By Walter Alexander

It’s the interventionalist’s Holy Grail: knowing which lesion is going to occlude—indicating a need for preventive stenting—and which one can be treated successfully with optimal medical therapy. It is a vision diametrically opposed to the “ocular stenotic reflex” approach, aptly named by Patrick W. Serruys, MD, of Erasmus University, Rotterdam, The Netherlands. Picture Charlie Chaplin as the crazed factory worker armed with a very big wrench in Modern Times: every protuberance was a nut to be turned.

The variety of imaging modalities available to interventionalists and imagists for revascularization decision making are numerous—from coronary flow reserve, fractional flow reserve (FFR), to positron emission tomography (PET) and optimal coherence tomography (OCT)—but they’re certainly no crystal ball. Is it just that we’re looking in the wrong place?

“We know that plaque composition is at the core of the disease, and that only progressive lesions result in acute events associated with plaque rupture,” Jagat Narula, MD, PhD, Ichan School of Medicine at Mount Sinai, New York, and editor-in-chief of JACC: Cardiovascular Imaging, told CSWN: Interventions. “With angiography or FFR, we are measuring the progressive nature of the disease; we are measuring the shrinking lumen. However, if you want to explore, classify, and risk stratify disease—and ultimately intervene—you need to go into the disease.” How to best do that is the central question considered in a new State-of-the-Art paper in JACC, co-authored by Dr. Narula and Pedro R. Moreno, MD, also of Mount Sinai.¹

Anatomy Versus Physiology

More than 25 years ago, the common belief was that acute myocardial infarction (MI) radiated from an easily visible obstruction in the plumbing, essentially. This began to show signs of cracking, however, when Glagov and colleagues described plaque remodeling. They demonstrated that a 75% cross-sectional area (CSA) stenosis was only a 50% diameter stenosis when viewed by angiography, suggesting that, in general, noninvasive imaging overestimates disease compared to coronary angiography.²

In 1988, John A. Ambrose, MD, working with Valentin Fuster, MD (also of Mount Sinai), published a retrospective study on the progression of coronary artery disease (CAD) that occurs between two cardiac catheterization procedures.³ The study encompassed two groups: 23 patients who had an MI between the procedures (Group I) and 15 who subsequently presented with at least one new total occlusion without an intervening MI (Group II). The revelation was that the MIs often developed from previously nonsevere lesions.

Only 22% (5/23) of the lesions in Group I patients were >70% at the time of the first catheterization. It was different in Group II, where 61% (11/18) were found to have been >70% during the initial catheterization. Moreover, the original narrowing in what would become the infarct-related artery leading to Q-wave MI in Group I patients was less severe (34%) than the arterial narrowing in patients who subsequently developed a non-Q-wave MI (80%). The only predictor of lesion evolution to >50% was proximal lesion location. The authors concluded that non-Q-wave infarction is usually preceded by a more severe pre-existing stenosis than Q-wave infarction.

What is it that explains the apparent evolution of MIs from less severely obstructing plaques? One account—a purely statistical one—was suggested in an interview with the University of Washington’s cath lab director, Steven Goldberg, MD: there are simply many more 30% lesions than 70% lesions. “Rupture of a plaque with subsequent clot formation and occlusion is statistically more likely with the 30% lesions because of their overwhelmingly large numbers.” On the other hand, he said, the development of non-ST-elevation segment MIs (NSTEMIs) may be explained by high-burden plaques and high-grade stenoses that make flow insufficient to supply myocardium. Statistics aside, the reason for STEMI evolution from less occlusive plaques may need another explanation.

