Impact of vascular remodeling on the coronary plaque compositions: An investigation with in vivo tissue characterization using integrated backscatter-intravascular ultrasound

Atherosclerosis 202 (2009) 476–482

Impact of vascular remodeling on the coronary plaque compositions: An investigation with in vivo tissue characterization using integrated backscatter-intravascular ultrasound Hiroki Takeuchi, Yoshihiro Morino ∗ , Takashi Matsukage, Naoki Masuda, Yota Kawamura, Satoshi Kasai, Tadashi Hashida, Daisuke Fujibayashi, Teruhisa Tanabe, Yuji Ikari Division of Cardiology, Tokai University School of Medicine, 143 Shimokasuya, Isehara 259-1193, Japan Received 30 January 2008; received in revised form 23 May 2008; accepted 25 May 2008 Available online 5 June 2008

Abstract Recent studies have indicated that positive remodeling is strongly associated with development of acute coronary syndrome (ACS). The aim of this study was to compare plaque composition of vascular remodeling patterns by an established in vivo tissue characterization method using integrated backscatter (IB)-intravascular ultrasound (IVUS). The study population consisted of 41 consecutive patients who received IVUS prior to percutaneous coronary intervention. Remodeling index (RI) was calculated as the external elastic membrane (EEM) area at the minimal lumen area (MLA) site divided by average EEM area at the proximal and distal reference sites. The patients were divided into two groups based on RI: positive remodeling (PR) defined as RI > 1 and non-PR as RI ≤ 1. A total of 21 areas centered at MLA per lesion site were evaluated by IB-IVUS at 1 mm intervals. The occupancy rate of four tissue types within atherosclerotic plaques was compared between the two groups. Percent lipid volume in the PR group (n = 20) was significantly greater than the non-PR group (n = 21) (40.5 ± 14.8% vs. 26.4 ± 15.9%, p < 0.001). In contrast, % fibrous volume in the PR group was significantly lower than the non-PR group (49.9 ± 9.4% vs. 56.1 ± 9.6%, p = 0.042). Percent dense fibrous volume and % calcified volume were slightly but significantly lower in the PR group compared with the non-PR group (dense fibrous: 6.8 ± 5.0% vs. 11.6 ± 8.4%, p = 0.034, calcified: 2.6 ± 2.0% vs. 5.1 ± 4.4%, p = 0.026). In conclusions, PR lesions contain more lipid-rich and less hard plaque components compared with non-PR lesions, which may account for the higher incidence of ACS and plaque vulnerability. © 2008 Elsevier Ireland Ltd. All rights reserved. Keywords: Vascular remodeling; Coronary disease; Plaque vulnerability; Integrated backscatter-intravascular ultrasound

Positive remodeling (PR) has been considered as a “compensatory process” to maintain a functional size of lumen as a safeguard against narrowing due to atherosclerotic progression with plaque accumulation [1]. Despite structural compensation, these vascular walls may accommodate biologically active components, associated with acute coronary syndrome (ACS). The primal pathogenesis of ACS has been regarded as rupture or erosion of atherosclerotic plaque and subsequent thrombus formation. Initially, the issues of vascular remodeling and plaque vulnerability was mainly studied by examining the tissue pathology [2,3]. ∗

Corresponding author. Tel.: +81 463 93 1121; fax: +81 463 93 6679. E-mail address: [email protected] (Y. Morino).

0021-9150/$ – see front matter © 2008 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.atherosclerosis.2008.05.052

After the emergence and introduction of intravascular ultrasound (IVUS), which potentially allows visualization of underlying plaque and accurate measurements of lumen and vessel area, unlike angiography, the presence or absence of remodeling response can easily be assessed from the clinical standpoints [4]. Accordingly, several IVUS studies have proved that vascular PR is significantly associated with the development of ACS [3,5,6]. In other words, one hypothesis is that positively remodeled coronary vessels accumulate more lipid-rich components within the vascular wall than non-positively remodeled vessels. However, few studies have been done to prove this hypothesis because of the lack of tissue characterization capability in the conventional gray-scale IVUS approach. Recently, a novel in vivo

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

tissue characterization system, integrated backscatter (IB)IVUS, was developed. This equipment allows qualitative plaque assessment compatible with tissue-specific quantification [7], which can be potentially utilized for the verification of the described hypothesis. Therefore, the aim of this study was (1) to compare composition of plaques based on vascular remodeling patterns using IB-IVUS and (2) to clarify the basic pathogenesis of vascular remodeling by characterizing plaque components.

