Lesion morphology poorly predicts functional significance of intermediate coronary artery stenosis. The aim of this study was to determine whether a coronary artery–based myocardial segmentation method that quantifies subtended myocardium can improve the diagnostic accuracy of intravascular ultrasound (IVUS)–derived parameters for detecting ischemia-producing lesions. Coronary computed tomography angiography, IVUS, and fractional flow reserve (FFR) data were analyzed in 101 non-left main lesions (20% to 80% angiographic stenosis). Using the coronary artery–based myocardial segmentation method, total left ventricular myocardial volume ( V total ), myocardial volume subtended by the stenotic coronary segment ( V sub ), and V ratio (the ratio of the V sub to the V total ) were assessed. Both V sub >30.7 cm 3 and V ratio >25.4% were determinants of FFR ≤0.75 (area under the curve = 0.696 and 0.744). Overall, an IVUS-measured minimum lumen area (IVUS-MLA) ≤2.83 mm 2 predicted FFR ≤0.75 with a sensitivity 88% and specificity 73%. Among lesions with IVUS-MLA ≤2.83 mm 2 and FFR >0.75, 89% showed V sub <30.7 cm 3 . In 50 lesions with V sub >30.7 cm 3 , an IVUS-MLA ≤2.85 mm 2 predicted FFR ≤0.75 with sensitivity 85%, specificity 92%, positive predictive value 92%, and negative predictive value 85%. Conversely, in 51 lesions with a V sub ≤30.7 cm 3 , IVUS-MLA ≤2.67 mm 2 showed sensitivity 100%, specificity 69%, positive predictive value 38%, and negative predictive value 100% for predicting FFR ≤0.75. Body surface area, reference lumen diameter, and vessel area had modest correlations with V sub . In those lesion subsets, IVUS-MLA ≈2.8 mm 2 accurately predicted an FFR ≤0.75, whereas the clinical relevance of assessing and treating lesions with a smaller myocardial territory may be limited ( ClinicalTrials.gov number NCT1696006 ).
Although intravascular ultrasound (IVUS) is generally used to assess coronary lesion morphology and disease severity, its role in the clinical decision to treat or not to treat non-left main lesions has been limited. Because the hemodynamic significance is determined not only by the degree of stenosis but also by the size of the myocardial territory, using the IVUS-measured minimal lumen area (IVUS-MLA) as a single anatomical parameter, is inaccurate for identifying an ischemia-producing lesion with a fractional flow reserve (FFR) ≤0.75 to 0.80. In addition, FFR is determined by many clinical and local factors, such as age, gender, body surface area, lesion length, left anterior descending artery (LAD) involvement, lesion location, and vessel size. Most of these factors reflect the size of the supplied myocardium; it remains challenging to quantitatively assess the amount of the subtended myocardium. A semi-automated, coronary artery–based myocardial segmentation (CAMS) method was recently developed to quantify the myocardium subtended by a specific stenotic coronary segment. The aims of this study evaluating intermediate non-left main coronary artery lesions were to assess the impact of CAMS-derived indexes of myocardial size on the hemodynamic significance of individual lesions and to determine how best to use CAMS to improve the diagnostic accuracy of IVUS-MLA in predicting an FFR ≤0.75.
Methods
From February 2012 to November 2014, 816 patients were consecutively enrolled in the Perfusion CT Registry ( ClinicalTrials.gov , number NCT1696006 ). Invasive coronary angiography, IVUS, and FFR measurements were conducted at the Asan Medical Center (Seoul, Korea) in 139 patients who had at least 1 coronary artery with an intermediate stenosis (angiographic diameter stenosis [DS], 20% to 80% on visual estimation). The exclusion criteria were in-stent restenosis, left main coronary artery or side branch lesions, Thrombolysis In Myocardial Infarction flow <3, thrombus-containing lesions, tandem lesions, and cases in which the IVUS imaging catheter or FFR guidewire failed to cross the lesion. In addition, patients with any history of myocardial infarction, echocardiographic regional wall motion abnormality or ventricular wall thinning, left ventricular hypertrophy on echocardiography, and a coexistent left main coronary artery stenosis >30 and those with poor-quality images were excluded. Thus, a total of 72 patients (including 101 coronary arteries with 101 intermediate lesions) assessed by complete angiography, FFR, and IVUS were included in the current analysis. The study was approved by an institutional review committee and that the subjects gave informed consent.
