Introduction

There is increasing evidence that acute clinical manifestations of coronary atherosclerosis are caused by plaque rupture and subsequent thrombus formation resulting from exposure of plaque components to flowing blood [1–3]. Plaque rupture at culprit lesions is associated with larger infarcts and left ventricular dysfunction [4], and reduced left ventricular function is a major risk factor for sudden death after myocardial infarction (MI) [5]. Positive remodeling of atherosclerotic plaques suggests a potential surrogate marker of plaque vulnerability as an indicator of the risk of plaque rupture [6]. Intravascular ultrasound (IVUS) imaging technology allows direct visualization of the distribution and morphology of atherosclerotic plaques in vivo and has greatly contributed to our understanding of plaque characteristics [7–9]. It also helps to determine the inner and outer curve of the vessel precisely based on the vascular and perivascular landmarks, especially in the left anterior descending coronary artery [10]. Several studies using IVUS have identified eccentric morphology, a thin fibrous cap, large lipid core [11], expanding remodeling [12, 13], and spotty calcification [14] as indicators of plaque vulnerability. However, the interactions among these factors are not fully understood. Previous studies have demonstrated that high shear stress causes plaque rupture [15, 16]. An in vivo study showed that all coronary plaque ruptures were observed in the proximal portion of the plaque hill, suggesting that a localized elevation of shear stress might be related to coronary plaque rupture [17]. Therefore, we hypothesized that coronary plaque rupture was prone to be observed on the outer side of the atherosclerotic plaque where shear stress is relatively high. The purpose of this study was to assess the association between the spatial location of plaque rupture defined by IVUS and remodeling pattern in the setting of acute anterior MI.

Methods

Study patients

We included consecutive patients with a first ST-elevation acute anterior MI defined as the presence of continuous chest symptoms for more than 30 min, accompanied by ST-segment elevation of >0.2 mV in at least two contiguous electrocardiographic leads and by an increase in serum creatine kinase levels to more than twice the upper limit of normal. All patients received coronary reperfusion within 6 h after the onset of symptoms, and infarct-related arteries were confirmed by IVUS before any percutaneous coronary intervention (PCI). Patients with malignant disease, infectious disease, or inflammatory diseases such as collagen disease were excluded to avoid bias. We also excluded patients with cardiogenic shock because we performed reperfusion procedures as soon as possible without preintervention IVUS. Plaque rupture was assessed in 90 patients who met all inclusion criteria. Fifty-two patients showed plaque rupture at the culprit lesion detected by preintervention IVUS, and they were enrolled in the present study. The protocol for the study was approved by the ethics committee of the Yokohama City University Medical Center. We obtained written, informed consent from all participants before initial coronary angiography.

Study protocol

On admission, all patients received 200 mg aspirin, an intravenous bolus injection of 5000 IU heparin, and isosorbide dinitrate. In the catheterization laboratory, coronary angiography was performed via the femoral approach. Periprocedural intravenous heparin was given to maintain an activated clotting time ≥300 s. Intracoronary isosorbide dinitrate (2–2.5 mg) was administered before angiography to prevent coronary artery spasm. After careful manipulation of the guidewire and additional intracoronary isosorbide dinitrate (2–2.5 mg), the IVUS catheter was advanced distal to the culprit lesion.

All IVUS studies were performed with a commercially available system (Scimed; Boston Scientific, Boston, MA) before any balloon inflation or stent implantation. A 40-MHz IVUS catheter was advanced distal to the culprit lesion, and imaging was performed in a retrograde fashion to the aorto-ostial junction with an automatic pullback speed of 0.5 mm/s. While pulling back the catheter, we manually infused contrast medium or normal saline suitable for IVUS imaging, while carefully observing the lesion. This procedure enabled us to eliminate blood noise and to observe communication between the plaque and the coronary artery lumen [4]. The IVUS images were recorded on S-VHS videotape for off-line analysis.

Clinical histories, including cardiovascular risk factors and current medications, were obtained from detailed interviews and medical records. Blood samples were obtained from all subjects on admission and analyzed in the hospital laboratory.

Analysis of IVUS images

Morphological features on IVUS images were independently interpreted by 2 experienced observers blinded to the angiographic and clinical data. Images for which the 2 observers agreed on the diagnosis were included in the subsequent analysis. Evaluation of lesion morphology and other measurements during IVUS was done according to the American College of Cardiology Clinical Expert Consensus Document on Standards for Acquisition, Measurement, and Reporting of Intravascular Ultrasound Studies [18]. We defined lesions with plaque rupture on IVUS as follows: (1) lesions with fissure/dissection or (2) lesions without fissure/dissection but in which the injection of saline or contrast medium confirmed communication between the plaque and the coronary artery lumen [19] (Figs. 1, 2, 3). The eccentricity index was calculated as (maximum plaque plus media thickness minus minimum plaque plus media thickness) divided by maximum plaque plus media thickness. An eccentric plaque was defined by an eccentricity index of >0.5. Plaque composition was assessed visually and classified as echolucent when >75% of the plaque area was less bright than the adventitia; otherwise, plaque composition was classified as echodense.

