Skip to content

Ischemic heart disease — pathophysiology and plaque biology

TL;DR — Coronary atherosclerosis is a diffuse arterial process driven by retention of apoB-containing lipoproteins, maladaptive immune responses, smooth-muscle and extracellular-matrix remodeling, calcification, and thrombosis. Acute coronary syndromes arise through several substrates—most prominently fibrous-cap rupture and plaque erosion—rather than a single “vulnerable plaque” phenotype (Falk 1995, PMID 7634481; Braunwald 2013, PMID 23473110). PROSPECT showed that non-culprit lesions with large plaque burden, small lumen, and thin-cap fibroatheroma predict later events, but their low positive predictive value exposed the weakness of lesion-by-lesion prediction (Stone 2011, PMID 21247313). Systemic LDL lowering and anti-inflammatory therapy reduce events, supporting causal roles for lipid exposure and immune activation (CTT 2010, PMID 21067804; Ridker 2017, PMID 28845751). PREVENT reopened the controversy by randomizing prophylactic PCI of high-risk, non-flow-limiting plaques: at 2 years the target-vessel composite occurred in 0.4% with PCI versus 3.4% with medical therapy alone (absolute difference −3.0 points, 95% CI −4.4 to −1.8), an event rate so low in both arms that generalization beyond highly selected imaging-defined lesions remains uncertain (Ahn 2023, PMID 37271356; Park 2024, PMID 38604213).

From lipoprotein retention to plaque

The initiating lesion is not simply “cholesterol in the wall.” ApoB-containing particles cross the endothelium, become retained and modified in the intima, and elicit endothelial activation, monocyte recruitment, and macrophage lipid uptake. Repeated injury and repair produce a heterogeneous plaque containing lipid, inflammatory cells, smooth-muscle-derived cells, collagen, proteoglycans, neovessels, hemorrhage, and calcium (Ueki 2024, PMID 39111840).

Stage Dominant processes Clinically observable correlate
Endothelial dysfunction Reduced vasodilator reserve, adhesion-molecule expression, permeability Abnormal coronary physiology may precede focal stenosis
Lipoprotein retention ApoB-particle trapping and modification Non-calcified plaque on CCTA
Foam-cell expansion Monocyte recruitment, lipid uptake, cell death Lipid-rich plaque by OCT/NIRS
Fibroatheroma Necrotic core plus fibrous cap Low-attenuation plaque; thin-cap phenotype
Remodeling Outward enlargement can preserve lumen Large plaque burden despite modest angiographic stenosis
Calcification Microcalcification and later dense calcium CAC burden; biomechanical effects vary by pattern
Complication Rupture, erosion, hemorrhage, thrombosis ACS, silent healing, or rapid progression

Angiography is a luminogram and can underestimate disease when outward remodeling preserves luminal diameter. This explains why a lesion can be biologically advanced yet non-obstructive, and why systemic prevention treats risk that focal PCI cannot see (Stone 2011, PMID 21247313).

Plaque rupture, erosion, and other ACS substrates

Falk's synthesis established plaque disruption as a key substrate of coronary thrombosis: a thin or disrupted cap exposes thrombogenic material and permits clot formation (Falk 1995, PMID 7634481). Rupture typically involves a lipid-rich necrotic core with cap inflammation and matrix degradation.

Plaque erosion differs: thrombus overlies an intact fibrous cap, often on a proteoglycan-rich lesion, and endothelial injury or denudation is implicated. Its recognition by intravascular imaging matters because erosion may have a different demographic distribution and could support less implant-intensive strategies in selected cases, though this is not routine standard care (Braunwald 2013, PMID 23473110).