“Equalizing” was performed with the guidewire sensor positioned at the guiding catheter tip. A 0.014-inch FFR pressure guidewire (Radi; St. Jude Medical, Uppsala, Sweden) was then advanced distal to the stenosis. The FFR was measured at the maximum hyperemia induced by an intravenous infusion administered through a central vein of adenosine at 140 μg/kg/min and increased to 200 μg/kg/min.
Quantitative coronary angiography (QCA) was performed using standard techniques with automated edge-detection algorithms (CAAS-5; Pie-Medical, Maastricht, The Netherlands) in the angiographic analysis center of the CardioVascular Research Foundation (Seoul, Korea). Angiographic DS, minimal lumen diameter, lesion length, and the lumen diameters of the proximal and distal reference segments were measured.
After FFR assessment and intracoronary administration of nitroglycerin (0.2 mg), IVUS imaging was performed using motorized transducer pullback (0.5 mm/s) and a commercial scanner (Boston Scientific Scimed, Inc., Minneapolis, Minnesota) with a rotating 40 MHz transducer within a 3.2-French imaging sheath. Using computerized planimetry (EchoPlaque 3.0; Indec Systems, Mountain View, California), offline quantitative IVUS analysis was performed in a core laboratory at the Asan Medical Center. The proximal and distal reference segments were within 5 mm of the lesion. The proximal and distal reference external elastic membrane (EEM) and lumen areas were measured. Plaque burden at the MLA site was calculated as (EEM area – MLA) divided by EEM area.
Perfusion imaging was performed using second-generation dual-source computed coronary tomography (Definition Flash, Siemens, Germany). Data with fewest motion artifacts and clearest demarcation of coronary artery segments were transferred to customized software for CAMS analysis (A-View Cardiac, Asan Medical Center, Korea). After extracting the centerline of each coronary artery and the left ventricular myocardium on the computed tomographic images, a 3-dimensional Voronoi algorithm was used to assign the myocardial territories to the 3 major epicardial coronary arteries with the CAMS program. Briefly, the Voronoi algorithm is a mathematical algorithm that divides the area or space between predetermined points or lines according to the shortest distances from those points or lines. The left ventricular myocardial volume ( V total ) was divided into 3 major epicardial coronary artery territories based on the shortest distance from the coronary artery. The V sub was defined as the volume of the myocardium subtended by the stenotic coronary segment. The V ratio was defined as the ratio of the V sub to the V total . Figure 1 shows examples of angiographic, IVUS, and CAMS analysis. Computed tomography, angiography, and IVUS image analyses were performed by observes blinded to the other data.
Using SPSS (version 10.0; SPSS Inc., Chicago, Illinois), all values are expressed as the means ± 1 SD (continuous variables) or as counts and percentages (categorical variables). Continuous variables were compared by unpaired t tests; categorical variables were compared by the chi-square statistics or Fisher’s exact test. Receiver-operating characteristic curves were analyzed using MedCalc Software (Mariakerke, Belgium) to assess the best cutoff for the morphologic parameters to predict an FFR ≤0.75 with maximal accuracy. Multivariable regression analysis was performed to identify the independent determinants for FFR as a continuous variable and an FFR ≤0.75. A p value <0.05 was considered statistically significant.
Results
The baseline clinical characteristics and CAMS data in 72 patients are summarized in Table 1 . Compared with women, men had a larger body surface area (1.81 ± 0.14 m 2 vs 1.58 ± 0.12 m 2 , p = 0.001) and a greater V total (123.1 ± 25.1 cm 3 vs 97.7 ± 24.5 cm 3 , p = 0.009). Overall, 49 patients (68%) had LAD disease.
| Clinical characteristics | |
| Age (years) | 62.5±9.2 |
| Men | 64 (89%) |
| Diabetes mellitus | 26 (36%) |
| Hypertension ∗ | 33 (46%) |
| Current smoker | 35 (49%) |
| Hyperlipidemia † | 21 (29%) |
| Body mass index (kg/m 2 ) | 25.1±2.6 |
| Stable angina pectoris | 51 (71%) |
| Unstable angina pectoris | 21 (29%) |
| CAMS data | |
| Left ventricular myocardial volume (cc) | 120.3±26.1 |
| Myocardial volume of right coronary artery territory (cc) | 32.1±11.9 |
| Myocardial volume of left anterior descending artery territory (cc) | 52.1±14.8 |
| Myocardial volume of left circumflex artery territory (cc) | 34.1±11.5 |
| %Myocardium of right coronary artery territory (cc) | 27.2±8.2 |
| %Myocardium of left anterior descending artery territory (cc) | 43.9±7.3 |
| %Myocardium of left circumflex artery territory (cc) | 28.8±7.9 |
∗ Defined as receiving antihypertensive treatment, or having systolic blood pressure ≥140 mm Hg or diastolic blood pressure of ≥90 mm Hg.
