ABSTRACT
Background
Up to 50% of patients presenting with ST-elevation myocardial infarction (STEMI) have multivessel coronary artery disease (CAD). Randomized trials suggest that complete revascularization improves outcomes, but the mechanism and identification of patients who benefit remain unclear. This study aims to assess the association between blood and coronary imaging biomarkers and clinical events, to identify patient-, vessel-, and lesion-specific risk in STEMI patients with bystander disease.
Method
PICNIC is a multicenter, international, prospective, observational study enrolling 320 patients with STEMI and multivessel CAD undergoing primary PCI of the culprit vessel without complete revascularization. Participants will undergo blood sampling for inflammatory markers and coronary CT angiography (CTCA) to assess: (1) plaque burden and morphology, (2) artificial intelligence-enabled fractional flow reserve derived from CTCA (FFR CT ) analysis of plaque and hemodynamic features, and (3) fat attenuation index (FAI) to evaluate perivascular inflammation.
The primary analysis will evaluate the association between a composite 24-month clinical endpoint (including all-cause mortality, myocardial infarction, ischemia-driven revascularization as first layer and cardiac arrest, heart failure, stroke, and ventricular tachyarrhythmia (second layer)) and: (1) serum inflammatory markers, and (2) anatomical and physiological characteristics of non–infarct-related arteries (NIRA) assessed by CTCA, FFR CT , and FAI. Statistical and machine learning methods will be applied to determine which combinations of clinical, imaging, and biomarker data best predict patient-, vessel-, and lesion-specific risk.
Conclusion
PICNIC will characterize the anatomical, physiological, and inflammatory features of NIRA lesions in STEMI patients treated with culprit-only PCI in order to develop an AI-based risk prediction model. If such a model is successful it could be used to inform personalized revascularization strategies.
Background
Primary percutaneous coronary intervention (PPCI) is the gold standard of care for patients presenting with ST-elevation myocardial infarction (STEMI). , Up to 50% of STEMI patients undergoing PPCI have multivessel coronary artery disease (CAD), defined as significant disease in coronary territories not supplied by the culprit vessel. At least 9 randomized control trials (RCT) have shown that complete revascularization (CR) of bystander lesions, either at the time of PPCI or as a staged procedure, is safe and reduces the risk of subsequent MI and/or repeat revascularization ,,,,,,,, ( Table 1 ). Based upon this body of evidence, CR in this patient cohort now has a class 1a indication in both the European Society of Cardiology and ACC/AHA/SCAI guidelines. ,
Table 1
List of randomized-controlled trials comparing a strategy of complete revascularization (CR) vs culprit lesion (CL) only revascularization in patients presenting with ST-elevation myocardial infarction (MI) and multivessel CAD.
| Study | Intervention | Control | Definition of significant stenosis | Primary endpoint | Outcome |
|---|---|---|---|---|---|
| HELP AMI | CR at index procedure ( n = 52) | CL only PCI ( n = 17) | Not specified | Incidence of repeat revascularization | CR 17% vs CL 35%; P =.247 |
| Politi et al | NCL PCI either at index (CR; n = 65) or staged (SR; n = 65) | CL only PCI ( n = 84) | Visual estimation: DS > 70% | Composite of cardiac death, noncardiac death, in-hospital death, reinfarction, rehospitalization with ACS and new revascularization | CR 23% vs staged 13% vs CL 50% |
| PRAMI | NCL PCI during the index procedure (CR; n = 234) | CL only PCI ( n = 231) | Visual estimation: DS >50% | Composite of cardiac death, nonfatal MI, and refractory angina | CR 9% vs CL 23% (HR 0.35;95% CI 0.21-0.58) |
| CvLPRIT | NCL PCI at index procedure or index admission (CR; n = 138) | CL only PCI ( n = 139) | Visual estimation: DS >70% in 1 view >50% in 2 views | Composite of all-cause death, recurrent MI, heart failure, and ischemia-driven revascularization | CR 10% vs CL 21.2% (HR 0.45; 95% CI 0.24-0.84) |