1. Methods 1.1. Subject and study design Study population consisted of consecutive patients with stable angina who underwent elective percutaneous coronary intervention (PCI) from September 2006 to February 2007 at Tokai University School of Medicine. The diagnosis of stable angina was based on the presence of chest pain on effort that persisted more than 2 months, a positive stress test by ECG and/or nuclear medicine study, and significant diameter stenosis of more than 75% in at least one coronary artery on coronary angiography (CAG). Patients were excluded if they had lesions with chronic total occlusion (CTO) and presence of intimal calcification (arc. > 60◦ ) in the entire stenotic segments or poor image quality at IVUS imaging. Pre-interventional IVUS images were obtained from 59 consecutive patients. Eighteen patients, composed of 10 patients with CTO, 4 patients with severe intimal calcification, and 4 patients with poor image quality, were excluded from this study. Accordingly, a total of 41 patients that met the inclusion and exclusion criteria were studied. Informed consent was obtained from each patient before participation in the study. 1.2. Integrated backscatter system presets and imaging procedure We used a commercially available IVUS imaging system to create ultrasound tissue images of coronary arteries (Galaxy2TM , Boston Scientific, Natick, Massachusetts) and a 40-MHz intravascular catheter. Because we obtained radio frequency signal output, a personal computer with custom software (IB-IVUS, YD Co., Ltd., Nara, Japan) was connected to the IVUS imaging system. All patients were premedicated with aspirin (200 mg) and ticlopidine (200 mg) and received heparin (100 U/kg). The IVUS catheter was introduced into the culprit lesion through a 6–8 Fr coronary guiding catheter over a guidewire. To prevent coronary spasm, an intracoronary optimal dose of isosorbide dinitrate was injected prior to IVUS imaging. The transducer was pulled back automatically at 0.5 mm/s to perform the imaging sequence, which started at least 5 mm distal to lesion and ended at the aorto-ostial junction.

477

A single frame of each IB-IVUS image was divided into picture elements (minimum units for evaluation of IB-value) in the following sequence: (1) the image was divided into 256 vector lines (1.4 grade/line) and (2) 71 regions of interest were defined for each 50-␮m depth on each vector line. Accordingly, volumetric elements were expressed approximately as 2πr/256 × 0.05 mm × 1.0 mm (r = distance from the center of IVUS catheter). IB values were calculated as the average power of the ultrasound backscattered signals from a small volume of tissue. For that purpose, fast Fourier transform was used and measured in decibels. It was reported that the attenuation for the distance from lumen-intima border to media-adventitia border using a 40-MHz frequency catheter was 5.9 dB/mm. Therefore, we corrected each IB value by adding 0.59 dB/0.1 mm [7–9]. 1.3. Measurement of conventional IVUS parameter and remodeling index Quantitative measurements were obtained offline with an IB-IVUS computer assisted analysis system. First, we identified the target vessel and lesion, considered to be responsible for the angina presentation of the patient. Then, we identified the culprit lesion site and the proximal and distal reference sites. The culprit lesion site selected for measurement was the image slice with the smallest lumen cross-sectional area (CSA). The proximal and distal lesions were chosen as the most-normal-looking image slices within 10 mm proximal or distal from the lesion segment. For each culprit lesion and each proximal and distal reference site pair, external elastic membrane (EEM) CSA were manually measured by tracing the external edge of the ultrasonic media-adventitia border. Also, lumen CSA was measured, by tracing the leading edge of the lumen-intima border. The plaque CSA was calculated as EEM CSA minus lumen CSA. The plaque burden was calculated as plaque CSA/EEM CSA × 100 (%). The area of stenosis was calculated as lumen CSA/EEM CSA × 100 (%). For the volumetric analysis, a 20-mm-long lesion was identified, centered at the minimal lumen area (MLA) site. Lumen volume, EEM volume, and plaque volume were calculated using the Simpson’s rule. Remodeling index (RI) was calculated as the culprit lesion site EEM CSA divided by the average of the proximal and distal reference site EEM CSA as described previously [10]. PR was defined as a RI > 1. In case of RI ≤ 1, we exclusively define it as non-PR. 1.4. IB-IVUS images analysis Two-dimension (2D) color-coded maps using IB values were constructed in 21 consecutive IVUS image slices at 1-mm intervals, centered at the most severe stenotic site. The quantitative IB-IVUS analysis was performed by an experienced physician unaware of the clinical data. The slices were excluded for analysis if intimal calcification was present (arc. > 60◦ ) or a coronary bifurcation existed. The percentage of four tissue parameters under investigation

