Article Text

Download PDFPDF

Original article
Detection of disrupted plaques by coronary CT: comparison with angioscopy
  1. Mayu Nishio,
  2. Yasunori Ueda,
  3. Koshi Matsuo,
  4. Mitsutoshi Asai,
  5. Takayoshi Nemoto,
  6. Akio Hirata,
  7. Kazunori Kashiwase,
  8. Kazuhisa Kodama
  1. Cardiovascular Division, Osaka Police Hospital, Osaka, Japan
  1. Correspondence to Yasunori Ueda, Cardiovascular Division, Osaka Police Hospital, 10-31 Kitayama-cho, Tennoji-ku, Osaka, 543-0035 Japan; ueda{at}oph.gr.jp

Abstract

Background Disrupted plaques are the major cause of acute coronary syndrome (ACS). Although the detection of vulnerable plaques by coronary CT (CCT) has been examined and reported, there has been no report on the detection of disrupted plaques by CCT.

Objectives To test the ability of CCT to detect disrupted coronary plaques.

Methods 32 consecutive patients with suspected ischaemic heart disease who underwent successful coronary angioscopic examination and CCT were analysed. Yellow plaques of colour grade 1−3 and disrupted yellow plaques were examined by angioscopy. CCT findings (low attenuation, positive remodelling and ring-like enhancement) were examined for each site of yellow plaques.

Results In the 32 patients, 65 yellow plaques were detected. Higher-colour-grade yellow plaques and disrupted yellow plaques had a significantly higher incidence of CCT findings: low attenuation (grade 1 vs grade 2 vs grade 3, 18% vs 59% vs 69%; non-disrupted vs disrupted, 36% vs 66%), positive remodelling (24% vs 59% vs 75%; 33% vs 75%), and ring-like enhancement (0% vs 19% vs 25%; 6% vs 44%). Positive and negative predictive values for ring-like enhancement to detect disrupted plaque were 88% and 63%, respectively; those for the combined CCT findings (low attenuation, positive remodelling and ring-like enhancement) to detect disrupted plaque were 90% and 58%, respectively.

Conclusion CCT findings were associated with disrupted plaques confirmed by angioscopy. Ring-like enhancement had a high positive predictive value for detecting disrupted plaque.

  • Yellow plaque
  • disrupted yellow plaque
  • angioscopy
  • coronary CT
  • CT scanning
  • coronary angioscopy
  • coronary artery disease (CAD)
View Full Text

Statistics from Altmetric.com

Yellow plaques detected by angioscopy1–4 have high thrombogenic potential and a thin fibrous cap; therefore, they are regarded as vulnerable plaques and are believed to cause acute coronary syndrome (ACS) by their disruption and subsequent thrombus formation. Because the majority (>90%) of ACS culprits are yellow, both vulnerable plaques prone to rupture and those prone to erode would be detected as yellow plaques. Angioscopy can also identify unhealed disrupted plaques with thrombus differentiating them from old disruptions that have already healed and no longer have fresh thrombus. On the other hand, coronary CT (CCT) has become one of the most important diagnostic modalities for ischaemic heart disease and is widely used for screening and diagnosis of ACS patients. CCT can detect coronary stenosis with high positive and negative predictive values.5 6 Although plaque composition has no association with the degree of stenosis,7 CCT can also measure plaque volume, having a good correlation with intravascular ultrasound (IVUS) measurements, and can detect vulnerable plaques according to many recent reports.8–13 However, to the best of our knowledge, it has never been tested whether CCT can detect disrupted plaques, which is important in differentiating ACS from stable stenosis and in investigation of the process of ACS onset.

Therefore, in the present study, we have tested the ability of CCT to detect disrupted yellow plaques among all yellow plaques detected by angioscopy in patients with ischaemic heart disease.

Methods

Study patients

Thirty-two consecutive patients with suspected ischaemic heart disease who underwent successful coronary angioscopic examination within 2 weeks after successful CCT examination from April 2007 to June 2008 were analysed. This is a retrospective study; however, CCT is routinely used in our daily practice for patients with suspected ischaemic heart disease and a suspected symptom of angina or positive results of stress tests, and also for patients with high risk as a screening of coronary heart disease. Exclusion criteria for CCT are poor renal function, atrial fibrillation and apparently severe coronary calcification on chest x-ray. Angioscopic examinations are also performed routinely for the target vessels of intervention as a part of our daily practice.