Substrate Structural feature Typical detection Main uncertainty
Fibrous-cap rupture Discontinuity over necrotic core OCT; pathology Which ruptures become clinically occlusive
Plaque erosion Thrombus on intact cap OCT inference Whether stent-free treatment is durable
Calcified nodule Protruding calcific material OCT/IVUS Optimal device strategy and recurrence risk
Spontaneous dissection Intramural hematoma/intimal tear Angiography plus OCT/IVUS selectively Iatrogenic extension risk during imaging
Supply-demand injury No acute coronary thrombus required Clinical context and troponin kinetics Distinguishing type 2 MI from non-ischemic injury

The vulnerable-plaque hypothesis

The vulnerable-plaque model proposed that local morphology could identify the lesion destined to rupture. PROSPECT followed patients after ACS with three-vessel intravascular imaging: subsequent non-culprit events were associated with plaque burden ≥70%, minimal luminal area ≤4.0 mm², and thin-cap fibroatheroma (Stone 2011, PMID 21247313).

That study also exposed three limitations:

  1. Many morphologically high-risk plaques did not cause events during follow-up, so positive predictive value was low (Stone 2011, PMID 21247313).
  2. Plaques are dynamic; serial CT studies show progression, regression, and compositional change rather than a fixed binary state (Papadopoulou 2012, PMID 22421228).
  3. Patient-level systemic vulnerability—thrombotic state, inflammatory activity, and total plaque burden—may matter as much as any one lesion (Conti 2010, PMID 20556800).

AI-enabled CCTA can integrate quantitative plaque and hemodynamic features to predict ACS, but prediction performance must be tested prospectively across scanners, populations, and treatment eras before it becomes a treatment trigger (Koo 2024, PMID 38752951).

Low-attenuation plaque on CCTA predicted myocardial infarction in SCOT-HEART analyses, supporting incremental value beyond stenosis alone (Williams 2020, PMID 32174130). Yet an imaging association is not proof that mechanically treating the feature improves outcomes.

PREVENT and prophylactic lesion treatment

PREVENT randomized patients with high-risk, non-flow-limiting plaques to preventive PCI plus medical therapy or medical therapy alone. Its design operationalized vulnerability using intracoronary imaging and explicitly tested the step that observational studies could not: whether sealing selected plaques changes clinical outcomes (Ahn 2023, PMID 37271356).

PREVENT reported in 2024: among 1,606 randomized patients, the 2-year primary composite occurred in 3 (0.4%) with preventive PCI versus 27 (3.4%) with medical therapy alone (absolute difference −3.0 percentage points, 95% CI −4.4 to −1.8; p=0.0003), with no significant difference in death (0.5% vs 1.3%) or myocardial infarction (1.1% vs 1.7%) (Park 2024, PMID 38604213). A prespecified diabetes-stratified analysis found the same direction in both strata (0% vs 3.7% in diabetes; 0.5% vs 3.2% in non-diabetes, HR 0.16, 95% CI 0.05–0.55) with no significant interaction, so the result is not a diabetes-specific effect (Kim 2025, PMID 40439104). Composite construction dominated by revascularization and unstable-angina hospitalization in an open-label trial, very low absolute event rates, procedure-related harm, background medical-therapy intensity, and reproducibility of imaging selection all constrain generalization.

Question before preventive PCI Why it matters
Is the lesion truly high-risk by validated imaging criteria? Morphology has limited positive predictive value
Is the endpoint lesion-specific or patient-level? Systemic events may arise elsewhere
How intensive is background LDL and risk-factor treatment? Plaque composition is modifiable medically
What procedural harms occur immediately? A prophylactic procedure treats an asymptomatic lesion
Does benefit persist beyond the stent horizon? Neoatherosclerosis and new lesions remain possible

Inflammation as mechanism and treatment target

Human coronary plaques contain clonally expanded T cells with cross-reactivity to viral and self antigens, demonstrating structured adaptive immune activity rather than passive debris accumulation (Chowdhury 2022, PMID 35430876). Innate pathways, including inflammasome–IL-1β–IL-6 signaling, link cholesterol crystals and tissue injury to hepatic acute-phase responses and plaque activity (Ajala 2020, PMID 32880743).

