Non-ST Elevation Acute Coronary Syndromes: Targeted Imaging to Refine Upstream Risk Stratification

Non-ST-segment elevation acute coronary syndromes (NSTE-ACS), encompassing non-ST-segment elevation myocardial infarction (NSTEMI) and unstable angina (UA), lack the declarative electrocardiographic findings that readily identify patients with ST-elevation myocardial infarction (STEMI) yet comprise >70% of all ACS^1–3. Unlike STEMI patients who receive uniform treatment to restore flow in an occluded artery, NSTE-ACS patients (who often present first to an emergency department) have varying degrees of coronary obstruction, undergo more heterogeneous management and have worse long-term outcomes^{2, 4}. Despite guidelines, there remains inconsistent utilization of routinely-advocated medical therapies and invasive evaluation⁵ that reflect uncertainty in initial evaluation. Greater age and comorbidities plus varied coronary artery disease (CAD) severity further complicate decision-making. And while cardiogenic shock, heart failure and arrhythmias may be less than in STEMI, NSTE-ACS patients suffer more recurrent events and worse long-term outcomes¹.

As our understanding of NSTE-ACS pathophysiology has grown, several non-invasive imaging techniques have emerged that offer high-resolution characterization of relevant markers of pathophysiology like coronary atherosclerotic plaque and salvageable myocardium. Incorporating such techniques into evaluating the ED patient with suspected or established NSTE-ACS promises improved outcomes by demonstrating patient-specific, mechanistically-based targets for therapy. This review summarizes NSTE-ACS, current imaging approaches and emerging imaging advances to improve diagnosis, treatment and outcomes.

PATHOPHYSIOLOGY

NSTE-ACS is most commonly caused by disruption of a coronary artery atherosclerotic plaque, with myocardial ischemia and injury often resulting from partial or intermittent occlusion along the ischemic cascade⁶. Other causes beyond the focus of this work include embolism and revascularization⁷. Events in the coronary and downstream myocyte together offer targets for imaging in NSTE-ACS.

Events in the coronary artery

Coronary plaque formation has been well underway prior to NSTE-ACS presentation⁸. The plaque prone to cause ACS has a thin fibrous cap, large lipid pool and is susceptible to disruption via inflammation, metalloproteinases and collagenases (Fig. 1). Intra-plaque hemorrhage, plaque neovascularization and outward (positive) vessel wall remodeling further identify the vulnerable plaque. After a triggering event, a superimposed thrombus forms upon contact between luminal blood and the plaque’s highly thrombogenic lipid core; in 20–40% of ACS, coronary thrombosis occurs with only superficial plaque erosion (Fig. 2, Movies 1–4) rather than plaque rupture. Myocyte necrosis ensues via either subtotal occlusion or transient episodes of flow reduction (Fig. 1). In UA, subtotal occlusion and transient ischemia stop short of myocardial necrosis. Embolization of plaque and thrombotic materials may occlude the downstream microvasculature⁹. External compression by edematous tissue, in situ thrombosis, vasospasm, leukostasis and reperfusion injury exacerbate microvascular obstruction and myocyte necrosis, particularly after mechanical reperfusion.

Figure 1.

The typical pathogenesis of NSTE-ACS begins with an atherosclerotic plaque. Disruption of the vulnerable plaque results in thrombus formation as coagulation factors come into contact with intraplaque elements. Embolic debris may travel downstream and lodge in microvessels. Injured myocardium develops edema and subsequent necrosis that begins in the subendocardium and may extend outward if at-risk regions are not salvaged.

Figure 2.

ECG, CMR, ultrasound and angiographic findings are shown in a 69 year-old female with chest and epigastric pain that had resolved by the time she presented to the ED. Her ECG was concerning for injury (A) and initial troponin was mildly elevated at 0.14 ng/mL. However, because of lack of symptoms by the time of presentation, invasive angiography was deferred to the following morning. T2-CMR in vertical-long axis (B) and horizontal long-axis (C) planes showed increased signal intensity in apical myocardium, beyond T2 increase in the stagnant apical blood. Apical dyskinesis was also evident (end-diastolic (D, E) and end-systolic (F, G) frames), and LGE (H, I) showed a large area of signal enhancement in the apical myocardium as well as thrombus in the apical LV. Echocardiography without contrast (J) was also concerning for LV apical thrombus. Invasive coronary angiography (K) showed serial nonobstructive plaques in the left anterior descending coronary artery. The ACS event evident by myocardial imaging was ascribed to plaque erosion. See Movies 1–4.