Yellow plaques detected by angioscopy were included in this study; CCT images at the sites of yellow plaques were analysed. CCT findings were comparatively analysed with the angioscopy findings to test whether CCT findings could predict the presence of disrupted plaque detected by angioscopy.

ACS includes acute myocardial infarction defined by the Joint European Society of Cardiology/American College of Cardiology Committee, and unstable angina defined according to the Braunwald classification. Hypertensive patients were defined as those with systemic arterial pressure >140/90 mm Hg or those already taking antihypertensive drugs. Diabetic patients were defined as those with fasting blood glucose >126 mg/dl or those already taking oral drugs for diabetes mellitus or receiving insulin therapy. Written informed consent was provided by all patients.

CCT examination and evaluation

CCT was performed with a 64-detector row CT scanner (Light speed VCT, GE Healthcare, Milwaukee, Wisconsin, USA). All patients with a heart rate >60/min were given metoprolol 20 mg orally 2 h before the CT scan. At the time of examination, patients with heart rate >70 bpm were given 2 mg of intravenous propranolol. All patients received 0.3 mg nitroglycerin sublingually immediately before scanning. CCT data were acquired using an x-ray beam collimation width of 0.625×64 mm, gantry rotation time of 0.35 s, tube voltage of 100–120 mV and effective tube current of 280–750 mA using ECG modulation; the pitch ranged from 0.18 to 0.26 depending on the patient's heart rate. The estimated mean radiation dose was 15–18 mSv.

A non-contrast scan was initially performed to determine the anatomic landmarks for the contrast-enhanced study. Immediately thereafter, test bolus tracking with 15 ml of non-ionic contrast agent was applied to calculate the exact arrival time of contrast agent in the coronary arteries, with a region of interest in the proximal part of the ascending aorta. Finally, a contrast-enhanced scan with retrospective ECG gating was performed after administration of contrast medium (0.8 ml/kg body weight/12 s) during a single breath hold. Image reconstruction was performed with image-analysis software (CardIQ, GE Healthcare) on a dedicated computer workstation (Advantage Workstation Ver. 4.2, GE Healthcare). A standard kernel was used as the reconstruction filter. Depending on heart rate, either a half-scan or multi-segment reconstruction algorithm was selected or the optimal cardiac phase with the least motion artefacts was chosen individually.

Coronary arterial remodelling was evaluated by cross-sectional images. The remodelling index was calculated as the ratio of outer vessel diameter at the plaque site to the mean diameter of proximal and distal reference sites. Positive remodelling was defined as remodelling index >1.05.14 The CT attenuation value of a plaque was measured at five points and averaged. Low attenuation plaque was defined as the CT attenuation value <40 HU.13 Ring-like enhancement was defined as previously reported.15 Briefly, it is the presence of a ring with high attenuation around certain plaque, with its attenuation value higher than that of adjacent plaque but no greater than 130 HU to exclude calcium deposition.

Coronary vessel that was observed by angioscopy was analysed. CCT findings of low attenuation, positive remodelling and ring-like enhancement were evaluated for each site of yellow plaques detected by angioscopy. The sites of yellow plaques were recorded on the angiogram when examined by angioscopy and were determined on the CCT image using side branch, vessel morphology and/or stent as markers. Therefore, the evaluation of CCT was performed at those sites without knowing the angioscopic images or findings. CCT findings at the site of each yellow plaque and the angioscopic findings of the plaque were comparatively analysed after the independent evaluations.

Angioscopic examination and evaluation

Catheterisation was performed by a femoral, brachial or radial artery approach using a 6 or 7 Fr sheath and catheters. Intravenous heparin (100 U/kg) was administered at the beginning of catheterisation, and an additional dose was given at the time of percutaneous coronary intervention (PCI) as a component of standard care. GP-IIb/IIIa inhibitors were not used, because they were not approved in Japan. The coronary angiogram was recorded by the Advantx medical system (GE Healthcare).

