as measured by quantitative coronary angiography (QCA) using CASSII software (Cardiovascular Angiographic Analysis System; Pie Medical, Maastricht, The Netherlands).9 All patients had clinical evidence of myocardial ischemia or functional test results indicative of ischemia in the territory of the culprit vessel. The clinical setting included stable angina, unstable angina and non-ST-elevation myocardial infarction.

The inclusion criteria were clinical history or functional test results indicating myocardial ischemia on stress testing, stress echocardiography or single-photon emission tomography, at least one intermediate lesion, as described above, and a de novo lesion in a native coronary artery. Patients were excluded if they had an ST-elevation acute myocardial infarction, anycontraindication to anticoagulation, or in-stent restenosis saphenous vein graft lesions.

Diagnostic methods. All analyses were performed offline. The diagnostic angiograms were obtained using Digital Imaging and Communications in Medicine (DICOM)-compatible digital systems (H-5000, Phillips Medical Systems, Eindhoven, Holland; and Siemens Medical Systems Coroskop Top digitalacquisition DICOM matrix 512 x 521).

For calibration of the QCA software, we used the final portion of the empty guiding catheter in the angiographic projection that showed the lesion best, with no foreshortening and with the most severe degree of stenosis. The software automat ically provided thr ee parameters: (1) percentage of stenosis; (2) minimum luminal diameter (MLD); and (3) lesion length. We also used the reference segment diameters for comparative analyses with the IVUS results. The reference segment was defined as the segment least affected by atherosclerosis within a 10 mm span proximal and distal to the target lesion. This definition was carefully followed for both angiography and IVUS to ensure that the measurements were done at exactly the same site for both diagnostic methods. We used the classification system from the National Heart, Lung, and Blood Institute’s Coronary Artery Surgery Study for lesion localization and definition of the proximal, medial and distal portions of the arteries.10

Prior to beginning the IVUS studies, patients were given 10,000 U of unfractionated heparin for systemic anticoagulation. IVUS was performed using conventional 6 or 7 Fr guiding catheters and a 0.014 mm guidewire was positioned distally. IVUS catheters of 30 or 40 MHz (UltraCross®, Discovery or Atlantis®; Boston Scientific Corp., Natick, Massachusetts) were pulled back automatically at a constant speed of 0.5 mm/second. The following parameters were included in the IVUS analyses: MLA, reference segment plaque burden, luminal area of stenosis, remodeling index, atheroma eccentricity (eccentricity index), reference segment luminal diameter and MLD. The measurements were performed according to the guidelines of the American College of Cardiology for the acquisition, measurement and reporting of IVUS studies.11 We defined positive remodeling as having a remodeling index above 1.05, and negative as having an index below 0.95.12

Statistical analyses. We used the NCSS statistical software package (Kaysville, Utah) for data analysis. The data were described for the whole population and further stratified into two groups based on their MLA. Group 1 included lesions with a MLA ≤ 4.0 mm2, and Group 2 included lesions with a MLA > 4.0 mm2.

Continuous data with normal distribution were described as means ± standard deviation (SD), and categorical data were described as frequency among the two groups. For inferential analyses, comparisons of numerical variables between both groups were performed using the two-tailed unpaired t-test or the Mann-Whitney U-test for parametric and nonparametric variables, respectively. Comparisons of categorical data between groups were performed using the chi-squared or Fisher’s exact tests. Multiple logistic regression was performed using QCA-derived parameters to assess for angiographic predictors of severe luminal stenosis (MLA ≤ 4 mm2).