Events in the myocardium

Myocyte death may occur via oncotic or, to a lesser extent, apoptotic pathways^{10, 11}. An intact membrane infers potential for myocyte recovery; because water balance dictates many processes that impact membrane integrity, attention to edema is particularly relevant in recognizing at-risk but salvageable myocardium (Fig. 1). Impaired oxygen delivery arrests oxidative phosphorylation and blunts energy production required to prevent intracellular sodium and chloride (and, therefore, water) accumulation. Lactate-induced acidosis releases water molecules from proteins, increasing the intracellular fraction of unbound vs. bound water. Increased membrane permeability ultimately leads to physical disruption of the sarcolemma and cell death.

Timely reperfusion can salvage myocardium, yet reperfusion itself may contribute to cardiomyocyte death¹⁰. As post-ischemic blood flow is restored, hyper-osmotic extracellular fluid is replaced by normo-osmotic blood, creating an osmotic gradient favoring the movement of water into myocytes. Fragile sarcolemma resulting from energy-starved processes suffer mechanical stress-induced damage, and cells undergo necrosis¹¹. Reperfusion may also produce myocyte death via inflammation, oxidation, apoptosis and autophagy¹².

CURRENT DIAGNOSIS AND RISK STRATIFICATION IN NSTE-ACS

Diagnosis

Historical descriptors such as substernal chest pain or recurrent angina equivalents in patients with known CAD facilitate diagnosis, and high-risk physical examination markers are well-established. ACS mimics such as pericarditis, myocarditis or non-cardiac causes warrant consideration.¹ The ECG in NSTE-ACS may show ST-segment depression, T-wave flattening or inversion, or even transient ST-segment elevation; variation with symptoms favors an ACS diagnosis. Biomarkers such as cardiac troponins are crucial in recognizing myocardial injury and to differentiate NSTEMI from UA. Emerging biomarkers may better rule-in or rule-out ACS¹³; circulating biomarkers alone, however, cannot localize salvageable, at-risk myocardium or distinguish ACS mimics.

Risk stratification

Initial risk stratification is important in NSTE-ACS since the benefit of intensive therapies varies with risk¹⁴. Currently available risk stratification schemes (Table 1^15–17) offer good predictive value for adverse cardiac events at 30-days and 1 year.^18–20 However, risk scores do not identify targets to lower risk, underscoring the difference between risk markers and risk factors. Imaging may not only improve risk stratification but also treatment selection by directly visualizing treatable targets in higher risk individuals.

Table 1.

Risk Scores’ Mortality Predictive Power in Non-ST Elevation Acute Coronary Syndromes

	Major elements	Endpoint	C index
TIMI15	age, risk factors, prior coronary stenosis, ST-segment deviation, angina frequency, use of aspirin, serum biomarkers	all-cause mortality and non-fatal MI at one year	0.6619
PURSUIT16	age, sex, worst Canadian Cardiovascular Society-class in previous 6 weeks, signs of heart failure, ST-segment depression	30-day mortality	0.7918
GRACE17	age, Killip class, systolic blood pressure, ST-segment deviation, cardiac arrest during presentation, serum creatinine level, cardiac biomarkers, heart rate	6-month mortality	0.8018

CURRENT IMAGING APPROACHES IN NSTE-ACS

Myocardial imaging – contractile function and perfusion

Left ventricular ejection fraction (EF) is a critical parameter in conservative management pathways¹; echocardiography, nuclear scintigraphy, or invasive ventriculography most commonly provide EF in NSTE-ACS. More refined detection of regional dysfunction is afforded by strain measurement; Eek et al. reported that strain echocardiography was able to predict acute coronary occlusion with a sensitivity and specificity of 85% and 75%, respectively, in NSTE-ACS patients without prior MI²¹. In patients presenting to the ED with chest pain and no ST-segment elevation, visually-assessed regional wall motion abnormality using contrast echocardiography (CE) better predicted adverse effects vs. traditional clinical markers—more so than CE-derived myocardial perfusion²². In such patients, normal CE-based wall motion may have incremental negative predictive value over low TIMI score²³.