Angioscopic examination was performed in the target vessel after the PCI procedure. The angioscope RX-3310A and MV-5010A (Machida, Tokyo, Japan) and optic fibre DAG-2218LN (Machida) were used. The angioscopic observations were made while blood was cleared from the viewing field by the injection of 3% dextran-40 as previously reported.16 The intensity of yellow colour was classified into three grades (1, slight yellow; 2, yellow; 3, intensive yellow), comparing with standard colours as previously reported.3 Thrombus was defined as white or red material with cotton-like or ragged appearance or with fragmentation with or without protrusion into the lumen or adherent to the luminal surface.16 For each yellow plaque detected, its yellow colour grade and whether it had thrombus on it were determined. Yellow plaques with thrombus were defined as disrupted. The angioscopic definition of plaque rupture has been described previously,17 and the disrupted plaques without rupture were defined as eroded plaques. Angioscopic evaluations were made by two angioscopic specialists blinded to patients' characteristics and CCT findings. In case of disagreement a third reviewer served as an arbitrator. The inter- and intra-observer reproducibility for the interpretation of angioscopic images was 85% and 95% for plaque colour, and 90% and 100% for thrombus, respectively. We focused on grade 2/3 yellow plaques, because grade 3 yellow plaques were regarded thin-cap fibroatheroma (TCFA) and grade 2/3 yellow plaques had the fibrous cap thickness thin enough to rupture as demonstrated by optical coherence tomography (OCT) studies.18 19

Statistical analysis

Continuous data were presented as mean±SD. Comparison of frequencies between groups was done by the χ2 or Kruskal–Wallis test; p<0.05 was regarded as statistically significant. Analysis was done with SPSS V.16.0J for Windows.

Results

Frequency of CCT findings at yellow plaques and at disrupted plaques

Table 1 shows the characteristics of the patients. A total of 65 yellow plaques were detected in the 32 patients. Table 2 presents CCT findings in comparison with angioscopic findings. Figure 1 shows the overlapping of CCT findings. The frequencies of all three CCT findings (low attenuation, positive remodelling and ring-like enhancement) were significantly higher at the higher-colour-grade yellow plaques and at the disrupted versus non-disrupted yellow plaques. Two representative cases are presented in figures 2 and 3.

Table 1

Clinical characteristics of the patients (N=32)

Table 2

Frequency of coronary CT (CCT) findings among each category of yellow plaques

Figure 1

Overlapping of three coronary CT (CCT) findings. No yellow plaque had ring-like enhancement alone. Among 65 yellow plaques, 19 (29%) plaques did not have any of three CCT findings.

Figure 2

Representative case 1. A 48-year-old male patient with unstable angina. Coronary CT (CCT) detected a significant stenosis in the proximal left anterior descending coronary artery with a plaque that had low attenuation (CT value=11 HU), positive remodelling and ring-like enhancement. Angioscopic examination showed that the lesion detected by CCT had grade 3 ruptured yellow plaque with thrombus.

Figure 3

Representative case 2. A 62-year-old male patient with stable effort angina. Coronary CT (CCT) detected a significant stenosis in the proximal left anterior descending coronary artery with a plaque that had low attenuation (CT value=25 HU), positive remodelling and ring-like enhancement. Angioscopic examination showed that the lesion detected by CCT had grade 3 ruptured yellow plaque with thrombus.

Detection of yellow plaques and disrupted plaques by CCT findings

The incidence of grade 2 or grade 3 yellow plaques among the plaques with each CCT finding is presented in figure 4. The sensitivity, specificity, positive predictive value and negative predictive value of CCT findings to detect yellow plaques of each grade and grade 2/3 yellow plaques are presented in tables 3–6. None of the CCT findings were highly predictive of yellow plaques of each colour grade separately (tables 3–5). However, positive predictive values were very high (90–100%), although negative predictive values were very low (30–40%), for any of three CCT findings or for combinations of those findings to detect grade 2/3 yellow plaques that were regarded as vulnerable (table 6).

Figure 4

The frequency of yellow plaques of each yellow colour grade at the site of various coronary CT findings. About 60–70% of the plaques with low attenuation alone or positive remodelling alone were grade 2 or 3 yellow plaques. However, all of the plaques with both low attenuation and positive remodelling were grade 2 or 3 yellow plaques. LAP, low attenuation; PR, positive remodelling; Ring, ring-like enhancement.