478

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

(lipid, fibrous, dense fibrous, and calcified) was automatically calculated by IB-IVUS system after exact manual tracing. Because the media of coronary artery always presents a low echoic band, potentially identified with “lipid” by the current algorithm of IB-IVUS, the internal border of media (internal elastic membrane, IEM) was traced manually to eliminate the misjudgment as described in previous studies [7,11]. The acoustic shadows of the guidewire and calcifications were manually traced and excluded to minimize acoustic artifacts. For 2D analysis, the cross-section of the tightest lumen was selected as the MLA segment. Percent lipid area, % fibrous area, % dense fibrous area, and % calcified area were also determined at the MLA site. Furthermore, the percent content of each category within the 20-mm-long segments was obtained as % lipid volume, % fibrous volume, % dense fibrous volume, and % calcified volume. 1.5. Statistical analyses Statistical analysis was performed with SPSS, version 11.0 for Windows (SPSS Inc., Chicago, Illinois). Data were reported as mean ± standard deviation (S.D.). Continuous

variables were compared with unpaired t-tests. Otherwise, a Mann–Whitney U-test was used. p-Values <0.05 were considered statistically significant.

2. Results 2.1. Overall results Out of 41 patients with stable angina, 20 patients (49%) demonstrated PR on IVUS and 21 patients (51%) were found to have intermediate or negative remodeling (non-PR). The two groups were similar in characteristics of age, sex, coronary risk factors, types of lesion, medication, and blood lipid levels (Table 1). 2.2. Conventional IVUS measurements The quantitative 2D IVUS measurements at the MLA site and the reference sites are shown in Table 2. At the culprit lesion site, the lumen CSA was similar between the two groups. However, plaque CSA, EEM CSA, and plaque burden of the PR group were statistically significantly greater

Table 1 Baseline clinical characteristics PR (n = 20)

Non-PR (n = 21)

p

Age (years) Male gender, n (%)

64.9 ± 10.3 14 (70)

67.3 ± 10.3 17 (81)

0.442 0.429

Clinical history, n (%) Myocardial infarction Previous coronary bypass graft Hypertension Hyperlipidemia Current smoker Diabetes mellitus type 2

5 (25) 1 (5) 16 (80) 16 (80) 5 (25) 7 (54)

8 (38) 0 (0) 18 (86) 19 (90) 2 (10) 6 (46)

0.379 0.312 0.638 0.360 0.197 0.668

Number of coronary artery disease, n (%) 1 vessel 2 vessel 3 vessel

9 (45) 5 (25) 6 (30)

5 (24) 10 (48) 6 (29)

0.160 0.140 0.922

Target plaque location, n (%) LAD LCX RCA

8 (40) 5 (25) 7 (35)

7 (33) 6 (29) 7 (33)

0.668 0.802 0.913

Medication, n (%) Statins Anti-platelet agents ACE inhibitors AT1 antagonists Calcium channel blockers Beta-blockers Nitrates

10 (50) 20 (100) 9 (45) 7 (35) 11 (55) 11 (55) 7 (35)

14 (67) 21 (100) 10 (48) 10 (48) 6 (29) 15 (71) 8 (38)

0.291 1.000 0.871 0.425 0.091 0.288 0.842

Blood lipid levels (mg/dl) Total cholesterol Triglycerides HDL cholesterol LDL cholesterol

185.5 ± 31.1 149.7 ± 89.4 53.6 ± 15.2 109.9 ± 24.4

182.0 ± 40.4 167.2 ± 85.3 49.7 ± 17.5 111.0 ± 33.5

0.763 0.523 0.452 0.909

Values are n (%) or mean ± S.D. PR, positive remodeling; LAD, left anterior descending coronary artery; LCX, left circumflex artery; RCA, right coronary artery; ACE, angiotensin-converting enzyme; HDL, high-density lipoprotein; LDL, low-density lipoprotein.