| DANAMI-3-PRIMULTI | NCL PCI at index admission (CR; n = 314) | CL only PCI ( n = 313) | Angio-guided NCL PCI: DS > 90% FFR-guided NCL PCI: DS >50% and FFR ≤0.80 | Composite of all-cause death, reinfarction, and ischemia-driven revascularization | CR 13% vs CL 22% (HR: 0.56; 96% CI 0.38-0.83) |
| COMPARE-ACUTE | NCL PCI at index procedure or admission (CR; n = 295) | CL only PCI ( n = 590) | QCA FFR guided PCI: DS >50% and FFR≤0.80 | Composite of all-cause death, nonfatal MI, any revascularization, and cerebrovascular events | CR 7.8% vs CL 20.5% (HR: 0.35; 95% 0.22-0.55) |
| COMPLETE | NCL PCI at index admission or staged ( n = 2,106) | CL only PCI ( n = 2,025) | Visual estimation: angio-guided PCI: DS >70% FFR-guided PCI: DS>50-69% and FFR≤0.80 | Composite of cardiovascular death and MI; and composite of cardiovascular death, MI, and ischemia-driven revascularization | Outcome 1: CR7.8% vs CL 10.5% (HR: 0.74; 95% CI:0.60-0.91) Outcome 2: CR 8.9% vs CL 16.7% (HR: 0.51; 95% CI: 0.43-0.61) |
| FLOWER-MI | FFR guided NCL PCI at index procedure or admission ( n = 586) | CL only PCI ( n = 577) | Visual estimation: DS ≥50% and FFR ≤0.80 | Composite of death from any cause, nonfatal MI, unplanned hospitalization leading to revascularization | FFR-guided group 5.5% vs angiography-guided group 4.2% (HR: 1.32; 95% CI: 0.78-2.23) |
| FIRE (only 35.2% were STEMI patients) | CR either at index procedure or admission utilising FFR/iFR or QFR ( n = 720) | CL only PCI ( n = 725) | Visual estimation: DS 50-99% | Composite of death, MI, stroke, ischemia-driven revascularization | CR 15.7% vs CL 21% (HR: 0.73; 95% CI: 0.57-0.93) |
ACS, acute coronary syndrome; CI, confidence interval; DS, diameter stenosis; FFR, fractional flow reserve; HR, hazard ratio; NCL, nonculprit lesion; PCI, percutaneous coronary intervention.
Despite the wealth of data in support of CR, 4 areas of uncertainty remain when considering the application of CR in routine clinical practice. Firstly, the patient populations in the RCTs were highly selective. For example, of all the patients screened for enrolment, 19% were included in PRAMI, 35% in CvLPRIT, and 14% in DANAMI-3-PRIMULTI. Complex anatomy (mean SYNTAX score in COMPLETE was 14) including chronic total occlusions, heavily calcified lesions, left main and other complex bifurcations, patients in cardiogenic shock (a clinical scenario in which CR has been shown to be harmful ), and frail or comorbid patients were excluded. Furthermore, the results are not applicable to patients presenting 12 hours beyond symptom onset as there is currently no consensus regarding optimal timing and PCI strategy in such patients. Thus, the relevance of the outcomes of these highly selected groups to the wider STEMI population within routine clinical practice is uncertain.
Secondly, the definition of what constituted significant angiographic disease in noninfarct related arteries (NIRA) is inconsistent across the trials, in most cases being defined visually as either a 50% or 70% minimum diameter stenosis. Furthermore, in some trials, there was a requirement for positive invasive FFR as well as angiographic minimum stenosis of the bystander lesion(s), which resulted in lower inclusion rates for NIRA, so that 31% of patients randomized to CR did not qualify for NIRA revascularization in DANAMI-3-PRIMULTI, and 44% in COMPARE-ACUTE. We can assume that, in contrast to these trials, physiologically nonsignificant (ie, FFR negative) NIRA lesions may have been treated in the angiography-guided PRAMI and CvLPRIT studies, as well as in the COMPLETE trial, in which less than 1% actually underwent pressure wire assessment. Interestingly, the hazard ratios for recurrent MI and repeat revascularization in these trials were similar, despite differences in the criteria for revascularization. These data raise the question as to whether the benefit of CR in these trials is based upon anatomical or physiological criteria for defining bystander disease.