Table 3

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of coronary CT findings to predict grade 1 yellow plaques

Table 4

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of coronary CT findings to predict grade 2 yellow plaques

Table 5

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of coronary CT findings to predict grade 3 yellow plaques

Table 6

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of coronary CT findings to predict grade 2/3 yellow plaques

The incidence of ruptured and eroded yellow plaques among the plaques with each CCT finding is presented in figure 5. The sensitivity, specificity, positive predictive value and negative predictive value of CCT findings to detect disrupted plaques are presented in table 7. Although low attenuation and positive remodelling had moderately high (60–70%) positive and negative predictive values, ring-like enhancement with or without other CCT findings had very high (about 90%) positive predictive value. The absence of all three CCT findings was highly (84%) predictive of the absence of disrupted plaque.

Figure 5

The frequency of disrupted (ruptured or eroded) yellow plaques at the site of various coronary CT (CCT) findings. The plaques with low attenuation alone had plaque disruption in 29%; none had rupture, but erosion. The plaques with positive remodelling alone had disruption in about half (54%); 17% of the disruptions were rupture. The plaques with both low attenuation and positive remodelling also had disruption in about half (58%); 43% of the disruptions were rupture. The plaques with all three CCT findings (low attenuation, positive remodelling and ring-like enhancement) had disruption in 90%; two thirds (67%) of the disruptions were rupture. Ring-like enhancement was a strong predictor of disrupted plaque. LAP, low attenuation; PR, positive remodelling; Ring, ring-like enhancement.

Table 7

Sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of coronary CT findings to predict disrupted plaques

Discussion

We have tested, for the first time, whether CCT can detect disrupted plaques detected by angioscopy, and have shown high specificity and positive predictive values. The presence of ring-like enhancement was highly predictive (88%) of disrupted plaque; and the absence of all three CCT findings (low attenuation, positive remodelling and ring-like enhancement) was also highly predictive (84%) of the absence of disrupted plaque.

Detection of vulnerable plaques by angioscopy and CCT

Yellow colour of plaques detected by angioscopy has been associated with high thrombogenicity; furthermore, the higher yellow colour grade has been associated with higher thrombogenicity.3 Plaques with thinner fibrous cap are regarded as more vulnerable and are linearly correlated with higher plasma levels of inflammatory markers.20 The majority of grade 2 or 3 yellow plaques have fibrous cap thickness of <150 μm; grade 3 yellow plaques have fibrous cap thickness of about 50 μm,18 compatible with pathologically defined TCFA. As ruptured plaques at the culprit lesions of ACS had fibrous cap thickness of up to 150 μm when measured by OCT in living humans,19 the majority of grade 2/3 yellow plaques that had fibrous cap thickness of this range would have a chance to rupture. Therefore, the plaques with higher yellow colour grade would be more vulnerable, and grade 2/3 yellow plaques should be regarded as vulnerable plaques or clinically defined TCFA. Furthermore, angioscopically detected disrupted yellow plaques would include both ruptured and eroded plaques, because >90% of ACS culprits are disrupted yellow plaques by angioscopy, while 70% of them are pathologically ruptured but 30% eroded. Therefore, the vulnerable plaques that are prone to erode may not have the characteristics of TCFA, suggesting that positive remodelling or low attenuation may not predict that kind of plaque, while yellow colour may. This might be an explanation for the low negative predictive value of CCT findings to detect yellow plaques. Indeed, the risk of a future ACS event increases according to the increase in the number of yellow plaques as demonstrated in a prospective study.1 Therefore, yellow plaques can be regarded as vulnerable plaques defined by angioscopy.

On the other hand, plaques with low attenuation have been associated with the culprit lesions of ACS patients,10 12 although the substantial overlap of attenuation value between stable plaques and vulnerable plaques determined by IVUS has been reported.11 Positive remodelling as another sign of vulnerable plaques has been well correlated with low attenuation.9 The combination of positive remodelling, low attenuation and spotty calcification has been reported as highly specific for the culprit lesions of ACS patients.10 Ring-like enhancement has also been shown to be associated with TCFA in an OCT study.21 Therefore, these CCT findings may be the characteristics of vulnerable plaques.