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

479

Table 2 Intravascular ultrasound measurements PR (n = 20) Two-dimensional intravascular ultrasound measurements Remodeling index 1.23 Minimal lumen area site Lumen CSA (mm2 ) 3.1 12.0 Plaque CSA (mm2 ) 15.0 EEM CSA (mm2 ) Plaque burden (%) 77.9 Area stenosis (%) 22.1

Non-PR (n = 21)

p

± 0.14

0.82 ± 0.11

<0.001

± ± ± ± ±

1.5 4.7 4.9 10.2 10.2

2.5 5.5 8.0 66.7 33.3

± ± ± ± ±

1.1 3.3 3.8 10.5 10.5

0.170 <0.001 <0.001 0.001 0.001

Proximal reference Lumen CSA (mm2 ) Plaque CSA (mm2 ) EEM CSA (mm2 ) Plaque burden (%) Area stenosis (%)

8.0 5.0 13.1 40.2 59.8

± ± ± ± ±

3.6 2.0 4.7 12.5 12.5

6.3 5.2 11.6 45.4 54.4

± ± ± ± ±

3.3 2.8 5.9 10.5 10.5

0.117 0.808 0.372 0.158 0.145

Distal reference Lumen CSA (mm2 ) Plaque CSA (mm2 ) EEM CSA (mm2 ) Plaque burden (%) Area stenosis (%)

6.5 3.8 10.3 36.4 63.6

± ± ± ± ±

2.8 2.3 4.6 9.9 9.9

5.6 4.1 9.7 40.4 59.6

± ± ± ± ±

3.6 3.1 6.2 13.7 13.7

0.365 0.674 0.743 0.292 0.291

94.1 5.2 81.2 4.6 175.3 9.9

± ± ± ± ± ±

60.6 3.0 51.4 2.5 107.5 5.2

0.333 0.435 0.005 0.003 0.060 0.062

Three-dimensional intravascular ultrasound measurements (20-mm-long segment) Lumen volume (mm2 ) 112.6 ± 60.4 Lumen average area (mm2 ) 5.9 ± 2.7 125.2 ± 43.7 Plaque volume (mm3 ) 6.7 ± 1.6 Plaque average area (mm2 ) EEM volume (mm3 ) 237.8 ± 99.5 12.6 ± 4.1 EEM average area (mm2 )

Values are mean ± S.D. PR, positive remodeling; CSA, cross-sectional area; EEM, external elastic membrane.

than that of the non-PR group. Area stenosis of the PR group was statistically significantly smaller than the non-PR group. At both the proximal and distal reference sites, no significant differences were observed between the two groups in lumen CSA, plaque CSA, EEM CSA, plaque burden, and area of stenosis. In addition, three-dimensional (3D) IVUS measurements (20-mm-long segment) are shown in Table 2. Lumen volume was similar between the PR group and the non-PR group. There were no significant differences between the two groups with respect to EEM volume; however, EEM volumes of the PR group tended to be larger than that of the nonPR group. Plaque volume of the PR group was statistically significantly greater than that of the non-PR group.

(dense fibrous; 5.9 ± 6.3% vs. 12.8 ± 8.8%, p = 0.016, calcified; 1.8 ± 2.0% vs. 4.5 ± 4.0%, p = 0.012). Interestingly, similar results were obtained from the 3D analyses of the 20-mm-long lesion. Specifically, % lipid volume in the PR group was significantly greater than that in the non-PR group (40.5 ± 14.8% vs. 26.4 ± 15.9%, p = 0.006). In contrast, % fibrous volume in the PR group was significantly lower than that in the non-PR group (49.9 ± 9.4% vs. 56.1 ± 9.6%, p = 0.042). The percentages of the other categories of lesion characteristics in the PR group were significantly lower than those in the non-PR group (dense fibrous: 6.8 ± 5.0% vs. 11.6 ± 8.4%, p = 0.034; calcified: 2.6 ± 2.0% vs. 5.1 ± 4.4%, p = 0.026) (Fig. 2).