Four patients with 1 lesion each were excluded from the study, as 3 of the lesions had poor-quality IVUS data and 1 had a poor-quality angiogram. The final group was composed of 52 patients with 59 intermediate lesions. The average age of the patients was 61 ± 13 years, and 28 (54%) were men. Twenty-five patients (48%) had dyslipidemia, 14 (27%) were smokers, 32 (62%) had high blood pressure, 9 (17%) had diabetes and 22 (42%) had a family history of coronary artery disease. Thirty patients (58%) had acute coronary syndrome (TIMI RISK > 3 for unstable angina or non-ST-elevation MI) and underwent coronary angiography as part of an early invasive strategy. The remaining 22 patients had chronic stable angina (n = 14) or chest pain of recent onset (n = 8) and were submitted to coronary angiography due to the presence of myocardial ischemia during stress testing. Two of these patients showed ECG changes during treadmill testing; 12 had perfusion defects in the territory of the culprit lesion on technetium Tc 99m sestamibi (MIBI) SPECT, and 8 showed akinesia on stress dobutamine echocardiograms.

The IVUS data and lesion characteristics are summarized in Table 1. There was a high frequency of severe stenosis (MLA ≤ 4.0 mm2), as assessed by IVUS. Thirty-seven patients (71%) had at least 1 severe stenosis, and these stenoses accounted for 40 (68%) of the lesions in the overall group. Most of the lesions were eccentric and showed negative remodeling. The Group 1 lesions were more severe, as indicated by a higher percentage of luminal area stenosis and higher plaque burden. An important reference segment compromise was evident in the overall group, as shown by the average plaque burden of 33.67 ± 15.01% at these sites.

Table 2 describes the angiographic analyses. Fifty-three (90%) of the 59 lesions were located at the proximal or midportions of the artery (35 of 40 in Group 1 and 18 of 19 in Group 2). The Group 1 lesions were longer and had lower MLD than did the Group 2 lesions. Although the severity of stenosis was higher among Group 1 lesions, none of the lesions showed more than 70% stenosis.

A comparison of the IVUS and angiography results revealed a significant underestimation of the reference segment luminal diameter by angiography, even though the MLD measurements by the two methods were similar (Table 3). Linear regression analysis showed a weak correlation between the two methods for the assessment of reference segment luminal diameter (r = 0.4; p < 0.001).

Multiple logistic regression was performed to identify predictors of a MLA ≤ 4 mm2. The analysis included the QCA derived parameters of proximal reference diameter, interpolated reference diameter, distal reference diameter, MLD and percentage of stenosis. The only predictor of small MLA was the diameter of the distal reference segment. Figure 1 shows theresults of a receiver operating characteristic curve using the QCA-derived distal reference diameter to predict a MLA ≤ 4.0 mm2. A diameter < 2.42 mm was predictive of a small MLA on IVUS, with a sensitivity of 90% and a specificity of 55%. Conversely, a diameter of 3.25 mm excluded a small MLA with high specificity (90%), although the sensitivity was low (26%).


In this group of patients, 68% of the lesions (in 71% of the patients) diagnosed as moderate by conventional angiography had, in fact, severe stenosis by the adopted IVUS criterion (MLA ≤ 4.0 mm2).

Atherosclerotic involvement of the arterial wall in apparently normal coronary segments seems to be a ubiquitous phenomenon.4 The clinical importance of coronary narrowing has been questioned, since the majority of acute myocardial infarctions originate from previously nonobstructive lesions.13–20 This concept of lesion vulnerability has changed the clinical approach to diagnosing and treating coronary artery disease, since the odds of an adverse event are not directly related to the severity of coronary artery stenosis.21,22 For example, revascularization of stenotic lesions in clinically stable patients might not affect long-term survival.23–27 Despite its sectional design, the present study confirms the above concept, as the frequency of lesions with MLA ≤ 4.0 mm2 was greater than that of lesions with MLA > 4.0 mm2 in patients with stable angina (35% versus 6.7%; p = 0.04), but not in those with unstable angina (56.8% versus 60%). Furthermore, in the clinical setting of stable angina, 13 of 14 patients (93%) had a MLA ≤ 4.0 mm2.

Post new comment

  • Web page addresses and e-mail addresses turn into links automatically.
  • Lines and paragraphs break automatically.

More information about formatting options

Enter the characters shown in the image.