Myocardial perfusion imaging is more commonly performed with nuclear techniques such as single-photon emission computed tomography (SPECT-MPI).^24–26 Presence vs. absence of perfusion defects may be useful to distinguish high-risk from low-risk ED chest pain patients ²⁷.

SPECT may be constrained by poor spatial resolution–particularly for the NSTE-ACS patient with subendocardial ischemia²⁸. Prior CAD may make it difficult to know if a resting perfusion SPECT defect is due to prior infarction or represents a new region of at-risk myocardium. Use of positron emission tomography (PET)²⁹ and emerging SPECT tracers³⁰ may further refine recognition of myocardium at risk in ACS patients, and newer SPECT techniques have considerably shortened acquisition times^{31, 32}.

One such tracer is β-methyl-p-[¹²³I]-iodophenyl-pentadecanoic acid (BMIPP), a fatty acid analog which demonstrates reduced uptake in ischemic myocytes and indicates a cellular shift from fatty acid metabolism to glucose metabolism. In 448 ED chest pain patients without ST elevation, Kontos et al.³⁰ showed that BMIPP imaging was more sensitive (42.9% vs. 73.0%) and more specific (60.9% vs. 63.2%) in diagnosing ACS. “Ischemic memory” helped to maintain sensitivity if imaging was performed 12–30 hours after symptom resolution vs. within 12 hours of such (68.6% vs. 77.1%).

Myocardial perfusion imaging can also be performed with cardiac magnetic resonance (CMR), detecting patients with coronary artery stenoses of ≥70% with similar specificity and better sensitivity vs. SPECT in the recently-published landmark CE-MARC trial³³. Another recently-published multicenter, multivendor study showed superior sensitivity and inferior specificity of perfusion-CMR to detect CAD vs. SPECT³⁴. Kwong et al. demonstrated 84% sensitivity and 85% specificity for ACS using CMR without stress in 161 patients presenting to the ED with chest pain and no ST-segment elevation³⁵. CMR was more sensitive than ECG, troponin-I and TIMI scores in detecting ACS. Using adenosine stress perfusion CMR, Plein et al. showed even more accurate identification of NSTE-ACS patients requiring revascularization (sensitivity and specificity 97% and 83%, respectively)³⁶.

Late gadolinium-enhancement CMR (LGE) is an extensively validated technique to visualize in vivo irreversibly-injured myocardium^{37, 38}. Tissue characterization also assists in distinguishing true NSTE-ACS from ACS mimics: ACS-related hyperenhancement is subendocardial, LGE-positivity in myocarditis shows an epicardial or midwall distribution, and LGE-positivity in tako-tsubo cardiomyopathy is typically minimal^{39, 40}.

Coronary imaging

Noncontrast computed tomography (CT)-based coronary artery calcium scoring (CAC) reflects coronary artery plaque burden, demonstrating a positive predictive value (PPV) of nearly 100% for atheromatous coronary plaque⁴¹. Laudon et al. found that CAC≤36 among patients who presented to the ED with chest pain strongly predicted non-cardiogenic chest pain, while CAC>36 predicted cardiogenic chest pain as defined by exercise treadmill testing, SPECT MPI, stress echocardiography or invasive angiography⁴². It should be noted that the negative predictive value of CAC varies by the population in which it is used: 38% of Asian chest pain patients presenting to an ED who subsequently developed ACS had a calcium score of 0⁴³.