2.3. Qualitative and quantitative analysis by IB-IVUS

2.4. Investigation of uniformity of the tissue component around stenotic and neighboring segments

We compared each categorical IB values of the PR group to that of the non-PR group from 2D and 3D IVUS perspectives. Representative cases of PR and non-PR are shown in Fig. 1. At 2D analyses (MLA site), % lipid area in the PR group was significantly greater than that in the non-PR group (49.1 ± 19.1% vs. 26.9 ± 15.4%, p < 0.001); % fibrous area in the PR group was significantly less than that in the non-PR group (43.2 ± 13.6% vs. 56.7 ± 9.3%, p < 0.001). Likewise, the percentages of the other categories in the PR group were significantly lower than those in the non-PR group

The percent contents of each category were compared between 2D (MLA site) and 3D (20-mm-long lesion) analysis. Importantly, % lipid area by 2D analysis was strongly correlated with % lipid volume by 3D analysis (Fig. 3). This correlation applied to both the PR group (r = 0.891, p < 0.891) and the non-PR group (r = 0.905, p < 0.001). Similarly, a strong correlation of the fibrous tissue ratio was found between the 2D and 3D assessments (PR group: r = 0.807, p < 0.001; non-PR group: r = 0.818, p < 0.001).

480

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

Fig. 1. Representative cases of positive remodeling (PR) (upper) and negative remodeling (non-PR) (lower). Basic quantitative IVUS parameters and the corresponding images of grayscale IVUS and IB-IVUS at the minimal lumen area site are demonstrated.

3. Discussion The overall results in this study can be summarized as follows: (1) the PR group had more plaque burden, (2) there

was greater lipid content and less fibrous, dense fibrous and calcified tissue components in the PR group, and (3) plaque composition was approximately uniformly distributed in the adjacent segments in both groups. These results provide fun-

Fig. 2. Percent content of each histological category at a cross-section of minimal lumen area site (2D) and within a 20-mm-long lesion (3D). Two major components are lipid and fibrous tissue in the human coronary atherosclerotic lesions, whereas mixed and calcified tissue appears less frequently. Furthermore, the distributions of lipid and fibrous tissue differ between the PR group and the non-PR group. Percent content of each plaque component.

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

481

Fig. 3. Correlation of % lipid area at the MLA site with % lipid volume within a 20-mm-long lesion. A strong positive correlation was observed between these two measurements. Similar results were obtained in the comparison of % fibrous area and % fibrous volume. Correlation of % lipid content between 2D and 3D analysis.

damentally important insight into the relationship between plaque stability and vascular remodeling. “Vascular remodeling” is not merely a pathogenic concept, because this process has been proven from the clinical standpoint to be an important factor in development of ACS. In fact, several post-mortem pathological studies using coronary segments harvested from patients who died of coronary artery disease have shown that lesions with positive remodeling that had a larger lipid core were more vulnerable [3,12]. Further clinical study should be performed to confirm these pathological observations. Recently, ultrasound devices that detect tissue characteristics of plaques were developed and became available including IB-IVUS [7], Virtual HistologyTM (VH), and other custom designed software [13,14]. These modalities enable investigators to examine coronary plaque components in a series of clinical populations. Above all, IB-IVUS, used in this study, has the following advantages: (1) accurate auto pull-back system, (2) simple single parameter of IB values with high reproducibility, and (3) methodology validated by many studies [15–18]. Several investigations have focused on association of coronary plaque composition and vascular remodeling using VH-IVUS [19,20]. However, these VH analyses failed to demonstrate consistent results with the previous pathological assessments as well as to agree with current clinical consensus. In fact, they concluded that the lesions with positive remodeling had “less necrotic core” components than those with intermediate or negative remodeling, evoking a great controversy. One potential factor generating these discrepancies may be the enrollment of significant proportion of ACS patients. Basically, ACS lesions contain thrombi that interfere with accurate tissue characterization, because a specific spectrum/algorithm of VH or IB values has not been determined for a thrombus within a coronary artery. Consequently, such tissue components can be misjudged and assigned to dif-