CANTOS provided causal therapeutic evidence: canakinumab reduced recurrent cardiovascular events without lowering lipids, while increasing fatal infection risk (Ridker 2017, PMID 28845751). COLCOT and LoDoCo2 then showed that inexpensive colchicine reduced composite cardiovascular events after MI and in chronic coronary disease (Tardif 2019, PMID 31733140; Nidorf 2020, PMID 32865380).

The sequence across CANTOS, CIRT, COLCOT and LoDoCo2 is reviewed as evidence that target and population specificity matter more than the broad label “anti-inflammatory” (Samuel 2021, PMID 33687270).

Mechanistic imaging has not yielded a single simple explanation. In COLOCT, 12 months of colchicine after ACS increased minimal fibrous-cap thickness versus placebo (difference 34.2 µm, 95% CI 9.7–58.6; p=0.006) and reduced lipid arc and macrophage extension (Yu 2024, PMID 39166327). A LoDoCo2 CT subanalysis of 151 participants found no difference in pericoronary adipose-tissue attenuation (−79.5 vs −78.7 HU; p=0.236) but more calcified plaque with colchicine (volume 169.6 vs 113.1 mm³; p=0.041) (Fiolet 2025, PMID 40393691). Local plaque change is therefore measurable but modality- and population-dependent, and no single imaging change has been shown to mediate the event reduction.

Plaque regression and stabilization

LDL lowering changes both event probability and plaque phenotype. Serial imaging studies generally show modest change in plaque volume but larger shifts in composition, cap thickness, and lipid burden; therefore “stabilization” is not synonymous with dramatic angiographic regression (Ueki 2024, PMID 39111840).

Therapeutic axis Expected biologic effect Clinical evidence type
Statin/ezetimibe/PCSK9 inhibition Lower apoB flux; reduce lipid content; favor thicker cap Serial imaging plus large outcomes trials
Smoking cessation Reduce endothelial, inflammatory, and thrombotic activation Cohort and secondary-prevention evidence
Blood-pressure control Reduce endothelial injury and wall stress Outcomes trials across high-risk populations
Antiplatelet therapy Reduce consequences of plaque complication ACS and secondary-prevention trials
Colchicine/IL-1β inhibition Attenuate inflammatory amplification CANTOS, COLCOT, LoDoCo2
PCI Treat a focal lumen/lesion; does not remove systemic substrate Stable-CAD and ACS trials

This systemic–focal distinction explains why PCI can be lifesaving for an occlusive culprit lesion in ACS, relieve angina from a flow-limiting stenosis, yet fail to reduce spontaneous MI or death when applied routinely to stable stenoses (Boden 2007, PMID 17387127; Maron 2020, PMID 32227755).

Translating mechanism into a valid intervention

Mechanistic observation Necessary next experiment Failure if skipped
Marker associates with events External prospective validation Overfit risk prediction
Marker localizes a lesion Show reproducibility and positive predictive value Many unnecessary procedures
Drug changes a biomarker Demonstrate hard clinical outcomes Surrogate-only success
PCI seals a high-risk plaque Compare immediate harm and long-term lesion/patient events Target-lesion benefit obscures systemic risk
Immune target reduces MACE Quantify infection and competing mortality Net harm despite pathway proof
Imaging feature regresses Link change to outcomes independently of treatment Cosmetic surrogate interpretation

The pathway from plaque association to patient benefit is therefore longer than the pathway from stenosis to technical feasibility. PREVENT is important precisely because it crosses from morphology to randomized treatment, while CANTOS is important because it crosses from inflammatory association to outcomes (Ahn 2023, PMID 37271356; Ridker 2017, PMID 28845751).

