Discussion
The major findings of this pilot study include the following: 1) the PMR of coronary HIP was lowered by statin therapy, which was also associated with decreases in LDL-C and hs-CRP as well as a decrease in the percentage of TAV and percentage of LAP volume as evaluated by CTA; 2) the percentage of change in the PMR was greater in the statin group with PMR ≥1.4 segments; and 3) in the control group, which did not receive statins or other LDL-lowering agents, the PMR was higher at 12 months, especially in patients who had HIP with PMR ≥1.4. The present study suggests the feasibility of using serial CMR examinations using noncontrast T1WI in clinical trials designed to assess changes in coronary plaque characteristics (Central Illustration).
(Enlarge Image)
Central Illustration.
Statin and HIP on CMR
Statin therapy reduces the plaque-to-myocardium signal intensity ratio (PMR) of coronary high-intensity plaque (HIP) through its plaque-stabilizing effects in association with decreases in low-density lipoprotein cholesterol, high-sensitivity C-reactive protein (hs-CRP), and low attenuation plaque (LAP) volume. In contrast, the PMR in the control group increased without significant morphological changes detected by computed tomography angiography measures. CMR = cardiac magnetic resonance; IPH = intraplaque hemorrhage; T1WI = T1-weighted imaging.
Observations from carotid plaque magnetic resonance imaging and histopathological validation studies, as well as studies using optic coherence tomography or specimens obtained through an aspiration catheter during PCI after ACS suggest that a coronary HIP may represent IPH and thrombus formation. Kawasaki et al. systematically evaluated the components of HIPs in patients undergoing CTA and IVUS. Coronary HIPs were closely correlated with IVUS attenuation, as well as with positive vascular remodeling and lower CT density on CTA. The estimation of HIP-PMR provides a useful quantitative measure that can be repeatedly analyzed without the need for radiation exposure, contrast agents, or invasive procedures. Therefore, it is reasonable to use HIP-PMR for the serial evaluation of plaque characterization, especially in response to plaque-modifying agents.
Numerous carotid magnetic resonance studies have demonstrated the critical role of IPH in plaque instability and acute carotid vascular events. Statin therapy has been posited to prevent neovascularization and limit the cholesterol content of red blood cell membranes and the phospholipid ratio. Carotid IPH was less frequently observed in patients on statins before endarterectomy, and the use of statins before a transient ischemic attack or stroke was negatively associated with the presence of IPH. On the other hand, statin therapy has been associated with a decrease in the percentage of LAP volume as assessed by CTA. Komukai et al. and Hattori et al. demonstrated that stain-induced optical coherence tomography–verified increases in fibrous cap thickness were associated with decreases in serum atherogenic lipoproteins and inflammatory biomarkers. This suggests that statin therapy modifies plaque phenotype including its lipid-rich necrotic core, fibrous cap, and IPH, which in turn might have reduced PMR values in our study. Future studies should investigate whether the effect of statins on HIPs, as monitored by noninvasive CMR, is also associated with reductions in the risk of clinical events.
In our segment-based analysis, the degree of PMR attenuation after statin treatment was significantly greater in the high versus low PMR groups. Conversely, the magnitude of PMR increase was significantly greater in control patients with PMR ≥1.4 segments than control patients with PMR <1.4 segments (Figure 5). Taken together with our previous study demonstrating that coronary lesions with PMR ≥1.4 are at significantly higher risk of subsequent ACS, the present findings indicate that coronary lesions with PMR ≥1.4 may be associated with accelerated plaque instability as well as increased signal intensity, which may reflect an increase in the volume of the necrotic core with IPH. This supports recent findings that fateful plaques are usually large with large necrotic cores.
However, increases in the PMR were observed even in PMR <1.4 segments (12.0% increase from baseline) (Figure 5), which are considered at lower risk of coronary events on the basis of our previous study. This suggests that even HIP with PMR <1.4 might evolve into a high-risk plaque during follow-up. Given that atherosclerosis is a dynamic process, our focus must remain on the entire disease process. Early identification of patients with HIP regardless of stenosis severity or plaque burden may prove valuable in the risk stratification of patients with CAD or multiple cardiovascular risk factors, including diabetes mellitus. Additionally, the present study proposes that noncontrast T1WI can potentially be used for comparing plaque characteristics at different time points and may assist in assessing the efficacy of antiatherosclerotic pharmacological interventions.
Study Limitations
As an observational study with a small number of patients examined, there may be inherent flaws related to selection bias, spurious observations, unmeasured covariates, and nonrandom allocation to treatment. However, we sought to minimize these issues by using a propensity model for multivariate analysis and added a summary of coronary events in the 2 groups during follow-up on the basis of an exact logistic regression analysis (Online Table 2 http://content.onlinejacc.org/data/Journals/JAC/934226/05056_mmc1.docx?v=635718626142300000). These data suggest that patients with HIPs seem to be at a higher risk of future coronary events. Second, because plaque measurements using CTA were performed semiautomatically, there is a possibility of measurement error. Third, intensive statin therapy did not change TAV and vessel RI as detected by CTA in this study. This was inconsistent with previous IVUS studies demonstrating that intensive statin treatment induces reductions in TAV as well as an absolute decrease in vessel RI. CTA has lower spatial resolution than IVUS for the measurement of plaque volume, which may in part be related to these inconsistencies. Finally, Noyes et al. reported that plaque regression occurred after an average of 19.7 months of statin treatment. Because target LDL-C levels in this study were comparable with other intensive statin IVUS studies, mean changes in plaque volume and PMR at 1 year of follow-up were acceptable. However, studies of serial changes in the PMR of HIP beyond 1 year might provide additional insights.
Most importantly, HIP assessment may be more conveniently possible in Japanese subjects by virtue of the body habitus and its applicability elsewhere has yet to be determined. Therefore, the findings and proposals from the study are considered hypothesis generating.