Coronary computed tomographic angiography (CCTA) has recently emerged as a reliable non-invasive modality for assessing epicardial luminal stenosis, with multiple studies demonstrating high diagnostic performance compared to invasive angiographic measures^44–46 (Fig. 3, Movies 5–7). In ED patients with acute chest pain, negative initial troponin and ECG without diagnostic changes of ischemia, Hoffmann et al. reported that CCTA that showed no evident coronary plaque had 100% negative predictive value for ACS. Positive predictive value for ACS of coronary stenosis by CCTA was considerably lower (35%), particularly in patients >65 years of age⁴⁷. In a smaller ED study of biomarker-negative chest pain patients with not only nonischemic presenting ECG but also normal or nondiagnostic treadmill stress ECG, CCTA fared better: positive and negative predictive values for ACS were 87% and 100%, respectively⁴⁸. And in a landmark multicenter trial, Litt et al. showed that low-to-intermediate-risk ACS patients randomized to undergo CCTA had a higher rate of discharge from the ED, shorter length of stay, and higher rate of coronary disease detection vs. those assigned to traditional care⁴⁹. Similar results of the ROMICAT II trial⁵⁰ are anticipated.

Figure 3.

A 59 year-old male with a history of coronary stent placement in the left anterior descending artery presented to the emergency department for evaluation of atypical chest pain. CMR showed a perfusion abnormality with stress (A) not present at rest (B) and corresponding to a region of infarct scar by LGE (C). SPECT imaging demonstrated a perfusion defect under stress (E) compared to at rest (F). Invasive coronary angiography (D) showed patent LAD stent. The presence of a perfusion abnormality in the absence of an anatomic target for repeat intervention led to his medical regimen being intensified. At a subsequent ED visit for chest pain, coronary CTA was performed that demonstrated a patent stent with mild proximal malapposition (G, arrow). As a result of imaging, the medical regimen was further optimized and repeat catheterization was deemed unnecessary. See Movies 5–7.

EMERGING IMAGING APPROACHES TO NSTE-ACS

CT perfusion

Early data suggest feasibility of CT perfusion (CTP) at both rest and stress vs. SPECT or CMR (Table 2)^51–56. Rest findings alone can be derived at the time of CCTA. Ongoing multicenter studies (clinicaltrials.gov NCT00934037 and NCT01334918) will assess the robustness of early single center data.

Table 2.

Studies Comparing Stress CT Perfusion to Other Modalities

Author	N	Comparison	Major Findings
George56	40	CTP vs. SPECT	Sensitivity/specificity 70/51%
Blankstein55	34	a)CTP vs. ICA, b)CTP vs. SPECT	Sensitivity/specificity: a)92/67%, b)84/80%
Okada54	47	CTP vs. SPECT summed stress score	Agreement in 107/141, correlation r=0.56
Tamarappoo53	30	Agreement for perfusion deficit	Good agreement, κ=0.71
Ho52	35	Dynamic CTP vs. SPECT	Sensitivity/specificity 83/78%
Ko51	41	CTP vs. CMR	Sensitivity/specificity 89/78%

Plaque characterization

In addition to stenosis severity assessment, CCTA provides non-invasive visualization of coronary artery plaque and wall characteristics, in a manner similar to invasive atherosclerosis imaging methods⁵⁷. Limited studies have investigated the clinical utility of CCTA to detail these atherosclerotic plaque characteristics (APCs, Fig. 4) as predictive metrics to identify individuals at heightened risk for unheralded or recurrent NSTE-ACS. APCs examined by CCTA may include plaque burden, plaque composition including low attenuation plaques (LAP, consistent with lipid-rich necrotic cores) and spotty calcifications, positive remodeling (PR) and other features including ulcer-like enhancement spaces within plaque (i.e., “napkin-ring sign”)⁵⁸. In the largest of these studies, Motoyama et al. demonstrated predictive value of APCs for future ACS, focusing on LAP with Hounsfield unit (HU) densities <30 and PR with remodeling indices >1.10⁵⁹. Importantly, all events observed in this study were for patients with <75% stenosis. Similarly, Kristensen et al. who performed CCTA at the time of NSTEMI showed that baseline volume of non-obstructive plaque, especially noncalcified plaque, predicted recurrent adverse events⁶⁰. While constrained by a small number of events, both studies suggest a potential prognostic role for CCTA-based APC assessment that may identify plaques that subsequently cause ACS.