ferent plaque components. Accordingly, we excluded ACS patients from our analysis. Ideally, future studies in patients with unstable presentation will be quite important because a series of these lesions are truly “vulnerable”. Further continuous approaches to establish discrimination of thrombi must be required for ultrasound-based tissue characterization. Importantly, our study showed that PR lesions had a greater content of lipid and lower content of hard plaque tissue component, consistent with previous pathological studies and clinical experiences [3,12,21], which explain the potential mechanism behind the high incidence of ACS seen in the PR lesion. In contrast, intermediate and negatively remodeled lesions had frequent contents of hard plaque tissue components, suggesting clinically more stable lesions than the PR lesions. We believe that these results provide important new information that clarifies the current controversy. Although no particular association was observed between remodeling pattern and medication usage in this study, several previous studies had indicated that some pharmacological intervention might affect the patterns of remodeling and changes in plaque compositions, including angiotensin-converting enzyme inhibitors and statins [22,23]. Other interesting findings of this study were the similarity of plaque compositions between a single cross-section of the culprit lesion and an average of 20-mm-long segment spanning the culprit lesion. This similarity indicated that the plaque composition is relatively uniform throughout the stenotic segments along the long axis of the coronary artery. Therefore, we can estimate plaque composition throughout a lesion and thereby the stability of the lesion from analysis of a single slice at the culprit lesion site. 3.1. Study limitations There are several limitations in this study. First, the sample size was relatively small. To reduce selection bias, we

482

H. Takeuchi et al. / Atherosclerosis 202 (2009) 476–482

enrolled consecutive patients seen daily at the clinic. Second, the acoustic shadow of calcification interferes with analysis of RF signals at the external side of calcification. To minimize this effect: (1) we excluded lesions with moderate to severe calcification and (2) the acoustic shadow of calcifications was manually traced and excluded from analysis. Therefore, it was not possible to address the impact of calcification on the plaque vulnerability in this study, although numerous studies have indicated an association between calcification and plaque vulnerability [24,25].

4. Conclusions PR lesions contain more lipid-rich components compared with non-PR lesions, which is linked to higher incidence of ACS and plaque vulnerability for PR lesions. Conversely, non-PR lesions are composed mainly of hard tissue that potentially accounts for the stability of these lesions. Similar plaque components are observed at MLA sites as well as surrounding adjacent segments.

References [1] Glagov S, Weisenberg E, Zarins CK, Stankunavicius R, Kolettis GJ. Compensatory enlargement of human atherosclerotic coronary arteries. N Engl J Med 1987;316:1371–5. [2] Fuster V, Badimon L, Badimon JJ, Chesebro JH. The pathogenesis of coronary artery disease and the acute coronary syndromes (1). N Engl J Med 1992;326:242–50. [3] Varnava AM, Mills PG, Davies MJ. Relationship between coronary artery remodeling and plaque vulnerability. Circulation 2002;105:939–43. [4] Nissen SE, Yock P. Intravascular ultrasound: novel pathophysiological insights and current clinical applications. Circulation 2001;103:604–16. [5] Schoenhagen P, Ziada KM, Kapadia SR, Crowe TD, Nissen SE, Tuzcu EM. Extent and direction of arterial remodeling in stable versus unstable coronary syndromes: an intravascular ultrasound study. Circulation 2000;101:598–603. [6] Nakamura M, Nishikawa H, Mukai S, et al. Impact of coronary artery remodeling on clinical presentation of coronary artery disease: an intravascular ultrasound study. J Am Coll Cardiol 2001;37:63–9. [7] Kawasaki M, Takatsu H, Noda T, et al. In vivo quantitative tissue characterization of human coronary arterial plaques by use of integrated backscatter intravascular ultrasound and comparison with angioscopic findings. Circulation 2002;105:2487–92. [8] Bridal SL, Fornes P, Bruneval P, Berger G. Parametric (integrated backscatter and attenuation) images constructed using backscattered radio frequency signals (25–56 MHz) from human aortae in vitro. Ultrasound Med Biol 1997;23:215–29. [9] Lockwood GR, Ryan LK, Hunt JW, Foster FS. Measurement of the ultrasonic properties of vascular tissues and blood from 35–65 MHz. Ultrasound Med Biol 1991;17:653–66.