† Defined as total cholesterol >200 mg/dl, or receiving anti-lipidemic treatment.
The QCA, IVUS, FFR, and CAMS data in 101 intermediate coronary lesions are summarized in Table 2 . The MLA was located in proximal and mid-segments in 81 lesions (81%). Table 3 lists CAMS data based on lesion location. Table 4 lists correlations between morphologic factors and CAMS-derived myocardial volumes.
| Quantitative coronary angiographic data | |
| Proximal reference lumen diameter (mm) | 3.2±0.5 |
| Distal reference lumen diameter (mm) | 2.9±0.5 |
| Averaged reference lumen diameter (mm) | 3.1±0.5 |
| Minimal lumen diameter (mm) | 1.7±0.6 |
| Diameter stenosis (%) | 45.0±16.7 |
| Lesion length (mm) | 16.8±9.1 |
| Intravascular ultrasound data | |
| Proximal reference lumen (mm 2 ) | 10.5±4.7 |
| Proximal reference external elastic membrane (mm 2 ) | 18.8±6.5 |
| Minimal lumen area (mm 2 ) | 3.5±2.2 |
| External elastic membrane at the minimal lumen area site (mm 2 ) | 12.9±4.5 |
| Plaque+media at the minimal lumen area site (mm 2 ) | 9.4±4.4 |
| Plaque burden at the minimal lumen area site (%) | 70.9±14.7 |
| Distal reference lumen (mm 2 ) | 7.0±3.5 |
| Distal reference external elastic membrane (mm 2 ) | 11.7±5.1 |
| FFR data | |
| FFR at maximal hyperemia | 0.81 ± 0.15 |
| FFR ≤0.75 | 45 (45%) |
| CAMS data | |
| V sub (cc) | 31.7±14.3 |
| V ratio (%) | 26.9±10.8 |
| Involved coronary segment | Lesion No | Myocardial volume subtended to post-stenotic segment (cc) | %Myocardium subtended to post-stenotic segment (%) | ||
|---|---|---|---|---|---|
| Mean ± SD | >30.7cc, N (%) | Mean ± SD | >25.4%, N (%) | ||
| Proximal right | 11 | 28 ±14 | 6 (43%) | 24±11 | 4 (29%) |
| Mid right | 17 | 27±10 | 6 (30%) | 23±6 | 7 (30%) |
| Distal right | 6 | 27±9 | 2 (22%) | 265±8 | 4 (44%) |
| Proximal left anterior descending | 27 | 44±12 | 27 (90%) | 38±8 | 29 (97%) |
| Mid left anterior descending | 22 | 36±12 | 19 (60%) | 30±8 | 20 (63%) |
| Proximal left circumflex | 4 | 26±11 | 4 (36%) | 22±8 | 4 (36%) |
| Distal left circumflex | 12 | 20±15 | 5 (23%) | 16±8 | 2 (9%) |
| Ramus intermedius | 2 | 23±2 | 0 (0%) | 20±2 | 0 (0%) |
| V total | V sub | V ratio | ||||
|---|---|---|---|---|---|---|
| 72 Patients | 101 Lesions | 101 Lesions | ||||
| r | p value | r | p value | r | p value | |
| Age | -0.149 | 0.211 | -0.202 | 0.016 | -0.130 | 0.124 |
| Body surface area | 0.614 | <0.001 | 0.476 | <0.001 | 0.153 | 0.071 |
| Body mass index | 0.331 | 0.005 | 0.145 | 0.088 | -0.043 | 0.610 |
| Averaged reference lumen diameter, mm ∗ | 0.326 | <0.001 | 0.245 | 0.003 | ||
| Averaged reference external elastic membrane area, mm 2 | 0.190 | 0.068 | 0.150 | 0.151 | ||
| Minimal lumen area | -0.107 | 0.309 | -0.053 | 0.613 | ||
| External elastic membrane at minimal lumen area site, mm 2 | 0.350 | 0.001 | 0.379 | <0.001 | ||
| Plaque burden at minimal lumen area site, % | 0.333 | 0.001 | 0.300 | 0.003 | ||
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