Figure 4.

CCTA demonstrates adverse atherosclerotic plaque characteristics such as: A) stenosis of the lumen, where luminal diameter at the site of stenosis shows significant and progressive narrowing compared to the proximal more normal-appearing coronary segment, B) low x-ray attenuation plaques (LAP) indicated by intra-plaque Hounsfield unit (HU) values <30 and C) positive vessel wall remodeling (PR) defined by an intraplaque arterial diameter-to-normal coronary segment diameter >1.10.

Recently, Ozaki et al. called into question CCTA’s ability to distinguish intact fibrous cap plaques from culprit lesions in a mixed group of ACS and stable angina patients who additionally underwent optical coherence tomography, angioscopy and intravascular ultrasound⁶¹. While LAP and PR by CCTA occurred more commonly in ACS with ruptured vs. intact fibrous cap, neither distinguished between intact fibrous cap ACS and stable lesions. Noting that a significant proportion of NSTE-ACS may result from erosion of thrombogenic plaques with intact fibrous cap, further work is needed for noninvasive techniques to be able to demonstrate such micron-scale characteristics.

The limits of agreement for APCs by CCTA vs. intravascular ultrasound (IVUS)—including plaque volume by composition—are wide, and can range up to 269% in our own laboratory’s experience⁶². Similarly, although plaques with large (>10%) intraplaque necrotic cores demonstrate lower HU values, a wide range of HU values in necrotic core overlaps with those of non-necrotic core fibrous plaque components⁶³. Thus, APCs by CCTA may represent a novel approach for enhanced risk stratification, but required still are i) confirmation of the prognostic relevance of APCs in larger prospective studies, ii) determination of the independent and incremental value of APCs beyond stenosis severity alone and iii) therapeutic outcomes studies that stratify treatment approaches based upon CCTA-based APCs.

One other exciting CCTA development that offers a physiological assessment of anatomic stenosis is the non-invasive estimation of fractional flow reserve (FFR) via computational fluid dynamics. FFR, commonly regarded as the reference standard for diagnosis of lesion-specific ischemia, has been demonstrated to improve event-free survival for stable patients with multivessel disease⁶⁴. FFR derived from CT (FFRCT) is a non-invasive method that does not require any additional imaging, medication administration or radiation exposure. Encouraging results from a recent prospective, multicenter study of FFRCT vs. invasive FFR⁶⁵ behoove its evaluation in the broader ED chest pain population.

Myocardial imaging –necrosis and edema

Ideally, non-invasive myocardial imaging in NSTE-ACS patients would identify not only necrotic myocardium (e.g. using standard LGE-CMR) but also at-risk, not yet irreversibly-injured muscle. Imaging myocardial water may provide this information based on classic preclinical studies that identified the edematous myocyte as potentially salvageable¹¹. Given its sensitivity to not only total change in myocardial water content but also changes in protein-binding of water⁶⁶, T2-CMR holds significant appeal (Fig. 5, Movies 8–9).

Figure 5.

ECG, CMR and angiographic findings are shown in a 50 year-old diabetic male who presented with intermittent chest pressure; initial troponin-I values were 0.44 and then 0.21 ng/mL (limit <0.11), with no symptoms upon arrival post-transfer from a regional medical center. Invasive angiography was planned for the following morning, preceded by CMR with T2-weighted imaging as part of a research protocol. While ECG was unremarkable (A), T2-CMR in vertical-long axis (B) and horizontal long-axis (C) planes showed increased signal intensity in septal and apical myocardium. Regional wall motion appeared normal in these planes (end-diastolic (D, E) and end-systolic (F, G) frames), and LGE (H, I) showed no evident myocardial injury. Invasive coronary angiography later that morning (J) showed high-grade stenoses in the LAD and first diagonal branch treated with percutaneous coronary interventions. Interventions aimed at rescuing the imaged salvageable myocardium were consideredand another stent was placed in the proximal RCA (K) the following day. See Movies 8–9.