Thirdly, the underlying mechanism(s) that drives the benefit in CR across these trials remain(s) unresolved, if defined by which specific clinical event was reduced by the intervention. For example, neither PRAMI nor CvLPRIT reported dominant differences in specific components of their composite clinical endpoints, whereas ischemia-driven revascularization was the clear driver for benefit in both DANAMI-3-PRIMULTI and COMPARE-ACUTE, whereas lower incidence of new MI was the main driver of benefit in COMPLETE.
Finally, 2 important observations that stimulate the PICNIC hypothesis are that: (a) most patients enrolled into the above trials did not experience a clinical event during follow up, even in the culprit-only arms and (b) less than half of the clinical events were eliminated in the CR groups.
Limitations of the current trial data raise uncertainty about the concept that all bystander disease should be subject to CR in STEMI patients. We speculate that, rather than a “one size fits all” approach to CR, a tailored approach based upon personalized detection of patient-, vessel- and possibly lesion-level risk would be dominant. The PICNIC study, therefore, aims to determine the anatomical, physiological, and inflammatory features of lesions in the NIRA of patients presenting with STEMI who are treated with culprit-only PCI, and to identify potential associations with subsequent clinical events. In order to achieve this, we will employ novel imaging technology in the form of both fat attenuation index (Caristo) and FFR CT -mediated adverse plaque and hemodynamic characteristics (HeartFlow) to assess vessel- and lesion-specific features of risk.
Methods
The PICNIC study is an investigator led, multicenter, prospective, observational study being performed in 3 UK and 3 South Korean sites. The UK part of the study received full ethical approval in July 2024 (Queen Square Research Ethics Committee, London [REC Reference 24/LO/0394, IRAS project ID: 341232]) and is registered at ClinicalTrials.gov (NCT06506448). Enrolment started in February 2025 and 28 patients have been recruited to date. Recruitment is expected to conclude by December 2026. The follow-up end date will be December 2029 and primary endpoint results will be reported in 2030. A study flowchart is shown in Figure .
Study flow. STEMI. AIQCPHA, artificial intelligence coronary plaque & hemodynamic analysis; FAI, fat attenuation index; FFRCT, fractional flow reserve derived from CT; PPCI, primary percutaneous coronary intervention; ROC, receiver operating characteristic; STEMI, ST elevation myocardial infarction.
Study hypothesis
We hypothesize that the susceptibility of nonculprit disease to ischemic events after PPCI is variable between individuals, and that this may be vessel- and, perhaps, lesion-specific. Specifically, we postulate that this susceptibility may be related to multiple factors including anatomical and physiological vulnerability, and local vascular inflammatory status. To test this hypothesis, we will systematically examine the following parameters in each bystander coronary vessel in patients who present with STEMI and have undergone PPCI: (1) blood biomarkers of systemic inflammation; (2) plaque anatomy including lesion severity and markers of lesion vulnerability on CTCA; (3) assessment of individual coronary vessel inflammation using CT-derived fat attenuation index (FAI), and (4) vessel- & lesion-specific anatomical and physiological parameters derived by HeartFlow including FFR CT , Δ FFR CT , wall shear stress and axial plaque stress, as described below.
Using both traditional statistical assessment and an AI-driven model, incorporation of clinical, inflammatory, and imaging parameters, our aim is to develop a risk assessment tool for individuals, vessels and lesions that are at the greatest risk of being associated with subsequent clinical event(s). If we are able to construct a model that predicts vessel- and/or lesion-specific risk, it may be plausible to design a randomized trial of CR that only targets the highest risk bystander lesions, rather than the strategy of stenting all lesions of a certain diameter stenosis.
