Cardiovascular Research | 2019
Noninvasive detection of perivascular inflammation by coronary computed tomography in the CRISP-CT study and its implications for residual cardiovascular risk
Abstract
Unstable non-obstructive atherosclerotic plaques account for a significant proportion of acute coronary syndromes, especially when they get inflamed. Unfortunately, most of such plaques cannot be identified on non-invasive testing which rely on detection of flow-limiting stenosis. While certain inflammatory biomarkers like C-reactive protein increase risk of a major cardiovascular event, they are considered relatively non-specific to precisely detect actual inflammatory activity at the level of an individual coronary artery. In recent years, there is an increased understanding that the adipose tissue adjacent to the coronary arteries and the heart is an active area, generating many inflammatory agents within the local milieu. This is suspected of promoting progression of atherosclerotic changes within the coronary arteries, with resultant major adverse cardiac events. Coronary computed tomography angiography (CTA) is an established and highly sensitive non-invasive technique to detect obstructive and non-obstructive coronary artery disease (CAD). However, it has not been well-studied as a technique to detect inflammatory changes (or its impact on future coronary events) in the surrounding fatty tissue. Recently, a new index, fat attenuation index (FAI), has been developed to detect and quantify changes in perivascular fat resulting from ongoing inflammation, using clinically acquired coronary CTA. It appears that the inflamed coronary arteries release signals into the surrounding adipose tissue, which results in inhibition of local adipogenesis. This ultimately alters perivascular fat composition adjacent to the inflamed coronaries and shifts the attenuation on coronary CTA from the lipid to the aqueous phase. Perivascular FAI captures these inflammation-induced changes on routine coronary CTA, resulting in earlier detection of coronary inflammation. However, the index’s utility for clinical risk stratification was uncertain, prompting the CRISP-CT study, which tests the association between FAI and future fatal myocardial infarctions. To test this hypothesis in patients with suspected CAD, we collected data from the two cohorts of consecutive patients undergoing standard coronary CTA: 1872 patients in Germany from 2005 to 2009 (derivation cohort) and 2040 patients at Cleveland Clinic from 2008 to 2016 (validation cohort). The median ages were 62 years and 53 years, respectively. Perivascular FAI was measured around the three large epicardial coronary arteries—proximal right, left anterior descending, and left circumflex. Subsequently, the incremental prognostic utility of perivascular FAI was studied for all-cause and cardiac mortality upto a median of 72 months (range 51–109) in the derivation cohort and 54 months (range 4–105) in the validation cohort. Perivascular FAI values around the proximal right coronary artery and left anterior descending artery were predictive of all-cause and cardiac mortality in both cohorts. We subsequently used the perivascular FAI value around the right coronary artery as a marker of coronary inflammation [hazard ratio (HR) for cardiac mortality of 2.15 [95% confidence interval (CI) 1.33–3.48] in the derivation cohort and 2.06 (95% CI 1.50– 2.83) in the validation cohort]. The study also demonstrated that the optimal perivascular FAI cut-off in the derivation cohort was -70.1 Hounsfield units or higher [HR = 9.04 (95% CI 3.35–24.40) for cardiac mortality and HR = 2.55 (95% CI 1.65–3.92) for all-cause mortality]. This cut-off value was confirmed in the validation cohort [HR = 5.62 (95% CI 2.90–10.88) for cardiac mortality and HR = 3.69 (95% CI 2.26–6.02) for all-cause mortality]. Perivascular FAI significantly improved risk stratification in both cohorts beyond current risk models. These findings could be of potential significance because almost 50% of the myocardial infarctions occur in the absence of substantial coronary stenosis. The CRISP-CT study also highlights the importance of diligent translational research which resulted in the development of an imaging biomarker in a basic science realm, with subsequent application, and validation in a large human cohort across continents. The perivascular FAI potentially identifies higher-risk patients who may have highly inflamed and potentially unstable atherosclerotic plaques. Routine coronary CT angiography could help potentially identify the residual inflammatory risk that is captured by the perivascular FAI. These are patients who might not get recognized early and the use of this new biomarker could potentially help institute intensive preventative strategies. Future studies in this field should encompass the basic and clinical science arena. In the basic arena, FAI could be used to test for personalized triggers of inflammation (either environmental or genetic). In addition, it