Combining LGE and T2 may enhance our understanding of the time course of myocardial changes in ACS. Dall’Armellina et al. performed 4 serial CMR exams that included T2-weighted imaging in MI patients, showing that edema persists for at least one week post-MI and allows for retrospective identification of this signature⁶⁷. Abdel-Aty et al. showed in a landmark study that this combination could distinguish old from acute infarction⁶⁸. Cury et al. showed the utility of T2-CMR in accurate detection of ACS in the ED-chest pain patient⁶⁹; a CMR protocol that included T2-weighted imaging, perfusion, cine and LGE acquisitions improved the c-statistic over traditional risk assessment from 0.695 to 0.958 (p<0.0001). Berry et al. performed bright-blood T2-weighted CMR in acute MI patients and showed excellent delineation of myocardial salvage in comparison to traditional measures of area at risk⁷⁰. Our group has evaluated the potential utility of T2-CMR in upstream risk stratification, showing that presence of edema had incremental value over clinical predictors such as TIMI risk score and troponin-I level to identify NSTE-ACS patients that subsequently require coronary revascularization⁷¹. T2 mapping offers a further advance in quantifying myocardial edema and overcoming limitations of traditional T2-weighted imaging^{72, 73} (Fig. 6).

Figure 6.

A 56 year-old male with a history of MI, CAD, and multiple coronary stents presented to the emergency department for evaluation of chest pain with epigastric symptoms. ECG showed right bundle branch block as well as ST-segment depression and T-wave inversion in leads V1 and V2 (A). His TIMI score was 4 and initial troponins were negative. Upon CMR examination, mild wall motion abnormalities were seen (end-diastolic (B) and end-systolic (C) frames in the vertical long-axis view). T2 mapping revealed elevated T2 values in areas absent of any LGE enhancement, suggestive of myocardium at risk (T2 and LGE images with vertical long-axis (D, E), 3-chamber (F, G), and short-axis (H, I) views). Note the region of LGE hyperenhancement in (I) without a corresponding T2 elevation in (H) indicating the presence of an old infarct and enabling identification of myocardium at risk amidst multiple anatomic targets for revascularization. Invasive angiography showed severe triple-vessel disease with chronic total occlusion of mid RCA with collateral flow and moderate in-stent restenosis throughout the circumflex territory (J).

CONCLUSIONS AND FUTURE DIRECTIONS

Cost considerations

The cost-effectiveness of imaging remains an important factor in determining its use in the ED for NSTE-ACS patients. Udelson et al. in a randomized trial showed lower hospitalization rates and increased discharge home using SPECT-MPI to rule out cardiac ischemia in patients presenting to the ED with chest pain and non-diagnostic ECG⁷⁴. Similarly, Stowers et al. showed lower treatment costs and length of stay in ED patients with chest pain and non-diagnostic ECG by incorporating SPECT⁷⁵.

Goldstein et al. randomized ED chest pain patients to CCTA or standard evaluation, and further prescribed to the CCTA group i) discharge of patients with minimal disease, ii) catheterization for those with stenosis >70%, and iii) stress testing for individuals with intermediate lesions or non-diagnostic scans⁷⁶. While diagnostic accuracy was similar in both groups, the CCTA cohort had shorter median time to definitive diagnosis (3.4 hours vs. 15.0 hours) and reduced costs of care ($1,586 versus $1,872).

Miller et al. recently demonstrated in a randomized trial of intermediate-risk ED patients that stress CMR delivered in an observation unit offers cost savings during the index visit and incurred lower total costs over the ensuing year ($3,101 vs. $4,742) without an increase in adverse cardiac events⁷⁷. The same group subsequently showed that CMR does not afford such savings in low-risk ED patients⁷⁸, underscoring the need for careful clinical evaluation before implementing a testing strategy.

Cost data from retrospective studies caution against overuse of advanced imaging^{79, 80}. Implementation logistics also require consideration: access to the most appropriate technology and expertise is required at the right time for the right patient. Eliminating barriers as was done with deployment of coronary angiography in STEMI may be needed for non-invasive imaging to deliver on the promise of reducing uncertainty in NSTE-ACS diagnosis and management.