Fig. 1
figure 1

Representative case of plaque rupture on the myocardial side. Upper panel shows a preintervention angiographic image. i An intravascular image of plaque rupture. Plaque ulceration with fissure (white arrow) on the myocardial side and ruptured cavity (arrowheads) are observed. Lower panel shows the postintervention angiographic image and branching landmarks seen by intravascular ultrasound for spatial orientation in the coronary artery. a First diagonal branch (asterisk), b second diagonal branch (asterisk), c major septal branch (asterisk), d myocardial bridge (arrowheads) indicates that this side is epicardial

Full size image
Fig. 2
figure 2

Representative case of plaque rupture on the epicardial side (ii). Plaque ulceration with fissure (white arrow) on the epicardial side and ruptured cavity (arrowheads) are observed. e–g Branching landmarks seen by intravascular ultrasound for spatial orientation in the coronary artery. e First major septal branch (asterisk), f first diagonal branch (asterisk), g second septal branch (asterisk), h this panel indicates that left side is epicardial

Full size image
Fig. 3
figure 3

Representative case of plaque rupture on the lateral side (iii). Plaque ulceration with fissure (white arrow) on the lateral side and ruptured cavity (arrowheads) are observed. i, j Branching landmarks seen by intravascular ultrasound for spatial orientation in the coronary artery. i First major septal branch (asterisk), j first diagonal branch (asterisk), this panel indicates that left side is epicardial

Full size image

Using the pericardium, side branches, coronary veins, and myocardium as landmarks, we divided the plaques into 3 categories according to the spatial orientation of the plaque rupture site (Figs. 1, 2, 3). Spatial orientation in the vessel was as follows: myocardial quadrant (inner curve of the vessel), epicardial quadrant (outer curve of the vessel), and lateral quadrants (2 quadrants intermediate).

External elastic membrane cross-sectional area (EEM CSA) and lumen CSA (in mm squared) at the lesion site and at the proximal and distal reference sites were analyzed using planimetry software (TapeMeasure, Indec Systems, Capitola, CA). The reference segments were the most normal-looking CSA within 5 mm proximal and distal to the lesion but before any side branch. Remodeling was considered positive when the lesion EEM CSA was greater than both the proximal and distal reference EEM CSAs [20]. The remodeling index was determined by dividing the EEM CSA at the lesion site by the mean value of the EEM CSA at the proximal and distal reference sites.

Statistical analysis

Statistical analysis was performed with StatView 5.0 software (SAS Institute, Cary, NC). Results are expressed as mean ± SD for continuous variables. Qualitative data are presented as numbers (%). Continuous variables were compared by means of Student’s t test, and categorical data were compared by the chi-square test or Fisher’s exact test between the groups. For all analyses, values of p < 0.05 were considered to indicate statistically significance.

Results

Coronary artery lesions were successfully observed before balloon inflation or stent implantation in all patients by IVUS, without procedural complications. Patients were classified according to the presence or absence of positive remodeling. Thirty-five patients showed positive remodeling (positive group), and 17 did not show positive remodeling (non-positive group). Baseline patient characteristics are summarized in Table 1. There were no statistically significant differences in age, gender, presence of preinfarction angina, risk factors, and laboratory results between the groups. The preintervention IVUS findings are summarized in Table 2. Plaques in the positive group showed a higher proportion of soft plaque morphology (94 vs. 65%, p = 0.0016) and had a larger vessel area at the culprit-lesion site (19.2 ± 4.4 vs. 15.2 ± 4.7 mm2, p = 0.0053) than plaques in the non-positive group. In 52 plaque ruptures, 27 were oriented toward the epicardial side (52%), 18 toward the myocardial side (35%), and 7 in the 2 lateral quadrants (13%). Among 35 plaques in the positive group, plaque rupture was observed in 21 (52%) on the epicardial side, 12 (34%) on the myocardial side, and 2 (6%) on the lateral side. Among 17 plaques in the non-positive group, plaque rupture was observed in 6 (35%) on the epicardial side, 6 (35%) on the myocardial side, and 5 (30%) on the lateral side. The orientations of plaque rupture were significantly different with an increased incidence of plaque rupture on the epicardial side in the positive group (p = 0.047) (Table 2). Among 33 plaques that showed eccentric plaque, the distribution of atherosclerotic plaque was not significantly different between the groups (data not shown).