Dr. Narula recently showed that the majority of event-causing coronary lesions enlarge substantially over time before plaque rupture.⁴ His colleague, Dr. Moreno, essentially finds the same phenomenon when he reconsiders his paper presented more than a decade ago.⁵

This reconciles reports of nonobstructive lesions leading to acute events with significant luminal stenosis. Serial intravascular imaging studies by Gregg W. Stone, MD, professor of medicine at Columbia University’s College of Physicians and Surgeons, demonstrated that lesions leading to major adverse coronary event- (MACE) are characterized by modest luminal stenosis, large plaque burden, and thin fibrous caps.⁶ Their volume can expand considerably, write Drs. Narula and Moreno in their State-of-Art review paper and that most lesions “are huge and significantly obstructive” at the time of most acute events. They wrote, “What the evidence has been increasingly revealing is that what leads to plaque rupture in patients with ACS, regardless of the initial luminal stenosis, is the expanding plaque burden and thinning fibrous cap.”

So, plaque composition is the principal determinant of atherothrombosis. When the situation is not an expanding plaque burden and thinning fibrous cap but rather obstructive lesions that are predominantly fibrotic, they may cause angina and require revascularization, but they do not lead to acute coronary syndromes (ACS).

FFR Insights from Chronic Stable Angina Trials

Being able to assess fractional flow reserve was a big step forward from angiography, write Drs. Moreno and Narula. The COURAGE trial, which compared angiographically-driven percutaneous coronary intervention (PCI) versus medical therapy in patients with chronic stable angina, found no difference between the strategies, with about a 20% risk of stroke, death, and MI after 5 years no matter which approach was taken. To the interventional community, COURAGE (in which FFR was not measured) indicated that not every plaque, even if obstructive, needs stenting.

The subsequent FAME trial did assess FFR prospectively. It randomized patients to angiographically-guided PCI or FFR-guided PCI and found that among lesions morphologically obstructive by angiography, 37% were FFR negative. Avoiding PCI in these lesions resulted in an absolute 5% reduction in death, MI, stroke and revascularization at 1 year. “So if you tailor your intervention to blockages that are severe according to angiography and FFR, you will benefit the population. But if you don’t do FFR, you will include 37% of lesions that may not benefit,” Dr. Moreno said. However, more is needed to complete the story, insists Dr. Narula.

Fine-Tuning FFR

Following from these results, FAME II randomized patients with FFR-positive lesions to optimal medical therapy (OMT) versus PCI, excluding all of the lesions that were obstructive based only on angiography. The Data Safety and Monitoring Board stopped enrollment in the trial after 9 months because of a large increase in hospital readmissions and revascularizations in the OMT arm. Two years after randomization, 11.1% of the OMT patients had undergone urgent revascularization, versus only 1.6% of the PCI group. The composite endpoint rates were 12.7% and 4.3% in the OMT and PCI groups, respectively (p < 0.001). There were no differences in death or MI.⁷

FAME II investigators reported that the number needed to treat (NNT) was fairly high: 10 FFR-positive patients had to be treated to prevent one urgent revascularization. “It’s still a high price to prevent one event,” Agreeing with Dr. Narula, Dr. Moreno added, “The field needs additional technology to identify the roughly nine plaques that will do fine with OMT.”

The limitation of FFR, Dr. Moreno said, is that it is derived purely from luminal obstruction and is therefore a surrogate for angiographic severity. If the reason that plaques grow and become bulky and then rupture is not because of obstruction but because of their composition, then the next step in identifying risk through invasive modalities will have to be the ability to characterize plaque composition.

“Then you can dichotomize which plaque will have the lipid-rich core associated with inflammation, growth, and rupture and those that will have a fibrotic or a fibro-calcific composition,” explained Dr. Moreno. “Those plaques are inert and don’t have life to grow from; they have few cells and very little lipid. Plaque expansion is mediated by lipid deposition and collagen expansion.”

FFR can identify hemodynamic impairment but can’t differentiate between the lipid-rich obstructive plaques versus the fibrotic or fibro-calcific plaques. “That’s where imaging comes in, specifically optimal coherence tomography OCT, because it can differentiate whether or not a plaque is fibrotic/fibro-calcific or lipid rich. Then we can tailor our treatment toward the high-risk patients,” claims Dr. Narula.

