Coronary High-Intensity Plaques at T1-weighted MRI in Stable Coronary Artery Disease: Comparison with Near-Infrared Spectroscopy Intravascular US.
Shunya SatoHidenari MatsumotoDebiao LiHidefumi OhyaHiroyoshi MoriKoshiro SakaiKunihiro OguraYosuke OishiRyota MasakiHideaki TanakaSeita KondoHiroaki TsujitaShigeto TsukamotoKoji IsodonoRyoji KitamuraYoshiaki KomoriNobuyuki YoshiiIkumi SatoAnthony G ChristodoulouYibin XieToshiro ShinkePublished in: Radiology (2021)
Background The histologic nature of coronary high-intensity plaques (HIPs) at T1-weighted MRI in patients with stable coronary artery disease remains to be fully understood. Coronary atherosclerosis T1-weighted characterization (CATCH) enables HIP detection by simultaneously acquiring dark-blood plaque and bright-blood anatomic reference images. Purpose To determine if intraplaque hemorrhage (IPH) or lipid is the predominant substrate of HIPs on T1-weighted images by comparing CATCH MRI scans with findings on near-infrared spectroscopy (NIRS) intravascular US (IVUS) images. Materials and Methods This study retrospectively included consecutive patients who underwent CATCH MRI before NIRS IVUS between December 2019 and February 2021 at two facilities. At MRI, HIP was defined as plaque-to-myocardium signal intensity ratio of at least 1.4. The presence of an echolucent zone at IVUS (reported to represent IPH) was recorded. NIRS was used to determine the lipid component of atherosclerotic plaque. Lipid core burden index (LCBI) was calculated as the fraction of pixels with a probability of lipid-core plaque greater than 0.6 within a region of interest. Plaque with maximum LCBI within any 4-mm-long segment (maxLCBI 4 mm ) greater than 400 was regarded as lipid rich. Multivariable analysis was performed to evaluate NIRS IVUS-derived parameters associated with HIPs. Results There were 205 plaques analyzed in 95 patients (median age, 74 years; interquartile range [IQR], 67-78 years; 75 men). HIPs ( n = 42) at MRI were predominantly associated with an echolucent zone at IVUS (79% [33 of 42] vs 8.0% [13 of 163], respectively; P < .001) and a higher maxLCBI 4 mm at NIRS (477 [IQR, 258-738] vs 232 [IQR, 59-422], respectively; P < .001) than non-HIPs. In the multivariable model, HIPs were independently associated with an echolucent zone (odds ratio, 24.5; 95% CI: 9.3, 64.7; P < .001), but not with lipid-rich plaque (odds ratio, 2.0; 95% CI: 0.7, 5.4; P = .20). Conclusion The predominant substrate of T1-weighed MRI-defined high-intensity plaques in stable coronary artery disease was intraplaque hemorrhage, not lipid. © RSNA, 2021 Online supplemental material is available for this article. See also the editorial by Stuber in this issue.
Keyphrases
- coronary artery disease
- high intensity
- contrast enhanced
- magnetic resonance imaging
- percutaneous coronary intervention
- diffusion weighted imaging
- coronary artery bypass grafting
- resistance training
- cardiovascular events
- magnetic resonance
- end stage renal disease
- computed tomography
- fatty acid
- coronary artery
- ejection fraction
- deep learning
- newly diagnosed
- chronic kidney disease
- aortic stenosis
- peritoneal dialysis
- healthcare
- cardiovascular disease
- prognostic factors
- total hip arthroplasty
- type diabetes
- patient reported outcomes
- social media
- transcatheter aortic valve replacement
- body composition
- acute coronary syndrome
- risk factors
- clinical evaluation
- light emitting
- machine learning
- amino acid