Mechanistic triangulation: substrate, trigger, and host

Layer Human evidence Unresolved point
Lipid retention and inflammation Atherosclerosis is a chronic inflammatory response to retained/modified lipoproteins rather than a passive lipid deposit (Ross 1999, PMID 9887164; Libby 2002, PMID 12490960). Relative contributions change by lesion stage and cannot be inferred from one circulating marker.
Rupture versus erosion Pathology and intracoronary imaging distinguish thin-cap rupture from erosion and other substrates (Falk 1995, PMID 7634481; Braunwald 2013, PMID 23473110). In-vivo classification is imperfect, and treatment-without-stent hypotheses require randomized safety evidence.
Patient versus lesion vulnerability PROSPECT linked plaque burden, small lumen, and thin-cap fibroatheroma to later lesion events, but most high-risk-appearing plaques did not cause events (Stone 2011, PMID 21247313). Low lesion-level positive predictive value favors systemic prevention unless a strategy trial proves otherwise.
CCTA phenotype Low-attenuation plaque burden predicted MI in SCOT-HEART beyond conventional measures (Williams 2020, PMID 32174130). Scanner/protocol variation and diffuse patient-level risk complicate focal intervention.
Adaptive immunity Coronary plaque T cells can be clonal and cross-react with viral and self antigens (Chowdhury 2022, PMID 35430876). Antigen findings do not establish a safe target or whether clones drive versus respond to plaque.
Somatic hematopoiesis CHIP provides a plausible bridge between aging, innate inflammatory signaling, and ASCVD (Zuriaga 2023, PMID 34879980). Mutation-specific treatment selection remains investigational.
Regression/stabilization Intensive lipid lowering changes plaque volume/composition and reduces events (CTT 2010, PMID 21067804; Ueki 2024, PMID 39111840). Imaging regression is not a validated surrogate for the full clinical effect.
Focal prevention PREVENT randomized PCI of non-flow-limiting vulnerable lesions on top of medical therapy and reported a lower 2-year target-vessel composite (0.4% vs 3.4%) (Ahn 2023, PMID 37271356; Park 2024, PMID 38604213). Target-lesion reduction must be separated from patient-level death/MI, bleeding, and repeat procedures.

The field therefore contains two nonexclusive causal models. The systemic-host model treats apoB exposure, immunity, thrombosis, metabolism, and behavior across the arterial tree; the focal-lesion model attempts to identify and seal the plaque most likely to fail. PROSPECT supports focal heterogeneity but also demonstrates prediction's low precision, while systemic LDL and inflammatory outcome trials show that modifying the host can reduce events without knowing the future culprit lesion (Stone 2011, PMID 21247313; Ridker 2017, PMID 28845751; Sabatine 2017, PMID 28304224).

Mechanistic validity requires four linked observations: target presence in human culprit tissue, pharmacologic target engagement, change in a credible intermediate, and net clinical benefit. CANTOS supplies clinical pathway proof for IL-1β; CIRT is the instructive negative control; colchicine's discordant outcome and imaging programs show why no single imaging change should be assumed to mediate benefit (Ridker 2017, PMID 28845751; Ridker 2019, PMID 30415610; Yu 2024, PMID 39166327; Fiolet 2025, PMID 40393691).

Cell identity and inflammatory memory complicate the linear plaque model

The response-to-retention model starts before macrophage foam cells: subendothelial binding of apoB-containing lipoproteins to arterial matrix is the initiating event, supported experimentally by apoB100 engineered for weak proteoglycan binding producing fewer lesions despite hyperlipidemia (Williams 1998, PMID 9812202). Retention creates a local depot that can aggregate, oxidize, recruit innate immunity, and sustain lipid uptake; plasma concentration changes flux into this process but does not identify which existing lesion will fail.

Necrotic-core formation is partly failed resolution. MerTK cleavage reduces macrophage efferocytosis; cleavage correlated with necrosis and ischemic symptoms in human carotid plaques, while cleavage-resistant MerTK in LDL-receptor-deficient mice produced smaller necrotic cores, thicker caps, and a more pro-resolving lipid-mediator balance (Cai 2017, PMID 28067670). This is mechanistic triangulation, not proof that systemic MerTK manipulation is safe in humans.