Intracoronary Imaging in ACS Patients

The progression of nonculprit lesions (NCLs) can lead to recurrent coronary events in ACS patients. In several ACS trials, all culprit and angiographically obstructive lesions had been already treated successfully, removing their role in triggering future events and allowing intravascular imaging studies of NCLs, Dr. Moreno said.

In PROSPECT (Providing Regional Observations to Study Predictors of Events in the Coronary Tree), those NCLs that caused future events after 3.4 years had an average percent stenosis of 32±21%, and almost all of them had less than 71% diameter stenosis. At the time of the event, however, they had expanded to a luminal diameter stenosis of 65±16%, showing that high-risk plaques—with angiographic stenoses ranging from mild to severe—can undergo rupture and thrombosis.

The strongest independent predictor among morphological features for clinical events (with a hazard ratio of 5) was plaque burden along with thin-cap fibroatheroma (verified by virtual histology intravascular ultrasound) and a minimal luminal area ≤4 mm2.5 The incidence of future MACE at 3.4 years, when all three of these parameters were present, was 18% (hazard ratio = 11.5). Such high-risk plaques were rare, however, occurring at a rate of 4.2%. The substantial evolution of plaques that were not critically occlusive at the outset confirms Dr. Narula’s expansion hypothesis, Dr. Moreno said. Similarly, a study by Motoyama et al. has shown a 22.5% risk of developing ACS after 2 years for low attenuated and positively remodeled, large, nonobstructive plaques identified by coronary computed tomography angiogram (CTA).⁸

Coming up will be Dr. Stone’s PROSPECT II, also a study of ACS natural history, which will be conducted in conjunction with the Uppsala Clinical Research Center of Uppsala, Sweden. It will use near infrared spectroscopy, which provides a lipid core burden index, a measure of the proportion of lipid in plaque, and IVUS to track 900 patients through the Swedish Cardiac Arrest Register (SCAR). In an interview for CSWN: Interventions, Dr. Stone said the goal is to learn how well detection of lipid-rich plaques correlates with future clinical events.

“But what’s even more exciting is the 300-patient substudy called PROSPECT ABSORB,” he said. “It will include patients who by gray-scale IVUS have plaque burdens equal to or greater than 70%, a marker that has been shown to be the single, most high-risk characteristic—it signifies about a 10% event rate at that lesion site over the next 3 years.” Participants will be randomized to either a bioabsorbable vascular scaffold or to guideline-directed medical therapy. “We hope it will show us, in a prospective fashion, that near infrared spectroscopy will add a substantial diagnostic value on top of gray-scale IVUS for predicting which lesions present future coronary risk.”

Dr. Stone then voiced the bedrock concern, echoed by all of the subjects interviewed: “It’s hard to recommend routine adoption of any invasive imaging modality unless you can show that it informs treatment that improves outcomes. That’s why ABSORB is a first step.” If it is positive, he said, it will likely lead to a large-scale randomized controlled trial of infrared spectroscopy combined, perhaps, with gray-scale IVUS. Dr. Stone and Dr. Narula have reviewed the current state-of-knowledge in their editorial published recently in JACC Cardiovascular Imaging,⁹ that reconciles the debate between Drs. Nicoli and Ambrose.¹⁰

Should We Leave the FFR-Positive Lesions to OMT?

With less than 15% of FFR-positive lesions leading to cardiovascular events, the option of treating them with OMT merited investigation, Dr. Moreno said. The YELLOW trial (reported by Drs. Kini, Narula, and Moreno in JACC last year¹¹) randomized patients with multivessel CAD and stable angina to high-dose rosuvastatin at 40 mg or standard-of-care statin therapy. After culprit lesions were treated with PCI, investigators used near infrared spectroscopy and IVUS to evaluate the FFR-positive NCLs. The outcome measure after 7 weeks, lipid content reduction, was lowered significantly in the intensive-therapy group. Also, in those plaques with substantial regression, there was a trend toward FFR improvement. Neither treatment affected fibrous plaques.

The implication, according to Dr. Moreno, is that high-dose statins and other improvements in medical therapy, potentially including PCSK9 (proprotein convertase subtilisin/kexin type 9) inhibitors, may have strong potential for further reduction of future events.