Table 1 Baseline clinical characteristics
Full size table
Table 2 Intravascular ultrasound findings
Full size table

Discussion

In this study, we examined the culprit plaque in acute MI and found that plaques with positive remodeling showed more frequent plaque rupture on the epicardial side of the coronary vessel wall than those without positive remodeling.

Positive remodeling is defined as a compensatory increase in local vessel size in response to increasing plaque burden [21]. Several studies have suggested that positive remodeling is a potential surrogate marker of plaque vulnerability [6, 22]. Previous studies demonstrated that positive remodeling and larger plaque areas were associated with acute coronary syndrome [12, 13] and suggested that positive arterial remodeling may be associated with an increased risk of plaque rupture [23]. In addition, recent optical coherence tomography studies demonstrated that the presence of microchannels in plaques correlated with a greater frequency of thin-cap fibroatheroma and positive remodeling [24, 25], suggesting an important role of vasa vasorum in plaque instability and coronary remodeling. Consistently, the biological implications of the relationship between coronary remodeling and plaque disruption suggest that large plaques with positive remodeling are more susceptible to mechanical forces that lead to plaque rupture. Imoto et al. [26] showed that positive remodeling led to a greater concentration of stress than did constrictive remodeling. In the present study, plaque rupture was observed equally in 3 categories (epicardial, myocardial, lateral sides) in patients without positive remodeling, whereas plaque rupture was more frequently on the epicardial side in patients with positive remodeling. These findings indicate that mechanical force such as high shear stress along the outer curve of the vessel wall, which corresponds to the epicardial side, may cause plaque rupture in plaques with positive remodeling.

Shear stress is one of the important physical factors in the process of atherosclerosis. Shear stress is defined as a stress produced by blood flow, which tends to cause vessel endothelium to slide or to be deformed. It has been recognized that regions of low shear stress are more susceptible to atheroma formation, and many studies have demonstrated an association between low shear stress and atherosclerotic deposits [27–30]. Furthermore, a recent study suggested that a localized elevation of shear stress might be related to coronary plaque rupture and showed that all coronary plaque ruptures were observed in the proximal of top portion of the plaque hill [17]. In addition to the process of atherosclerosis development and progression, several studies showed that lowered shear-stress lesions were associated with outward (positive, expansive) vascular remodeling [31, 32]. These studies indicate that low shear stress is relevant not only to the spatial orientation of atherosclerotic plaques, but also to the positive remodeling leading to increased plaque vulnerability.

Although the relationship between shear stress and plaque development is still evolving, there is little information in vivo on the relationship between plaque disruption and shear stress. A previous autopsy study showed differences in cell composition between upstream and downstream parts of plaques and suggested that high shear/high flow may be related to plaque instability [15]. Fukumoto et al. [17] demonstrated that a localized elevation of shear stress on the plaque surface was related to coronary plaque rupture in patients using a simplified computational analysis from three-dimensional IVUS plaque images. Our study showed that plaque rupture was observed more frequently on the epicardial side than myocardial and lateral sides in plaques with positive remodeling, further extending the previous findings regarding the association between shear stress and plaque rupture. Additionally, elucidation of the geometry of plaque rupture would contribute to future understanding of the coronary artery disease.

Atherosclerotic plaque is ubiquitous even in coronary arteries without angiographically significant stenosis [33]. Our findings suggest that coronary plaques located in the epicardial side of the coronary vessel wall with positive remodeling represent the high-risk for plaque rupture and may contribute to the management of the patients with coronary artery disease and prevention of coronary events. A recent study demonstrated that aggressive lipid-lowering therapy achieved significant regression of the coronary plaque volume with negative vessel remodeling in patients with acute coronary syndrome [34]. In a serial 3-vessel IVUS and optical coherence tomography study, Yamada et al. demonstrated that percent changes in fibrous cap thickness correlated negatively with the percent changes of EEM CSA during 6-month follow-up. These results provided evidence that administration of statins might reverse the remodeling process of atherosclerosis and reduce plaque vulnerability [35].

Our study had several limitations. This was a single-center, retrospective study with a relatively small number of patients. We enrolled only patients with anterior acute MI because arterial length and branching patterns differ among the left anterior descending, left circumflex, and right coronary arteries. Because of the differences in distortion among the three coronary arteries, the effect of endothelial shear stresses may differ. Therefore, our results cannot be extrapolated to left circumflex and right coronary arteries. Although we carefully examined lesions by flushing the surrounding region with saline or contrast medium, lesions with small ruptured plaques may have been misread as non-ruptured plaques. Finally, in the present study, we did not calculate the distribution of the shear stress on the plaque surface, so these results are largely conjectural.

Conclusion

In the setting of anterior MI, plaque rupture of atherosclerotic plaque with positive remodeling was located more frequently on the epicardial side of the coronary vessel wall.