Plaque cell labels are also unstable. Lineage tracing found conventional markers missed more than 80% of smooth-muscle-derived cells in advanced mouse lesions; KLF4 deletion reduced macrophage-like/mesenchymal transitions, lesion size, and multiple instability indices (Shankman 2015, PMID 25985364). After MI, mouse bone-marrow transfer experiments further showed transmissible monocyte trained immunity involving KMT5A/H4K20 methylation, CNBP, and SYK, with higher SYK/KMT5A expression in classical monocytes from people with advanced STEMI lesions (Dong 2024, PMID 38085922). Whether this axis causes recurrent human events or is druggable remains unresolved.

Open questions

  • Can a patient-level model combining plaque burden, low-attenuation plaque, physiology, inflammation, and thrombogenicity reach sufficient positive predictive value to justify prophylactic intervention? (Stone 2011, PMID 21247313; Koo 2024, PMID 38752951)
  • Does preventive PCI add durable benefit beyond 2 years, and on patient-level death/MI rather than a target-vessel composite, when LDL, blood pressure, smoking, and inflammation are treated to contemporary targets? (Park 2024, PMID 38604213)
  • Which patients have plaque erosion that can safely be treated without stenting? (Braunwald 2013, PMID 23473110)
  • Why do colchicine event reductions not map consistently onto local imaging changes? (Yu 2024, PMID 39166327; Fiolet 2025, PMID 40393691)
  • Can clonal immune responses in plaque be targeted without the infection penalty observed with broad cytokine blockade? (Chowdhury 2022, PMID 35430876; Ridker 2017, PMID 28845751)