Dr. Moreno noted another promising invasive imaging modality, OCT, has resolution 10-fold that of IVUS, helping to identify fibrous plaque thickness and discriminate between fibrous/fibro-calcific or lipid-rich plaques. Dr. Narula has compared the relative merits of the imaging modalities in his editor’s page in JACC Cardiovascular Imaging.¹³

Systemic Approach Versus Lesion Approach

Despite the considerable progress detailed above in showing the lumen to be “an innocent bystander,” the authors are well aware of persistent unresolved questions and issues. They point to a void of understanding regarding the evolution of NCLs toward cardiac events and about MACE occurrence in previously asymptomatic subjects. Dr. Narula said that while advances have been made in detecting the plaques that are vulnerable and at imminently high risk, predictive strategies are lacking for the third of events caused by plaques that do not rupture but rather erode.

Noninvasive techniques, such as CT angiography, can detect plaques with a necrotic core, and systemic C-reactive protein levels suggest plaque inflammation, but they do not tip off clinicians as to which plaques will grow and become vulnerable or identify which ones will erode. Dr. Stone did note that in more than 100 patients with MI who were assessed with infrared spectroscopy, the lipid content was high in about 95%. “That might actually mean that the vast majority of lesions that are causing ACS are thin-cap fibroatheromas—more so than you would expect pathologically—or that lesions that are prone to erosion are also very lipid-rich,” he said. Optical coherence tomography may be valuable in this context, he added, but has not yet been studied.

Learning which NCLs will rear up and become menacing, Dr. Narula said, is a great challenge. Using the analogy of what he referred to as a “banana ripeness index,” learning which green ones to treat before they become brown would require “quite an exercise. You cannot put a patient on the table again and again and again to do imaging!”

Maybe where it’s all heading is toward a systemic approach rather than a treating-every-plaque-in-sight approach. Dr. Narula’s own journey as a researcher has radically transformed his thinking about the fundamental strategy for alleviating atherosclerotic disease. “As a person who has spent all his life researching vulnerable plaques—imaging them, identifying them and their pathogenesis and pathology—at the end I come to the conclusion that, while we may need to define strategy to manage those that might be at imminently high risk, it is more important to prevent plaques than to treat them.”

Dr. Narula has a few ideas about how to accomplish this: “Government has to subsidize lifestyle intervention through laws or incentives or any other means. We have to emphasize the prevention part and support services. Of course we need statins and other new drugs. But I’m willing to go a step further than prevention and say we need cardiovascular health promotion, with more stress on lifestyle behaviors, nutrition, and exercise—we need more bicycle lanes!”

---

References

1. Moreno PR, Narula J. J Am Coll Cardiol. 2013 July 29. [Epub ahead of print].
2. Glagov S, Weisenberg E, Zarins CK, et al. N Engl J Med. 1987;316:1371-5.
3. Ambrose JA, Tannenbaum MA, Alexopoulos D, et al. J Am Coll Cardiol. 1988;12:56-62.
4. Narula J, Nakano M, Virmani R, et al. J Am Coll Cardiol. 2013;61:1041-51.
5. Moreno PR, Purushothaman KR, Fuster V, et al. Circulation. 2002;105:2504-11.
6. Stone GW, Maehara A, Lansky AJ, et al. N Engl J Med. 2011;364:226-35.
7. De Bruyne B, Pijls NHJ, Kalesan B, et al. N Engl J Med. 2012;367:991-1001.
8. Motoyama S, Sarai M, Harigaya H, et al. J Am Coll Cardiol. 2009;54:49-57.
9. Sone GS, Narula J. JACC Cardiovasc Imag. 2013;6:1124-6.
10. Niccoli G, Stefanini GG, Capodanno D, et al. JACC Cardiovasc Imag. 2013;6:1108-1114.
11. Kini AS, Baber U, Kovacic JC, et al. J Am Coll Cardiol. 2013;62:21-9.