Primary analysis
The primary analysis is to define whether there is a correlation between a composite endpoint that includes all-cause mortality, myocardial infarction and unplanned revascularization at 24 months and (1) serum inflammatory markers, and (2) anatomical and physiological characteristics of coronary vessels and lesions in the NIRA as assessed by CTCA/FFR CT /FAI. This is with a view to the development of a risk model with a predefined discrimination/ calibration using the parameters we collect to predict clinical endpoints.
Secondary endpoints
The secondary endpoints are: (1) anatomical, physiological, and inflammatory features of lesions in the NIRA(s) at 12, 24 and 36 months; (2) association between anatomical, physiological, and inflammatory features and the risk of nonculprit lesions causing the more extensive composite of clinical events (death, myocardial infarction, unplanned revascularization, cardiac arrest, acute coronary syndrome, additional revascularization by CABG or PCI, rehospitalization for angina, heart failure, stroke, ventricular tachyarrhythmia) at 12, 24, and 36 months; (3) correlation between parameters of inflammation and CTCA based anatomical, hemodynamic, and plaque characteristics with individual components of the clinical event composite; (4) differences between patient groups with clinical events vs those without in terms of serum inflammatory markers, anatomical and physiological characteristics of coronary vessels/lesions in the NIRA; (5) the feasibility to build a multidimensional AI model that would best predict the risk of events in the NIRA vessels and NIRA lesions employing all the studied parameters (serum biomarkers, CTCA lesion severity, CTCA plaque characteristics, FFR CT derived plaque characteristics and FAI).
Patient population
Patients aged ≥18 years of age presenting with STEMI within 12 hours of symptom onset who have undergone PPCI and fulfilling the eligibility criteria ( Table 2 ) will be recruited. All patients will need to have undergone successful PCI to the culprit vessel, using stent(s), and have at least 1 coronary stenosis of ≥50% diameter stenosis by visual estimation in a NIRA with a minimum diameter of 2.5 mm. Key exclusion criteria include cardiogenic shock or ongoing hemodynamic instability, significant complication or poor outcome relating to the culprit lesion PCI, intention to perform complete revascularization. We will exclude patients who are considered likely to require surgical revascularization because (1) they are unlikely to be suitable for complete revascularization by PCI and (2) we will not be able to follow the natural history of bystander disease if they have undergone bypass graft surgery. Stent thrombosis cases are excluded because we seek to investigate de novo disease, and because such patients may have factors such as hyporesponsiveness to antiplatelet therapy that compounds variables.
Table 2
Inclusion and exclusion criteria.
|
Inclusion criteria
1. Age 18 to 85 years 2. Ability to provide written informed consent post primary PCI 3. Presentation with acute ST-elevation myocardial infarction within 12 hours of symptoms onset 4. Culprit lesion only primary PCI 5. Coronary stenosis of ≥50% diameter stenosis by visual estimation in the NIRA with a minimum diameter of 2.5 mm. |
|
Clinical exclusion criteria
1. Cardiogenic shock 2. Decompensated heart failure requiring intubation, inotropes, or IABP 3. Refractory ventricular arrhythmia 4. Previous coronary artery bypass surgery 5. Stent thrombosis and in-stent restenosis 6. An intention before inclusion into the study to revascularize a nonculprit lesion 7. Active malignancy or inflammatory disorders such as rheumatoid arthritis or inflammatory bowel disease 8. Severe valvular heart disease requiring surgery 9. Planned surgical revascularization 10. Active participation in another clinical trial 11. Life expectancy less than 12 months 12. Contraindication to CTCA such as presence of internal defibrillator, allergy to iodinated contrast, pregnancy, contraindication to beta blocker/sublingual nitrate administration, mechanical prosthetic heart valve, and renal impairment with creatinine >200 Anatomical exclusion criteria 1. NIRA stenosis ≥50% in the left main stem or ostia of both the left anterior descending and circumflex arteries 2. <TIMI III flow in the NIRA 3. Evidence of thrombus in the NIRA |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