References

  1. Falk E, et al. Coronary plaque disruption. Circulation. 1995;92:657-71. PMID 7634481
  2. Braunwald E, et al. Coronary plaque erosion: recognition and management. JACC Cardiovasc Imaging. 2013;6:288-9. PMID 23473110
  3. Stone GW, et al. A prospective natural-history study of coronary atherosclerosis. N Engl J Med. 2011;364:226-35. PMID 21247313
  4. Cholesterol Treatment Trialists’ (CTT) Collaboration, et al. Efficacy and safety of more intensive lowering of LDL cholesterol: a meta-analysis of data from 170,000 participants in 26 randomised trials. Lancet. 2010;376:1670-81. PMID 21067804
  5. Ridker PM, et al. Antiinflammatory Therapy with Canakinumab for Atherosclerotic Disease. N Engl J Med. 2017;377:1119-1131. PMID 28845751
  6. Ahn JM, et al. Preventive PCI or medical therapy alone for vulnerable atherosclerotic coronary plaque: Rationale and design of the randomized, controlled PREVENT trial. Am Heart J. 2023;264:83-96. PMID 37271356
  7. Park SJ, et al. Preventive percutaneous coronary intervention versus optimal medical therapy alone for the treatment of vulnerable atherosclerotic coronary plaques (PREVENT): a multicentre, open-label, randomised controlled trial. Lancet. 2024;403:1753-1765. PMID 38604213
  8. Kim MC, et al. Preventive percutaneous coronary intervention for non-flow-limiting vulnerable atherosclerotic coronary plaques in diabetes: the PREVENT trial. Eur Heart J. 2025;46:3181-3197. PMID 40439104
  9. Ueki Y, et al. Lipid-lowering Therapy and Coronary Plaque Regression. J Atheroscler Thromb. 2024;31:1479-1495. PMID 39111840
  10. Papadopoulou SL, et al. Natural history of coronary atherosclerosis by multislice computed tomography. JACC Cardiovasc Imaging. 2012;5:S28-37. PMID 22421228
  11. Conti CR, et al. Finding the vulnerable plaque. Clin Cardiol. 2010;33:320-1. PMID 20556800
  12. Koo BK, et al. Artificial Intelligence-Enabled Quantitative Coronary Plaque and Hemodynamic Analysis for Predicting Acute Coronary Syndrome. JACC Cardiovasc Imaging. 2024;17:1062-1076. PMID 38752951
  13. Williams MC, et al. Low-Attenuation Noncalcified Plaque on Coronary Computed Tomography Angiography Predicts Myocardial Infarction: Results From the Multicenter SCOT-HEART Trial (Scottish Computed Tomography of the HEART). Circulation. 2020;141:1452-1462. PMID 32174130
  14. Chowdhury RR, et al. Human Coronary Plaque T Cells Are Clonal and Cross-React to Virus and Self. Circ Res. 2022;130:1510-1530. PMID 35430876
  15. Ajala ON, et al. Targeting Inflammation to Reduce Residual Cardiovascular Risk. Curr Atheroscler Rep. 2020;22:66. PMID 32880743
  16. Tardif JC, et al. Efficacy and Safety of Low-Dose Colchicine after Myocardial Infarction. N Engl J Med. 2019;381:2497-2505. PMID 31733140
  17. Nidorf SM, et al. Colchicine in Patients with Chronic Coronary Disease. N Engl J Med. 2020;383:1838-1847. PMID 32865380
  18. Samuel M, et al. Lessons learned from large Cardiovascular Outcome Trials targeting inflammation in cardiovascular disease (CANTOS, CIRT, COLCOT and LoDoCo2). Future Cardiol. 2021;17:411-414. PMID 33687270
  19. Yu M, et al. Effect of Colchicine on Coronary Plaque Stability in Acute Coronary Syndrome as Assessed by Optical Coherence Tomography: The COLOCT Randomized Clinical Trial. Circulation. 2024;150:981-993. PMID 39166327
  20. Fiolet ATL, et al. Effect of low-dose colchicine on pericoronary inflammation and coronary plaque composition in chronic coronary disease: a subanalysis of the LoDoCo2 trial. Heart. 2025;111:1156-1163. PMID 40393691
  21. Boden WE, et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med. 2007;356:1503-16. PMID 17387127
  22. Maron DJ, et al. Initial Invasive or Conservative Strategy for Stable Coronary Disease. N Engl J Med. 2020;382:1395-1407. PMID 32227755
  23. Ross R, et al. Atherosclerosis--an inflammatory disease. N Engl J Med. 1999;340:115-26. PMID 9887164
  24. Libby P, et al. Inflammation in atherosclerosis. Nature. 2002;420:868-74. PMID 12490960
  25. Zuriaga MA, et al. Clonal hematopoiesis and atherosclerotic cardiovascular disease: A primer. Clin Investig Arterioscler. 2023;35:35-41. PMID 34879980
  26. Sabatine MS, et al. Evolocumab and Clinical Outcomes in Patients with Cardiovascular Disease. N Engl J Med. 2017;376:1713-1722. PMID 28304224
  27. Ridker PM, et al. Low-Dose Methotrexate for the Prevention of Atherosclerotic Events. N Engl J Med. 2019;380:752-762. PMID 30415610
  28. Williams KJ, et al. The response-to-retention hypothesis of atherogenesis reinforced. Curr Opin Lipidol. 1998;9:471-474. PMID 9812202
  29. Cai B, et al. MerTK receptor cleavage promotes plaque necrosis and defective resolution in atherosclerosis. J Clin Invest. 2017;127:564-568. PMID 28067670
  30. Shankman LS, et al. KLF4-dependent phenotypic modulation of smooth muscle cells has a key role in atherosclerotic plaque pathogenesis. Nat Med. 2015;21:628-637. PMID 25985364
  31. Dong Z, et al. Myocardial infarction drives trained immunity of monocytes, accelerating atherosclerosis. Eur Heart J. 2024;45:669-684. PMID 38085922