Cardiovascular disease remains the leading cause of death worldwide. Identifying individual patients at risk of a suture adverse events, including myocardial infractions (heart attacks), remains a major diagnostic challenge. Recent studies have shown that coronary plaques responsible for hear attack are composed of a large lipid core with a thin overlying fibrous cap. Although these features can be identified using invasive imaging modalities, non-invasive imaging options remain limited due to their poor spatial resolution. Recently the investigators have developed and validated a novel tool that will allow us to characterise coronary plaque composition based on dual source CT images. Our aim is to assess this tool within a cohort of patients who have already undergone a coronary CT as part of a previous study.

Reliable identification of coronary plaque at risk of causing future adverse cardiovascular events would allow patient-specific targeting of intensive therapy. The majority of events are precipitated by coronary plaque rupture, with ruptured plaques exhibiting a large necrotic, lipid core with superimposed thrombus. The proposed precursor lesion shares similar plaque compositional features to ruptured plaques but with a thin overlying fibrous cap and is termed a thin-cap fibroatheroma (TCFA)1. At present there is a major emphasis on imaging modalities that can identify these higher-risk plaque subtypes.

We have previously validated an invasive imaging modality, virtual-histology intravascular ultrasound (VH-IVUS) against histology, and found that VH-IVUS identified TCFA with a diagnostic accuracy of 74%2. In prospective studies, VH-defined TCFAs were associated with a 7x greater risk of future events than other plaque subtypes3. Although this technique shows promise in permitting plaque-based risk stratification, it is limited by its invasive nature. Thus, alternate, non-invasive imaging options are required for more widespread risk assessment.

Recently, we have devised a novel method of creating "Plaque Maps" using CT attenuation data individualised to each patient (Figure 1), which permits identification of coronary plaque components with a diagnostic accuracy of 75%-85%4. However, whilst CT Plaque Maps could identify fibroatheroma with similar diagnostic accuracy to VH-IVUS (79% vs. 74%), the spatial resolution of CT was unable to detect the thin fibrous cap and distinguish fibroatheroma from TCFA (Figure 2). More recently we have introduced necrotic core/fibrous plaque ratio as a possible Plaque Map surrogate for identification of TCFA. Using a ratio of >0.58, the sensitivity to detect a TCFA was 84% and specificity 75%; an improvement over all previously identified CT-defined features of high risk plaques4. Whether this novel strategy can prospectively improve identification of plaque vulnerability is unproven.

Inclusion Criteria:

• Age >18 years
• The ability to provide informed consent.
• Previously part of CTA accuracy validation study

Exclusion Criteria:

• The inability of the participant to provide informed consent

This study will be an observational, retrospective, single center cohort study based at Papworth Hospital NHS Trust, including up to a maximum of 100 patients. The patients have already been identified through their participation in a previous research study, as stated above. Additional inclusion criteria are; age >18 years and the ability to provide informed consent. The only exclusion criterion is the inability of the participant to provide informed consent.