Author + information
- Maros Ferencik, MD, PhD∗ (, )
- Stefan B. Puchner, MD and
- Udo Hoffmann, MD, MPH
- ↵∗Knight Cardiovascular Institute, Oregon Health and Science University, 3180 SW Jackson Park Road, Mail Code UHN62, Portland, Oregon 97239
We thank Dr. Wong for his interest in our paper (1). Among patients with acute coronary syndrome (ACS), 32 of 37 (86.5%) patients underwent invasive coronary angiography. Coronary revascularization was performed in 26 of 37 (70.3%) patients. There were no cardiovascular deaths, and 5 patients were diagnosed with acute myocardial infarction during the index hospitalization. These numbers of events were insufficient to evaluate the predictive value of high-risk plaque features for hard cardiovascular outcomes.
We agree with Dr. Wong that the ability of high-risk plaque features to predict long-term cardiovascular outcomes is of great interest in acute chest pain patients. In the ROMICAT II (Rule Out Myocardial Infarction/Ischemia Using Computer Assisted Tomography II) trial, the design of the study did not include follow-up beyond 28 days. Therefore, we were unable to study long-term outcomes.
Plaque burden of >70% was associated with an increased risk of future major adverse cardiovascular events in the PROSPECT (Providing Regional Observations to Study Predictors of Events in the Coronary Tree) trial (2). The primary goal of our study was to perform qualitative assessment of high-risk coronary plaque features that can be included in the routine clinical assessment. The measurement of plaque burden requires quantitative evaluation of coronary arteries with dedicated software. We performed quantitative analysis of coronary plaque in the ROMICAT II trial and demonstrated that the volume of coronary plaque with low HU (<60 HU), number of plaques with positive remodeling (remodeling index >1.1), and number of plaques with plaque burden >70% were predictive of ACS during the index hospitalization in the ROMICAT II trial (3). In the multivariable analysis, the volume of low-HU plaque and positive remodeling remained significant predictors of ACS after adjusting for minimal luminal area of <4 mm2. The final model did not include plaque burden, which was in part due to overlap between positive remodeling and large plaque burden.
We agree with Dr. Wong that the scan parameters can affect the measurements of plaque CT attenuation. The typical tube potential in the ROMICAT II study was 120 kVp. However, luminal contrast enhancement, reconstruction filters, and image noise in addition to scan parameters can also influence plaque CT attenuation. The adaptive thresholding, which takes into account CT attenuation of structures outside of the plaque, may provide more accurate classification of coronary plaque composition (4). We demonstrated in ex vivo coronary arteries that the threshold of <60 HU may be more appropriate for the detection of lipid core (5). In the quantitative analysis in the ROMICAT II trial, the volume of plaque with <60 HU was predictive of ACS (3). Nevertheless, we decided to use the threshold of <30 HU in our study, as this was the most often used value in the published reports. We decreased the influence of the error induced by the selected threshold by using qualitative assessment and defining high-risk plaque as the presence of any of 4 features. We observed significant overlap in the presence of positive remodeling, low-HU plaque, and napkin-ring sign, which decreased the influence of a single high-risk plaque feature (e.g., low-HU plaque) on the results.
Please note: Dr. Ferencik has received support from the American Heart Association (grant 13FTF16450001). All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- American College of Cardiology Foundation
- Puchner S.B.,
- Liu T.,
- Mayrhofer T.,
- et al.
- Ferencik M.,
- Puchner S.B.,
- Lu M.,
- et al.
- de Graaf M.A.,
- Broersen A.,
- Kitslaar P.H.,
- et al.