Molecular Imaging

While clinical trials of molecular imaging in NSTE-ACS ED patients have not yet been performed, the appeal of targeting relevant cellular and molecular pathophysiological processes may be realized with increased availability of such imaging approaches. With imaging agents that target specific molecules, cellular structures and processes can be probed at a level not possible with current imaging techniques.

Selectins, integrins and adhesion molecules may be suitable vascular imaging targets, coupling modalities like CMR with PET or SPECT with CTA. Similarly, plaque imaging may target species of destabilization⁸¹ or metabolism⁸². Proton-based CMR, with direct visualization of hydrogen in various forms may be used to track water molecules in the pathophysiologic changes of ischemia-reperfusion injury. Hofstra et al. used Tc-99m-labeled annexin to detect apoptotic myocytes in patients post-MI⁸³. Annexin can also be tagged with CMR to detect apoptosis⁸⁴. Finally, MR spectroscopy techniques to visualize subcellular energetics⁸⁵ and even acidosis⁸⁶ are exciting advances to refine preclinical understanding and to develop novel therapeutic strategies.

However, even with useful affinity ligands as imaging agents, molecular imaging still faces many challenges before it can be considered a useful tool in the ED. Pharmacokinetic concerns include target molecule abundance, in vivo specificity, binding stability, and target-to-background ratios. Limited spatial and temporal resolution, even with state-of-the-art technologies, may not be able to consistently image small-caliber, rapidly-moving coronary arteries.

Closing remarks

Patients presenting with chest pain may currently be assigned to intensive medical therapy alone despite having, for instance, obstructive single vessel disease⁸⁷. Practitioners are all too familiar with instances of ACS following institution of medical management for ‘stable’ CAD; these challenge the notions that current diagnostic modalities are sufficiently predictive, and that current therapies are sufficiently preventive (Fig. 7). And despite widespread use of high-sensitivity serum biomarkers, risk scores and early invasive cardiac catheterization, only 55% of patients with NSTE-ACS have disease requiring PCI or bypass surgery, while the balance receive medical therapy alone due to either extensive CAD or nonobstructive CAD despite laboratory evidence of myocardial necrosis^{88, 89}. Patterns of care for patients with NSTE-ACS have been described as ‘paradoxical’: only about 30% of high-risk patients are managed with an early invasive strategy compared to 75% of low-risk patients^{90, 91}. Increasing recognition of bleeding risk with aggressive use of anti-platelet and anti-coagulant therapies underscores the need for better selection of patients likely to benefit from early invasive management⁹².

Figure 7.

Limitations of current diagnostic and treatment approaches are highlighted in this 44 year-old nonsmoking male who presented with stable angina. CTA identified single-vessel coronary stenosis (A, arrow) in the right coronary artery (RCA) prompting immediate institution of aggressive medical therapy. After several weeks, he presented to the emergency department with unstable symptoms, and emergent coronary angiography identified plaque rupture at the site of RCA APCs that included low attenuation plaque with HU<30 (B, arrow) and positive remodeling (C) indicated by greater vessel wall diameter at the site of plaque (distal open arrows) vs. the reference segment (proximal closed arrows).

Non-invasive imaging targeting the underlying coronary and myocardial events that lead to adverse clinical sequelae can help reduce uncertainties in initial diagnosis and treatment. Recent advances in non-invasive plaque and flow assessment with CT and myocardium at-risk assessment with CMR warrant further prospective evaluation of their ability to i) improve selection of patients for more costly, higher-risk invasive and adjuvant medical therapies, ii) reduce morbidity and mortality via timely rescue of salvageable myocardium and iii) improve distinction of NSTE-ACS from ACS mimics that require distinct treatments.

While techniques for noninvasive detection of high-risk features have developed rapidly, trials addressing management of such high-risk features are now needed to deliver on the promise of predicting, preventing and reducing the economic and societal burden of NSTE-ACS. Delivery on this promise will also require improved therapeutic options for e.g. the non-critical coronary stenosis due to plaque with adverse features supplying at-risk but viable myocardium. Heterogeneous care that delivers inconsistent outcomes should sufficiently motivate the imaging community, clinicians and scientists to develop better solutions for these patients.