[10] Dangas G, Mintz GS, Mehran R, et al. Preintervention arterial remodeling as an independent predictor of target-lesion revascularization after nonstent coronary intervention: an analysis of 777 lesions with intravascular ultrasound imaging. Circulation 1999;99: 3149–54. [11] Kawasaki M, Ito Y, Yokoyama H, et al. Assessment of arterial medial characteristics in human carotid arteries using integrated backscatter ultrasound and its histological implications. Atherosclerosis 2005;180:145–54. [12] Burke AP, Kolodgie FD, Farb A, Weber D, Virmani R. Morphological predictors of arterial remodeling in coronary atherosclerosis. Circulation 2002;105:297–303. [13] Komiyama N, Berry GJ, Kolz ML, et al. Tissue characterization of atherosclerotic plaques by intravascular ultrasound radiofrequency signal analysis: an in vitro study of human coronary arteries. Am Heart J 2000;140:565–74. [14] Nair A, Kuban BD, Tuzcu EM, Schoenhagen P, Nissen SE, Vince DG. Coronary plaque classification with intravascular ultrasound radiofrequency data analysis. Circulation 2002;106:2200–6. [15] Kawasaki M, Sano K, Okubo M, et al. Volumetric quantitative analysis of tissue characteristics of coronary plaques after statin therapy using three-dimensional integrated backscatter intravascular ultrasound. J Am Coll Cardiol 2005;45:1946–53. [16] Sano K, Kawasaki M, Okubo M, et al. In vivo quantitative tissue characterization of angiographically normal coronary lesions and the relation with risk factors: a study using integrated backscatter intravascular ultrasound. Circ J 2005;69:543–9. [17] Kawasaki M, Bouma BE, Bressner J, et al. Diagnostic accuracy of optical coherence tomography and integrated backscatter intravascular ultrasound images for tissue characterization of human coronary plaques. J Am Coll Cardiol 2006;48:81–8. [18] Sano K, Kawasaki M, Ishihara Y, et al. Assessment of vulnerable plaques causing acute coronary syndrome using integrated backscatter intravascular ultrasound. J Am Coll Cardiol 2006;47:734– 41. [19] Fujii K, Carlier SG, Mintz GS, et al. Association of plaque characterization by intravascular ultrasound virtual histology and arterial remodeling. Am J Cardiol 2005;96:1476–83. [20] Surmely JF, Nasu K, Fujita H, et al. Association of coronary plaque composition and arterial remodelling: a virtual histology analysis by intravascular ultrasound. Heart 2007;93:928–32. [21] Hasegawa T, Ehara S, Kobayashi Y, et al. Acute myocardial infarction: clinical characteristics and plaque morphology between expansive remodeling and constrictive remodeling by intravascular ultrasound. Am Heart J 2006;151:332–7. [22] Rodriguez-Granillo GA, de Winter S, Bruining N, et al. Effect of perindopril on coronary remodelling: insights from a multicentre, randomized study. Eur Heart J 2007;28:2326–31. [23] Schoenhagen P, Tuzcu EM, Apperson-Hansen C, et al. Determinants of arterial wall remodeling during lipid-lowering therapy: serial intravascular ultrasound observations from the Reversal of Atherosclerosis with Aggressive Lipid Lowering Therapy (REVERSAL) trial. Circulation 2006;113:2826–34. [24] Ehara S, Kobayashi Y, Yoshiyama M, et al. Spotty calcification typifies the culprit plaque in patients with acute myocardial infarction: an intravascular ultrasound study. Circulation 2004;110:3424– 9. [25] Fujii K, Carlier SG, Mintz GS, et al. Intravascular ultrasound study of patterns of calcium in ruptured coronary plaques. Am J Cardiol 2005;96:352–7.