In the present study, yellow plaques of the higher yellow colour grade had a higher frequency of low attenuation, positive remodelling and ring-like enhancement, any of which had very high positive predictive value (90–100%) but low negative predictive value (30–40%) for detecting grade 2/3 yellow plaques. As mentioned above, angioscopically defined vulnerable plaques (=yellow plaques) may include both rupture-prone and erosion-prone plaques, while CCT defined vulnerable plaques may miss some of the erosion-prone plaques that do not have TCFA-like morphology. Figure 5 shows that 29% of yellow plaques with low attenuation and 16% of yellow plaques without any of three CCT findings were disrupted; however, none of them were ruptured. Therefore, positive remodelling may be an essential factor required for the yellow plaques to rupture; however, the yellow plaques without positive remodelling still have a chance to disrupt and cause erosive thrombosis. Although the yellow plaques with positive remodelling may rupture or erode, the rate of rupture among all disruptions increases by having an additional characteristic of low attenuation. Therefore, the rupture-prone plaques and erosion-prone plaques may have different morphologies within a same spectrum. Furthermore, the yellow plaques with ring-like enhancement would be more vulnerable or already disrupted compared with the yellow plaques with low attenuation and/or positive remodelling alone. Although the possible explanation for ring-like enhancement has been speculated previously21 as the reflection of increased vasa vasorum density, large lipid content or thrombosis, it has not been clarified whether it is a characteristic of vulnerable plaques or disrupted plaques.

Detection of disrupted plaques by angioscopy and CCT

Disrupted plaques are angioscopically defined as plaques with thrombus, which are commonly detected at the culprit lesions of ACS.16 22 Although patients are usually diagnosed as having ACS or stable angina according to the clinical symptoms, the true instability of the culprit lesion depends on the thrombogenicity of the culprit plaque, which becomes highly unstable when the plaque disrupts and gradually loses its instability by its healing, as shown in the healing process of acute myocardial infarction culprit lesions.22 As some additional factors (ie, thrombogenic potential of blood, thrombogenic potential of necrotic core that would be exposed to blood by plaque rupture, underlying stenosis or stenosis caused by the protrusion of the necrotic core at the site of plaque rupture and vasoconstriction) should be required for the disrupted plaques to cause ACS,23 24 the lesions of silent plaque disruptions are not always safe but may cause ACS later, as in a case reported previously,25 in which disrupted yellow plaque was accidentally found by angioscopy in the proximal left anterior descending coronary artery where stenosis was mild; the patient presented with unstable angina 2 weeks later, when the site of disrupted plaque had sub-occlusive stenosis. Therefore, the silently disrupted plaques may also have high risk of an ACS event as vulnerable plaques.

However, the correlation between CCT findings and the unhealed disrupted plaques that still have thrombogenicity has not been reported previously. We showed in the present study for the first time that ring-like enhancement had high positive predictive value for detecting unhealed thrombogenic disrupted plaques. In addition, the absence of all three CCT findings was highly predictive of the absence of disrupted plaque. Therefore, if the patients with suspected ACS had a plaque with ring-like enhancement, the patients would have a high probability of having ACS; and the absence of all three CCT findings could exclude ACS with high probability.

Study limitations

Our study population is limited to those who underwent CCT and angioscopic examination for clinical indications and thus a selection bias may be possible. Another possible selection bias would be that plaques in the small or tortuous vessels were not included in the present study. We cannot exclude the influence of PCI on the angioscopic findings, although PCI usually does not change the yellow colour grade or presence/absence of thrombus at the target lesions according to our experience.

Conclusions

CCT findings were associated with disrupted plaques confirmed by angioscopy. The presence of ring-like enhancement was highly predictive (88%) of disrupted plaque; and the absence of all three CCT findings (low attenuation, positive remodelling and ring-like enhancement) was also highly predictive (84%) of the absence of disrupted plaque.

References

View Abstract

Footnotes

  • Competing interests None.

  • Patient consent Obtained.

  • Ethics approval This study was conducted with the approval of the Osaka Police Hospital Ethical Committee.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.