NSTE-ACS: NSTEMI and unstable angina — diagnosis, risk and invasive timing

Overview and definitions

Acute coronary syndromes (ACS) are a spectrum: a recent change in symptoms or signs, with or without ECG changes and with or without an acute troponin rise (ESC 2023).^[1] Patients with suspected ACS may end with a diagnosis of acute myocardial infarction (AMI) or UA.^[1] ACC/AHA 2025 describes three related conditions on a continuum of severity: UA, non-ST-elevation MI (NSTEMI) and STEMI.^[2]

ESC 2023 says suspected ACS is typically classified first by the ECG at presentation, for initial management, and then by the presence or absence of troponin elevation once results are available.^[1] The final diagnosis rests on symptoms, ECG and troponin, plus imaging or angiography to clarify the mechanism and the type of MI.^[1]

Working diagnosis: NSTE-ACS

ESC 2023 defines the working diagnosis of NSTE-ACS as acute chest pain, or chest pain-equivalent signs or symptoms, without persistent ST-segment elevation or ST-elevation equivalents.^[1] The ECG may show transient ST elevation, persistent or transient ST depression, or T-wave abnormalities (hyperacute, inverted, biphasic, flat or pseudo-normalised T waves). It may also be normal.^[1] NSTEMI or UA will not be the final diagnosis in every patient with this working diagnosis.^[1]

NSTEMI versus unstable angina

^{[1] [2] [29]}

ACS can be dynamic, and patients can progress rapidly from one condition to another (UA, NSTEMI, STEMI) during evaluation and treatment (ACC/AHA 2025).^[2]

ACS is not the same as MI

ESC 2023 stresses that ACS and MI are closely related but not the same.^[1] ESC 2023 defines AMI as cardiomyocyte necrosis in the clinical setting of acute myocardial ischaemia, including atherothrombotic MI (type 1) and other causes of ischaemia and necrosis (types 2–5).^[1] These numbered types are the classification of the Fourth Universal Definition of MI (2018), on which ESC 2023 bases the MI diagnosis; that framing is now dated.^[1][29] The current Fifth Universal Definition of MI (2026) replaces the numerical classification with three clinical types: primary, secondary and procedure-related MI.^[29] ESC 2023 describes myocardial injury in the dated Fourth UDMI terms: troponin release from mechanisms other than ischaemia that does not meet MI criteria, acute or chronic depending on whether serial troponins change.^[1][29] The Fifth UDMI defines acute myocardial injury as a rise and/or fall in troponin with at least one value above the sex-specific 99th percentile upper reference limit (URL), from ischaemic or non-ischaemic mechanisms.^[29] It classifies acute myocardial injury as MI when it is shown to be ischaemic and the MI criteria are met.^[29] Chronic myocardial injury is considered when two or more values exceed that limit in a stable clinical setting, and confirmed when a cardiac or non-cardiac condition associated with cardiac remodelling is identified (Fifth UDMI).^[29]

^{[2] [29]}

Both guidelines centre on type 1 MI: ESC 2023 focuses largely on patients who will receive a type 1 diagnosis, and ACC/AHA 2025 focuses on type 1 AMI.^[1][2] ACC/AHA 2025 states that type 2 MI, spontaneous coronary artery dissection (SCAD) and MINOCA are covered in separate documents, so the type 2 MI and MINOCA sections below use ESC 2023.^[2]

Epidemiology

^{[1] [2]}

In the United States, chest pain is the second most common presenting complaint to emergency departments, and most of these patients will not have ACS (ACC/AHA 2025).^[2] Risk of cardiovascular complications is highest acutely, including before hospital presentation and during the early hospital phase.^[2]

ESC 2023 finds no data supporting different management of ACS by sex, yet several studies report that women are treated differently.^[1] They are less likely than men to receive invasive coronary angiography, timely revascularisation, cardiac rehabilitation and secondary prevention drugs.^[1] Women and men receive equal benefit from invasive and non-invasive strategies and, in general, should be managed similarly.^[1]

Pathophysiology

ACS is typically caused by rupture or erosion of an unstable atherosclerotic plaque, with associated partial or complete thrombosis and/or microemboli, resulting in diminished myocardial blood flow (ACC/AHA 2025).^[2] Exposure of plaque contents can activate the coagulation cascade and thrombosis, and the thrombus may compromise flow, leading to ischaemia and eventual myonecrosis.^[2]

Why is there no ST elevation? In NSTEMI the artery may be partially occluded, leading to subendocardial ischaemia. In STEMI the vessel is typically completely occluded, leading to transmural ischaemia and infarction (ACC/AHA 2025).^[2] Less common causes of ischaemia include coronary spasm, embolism and dissection.^[2]

In ESC 2023, which uses the dated Fourth UDMI types, type 2 MI has a different mechanism: ischaemic injury from an oxygen supply–demand mismatch that is not related to acute coronary atherothrombosis.^[1] ESC 2023 says it may occur with atherosclerosis plus a supply–demand imbalance, with the imbalance alone, with vasospasm or coronary microvascular dysfunction, or with non-atherosclerotic coronary dissection.^[1] Under the current Fifth UDMI (2026), spontaneous coronary artery dissection, embolism and vasospasm are acute coronary pathologies of primary MI.^[29] Under the Fifth UDMI, secondary MI is considered when acute myocardial injury results from a supply–demand mismatch caused by another acute condition and produces symptoms or signs of ischaemia.^[29] Where imaging is feasible and appropriate, it is confirmed by obstructive coronary artery disease (≥70% stenosis, or ≥50% if flow-limiting) without an acute coronary pathology, or a new or presumed new regional wall motion abnormality or absent viable myocardium in an ischaemic pattern.^[29] When imaging is not feasible or appropriate, clinical judgment is required.^[29]

Clinical presentation and bedside assessment

Acute chest discomfort, described as pain, pressure, tightness, heaviness or burning, is the leading presenting symptom of ACS (ESC 2023).^[1] ESC 2023 classifies chest pain as cardiac, possibly cardiac or likely non-cardiac, and says the descriptor "atypical" should be avoided.^[1] Chest pain-equivalent symptoms include dyspnoea, epigastric pain, and pain in the left or right arm or neck/jaw.^[1]

Presentations range from symptom-free patients to ongoing chest discomfort, cardiac arrest, electrical or haemodynamic instability, or cardiogenic shock.^[1] Patients with diabetes may more often present with non-specific symptoms, which can delay diagnosis and treatment (ESC 2023).^[1] Misdiagnosis or delay sometimes comes from an incomplete history or difficulty eliciting symptoms.^[1] ESC 2023 lists prolonged chest pain (over 15 min) and/or recurrent pain within 1 h as red flag symptoms that should prompt the public to seek urgent help.^[1]

ESC 2023 recommends prompt assessment of vital signs at first medical contact (FMC), at the same time as the first ECG.^[1] Physical examination is recommended, both to exclude differential diagnoses and to find very high-risk and high-risk features. Diagnosis and initial short-term risk stratification should rest on the combination of history, symptoms, vital signs, other physical findings, ECG and hs-cTn (Class I, Level B).^[1] The focused examination checks all major pulses, blood pressure in both arms, heart and lung auscultation, and signs of heart failure or circulatory compromise.^[1]

Patients without ST elevation but with ongoing ischaemic symptoms should be triaged before hospital as for the STEMI pathway, because they also face immediate risks, including ventricular arrhythmias (ESC 2023).^[1]

The ECG

ESC 2023 recommends 12-lead ECG recording and interpretation as soon as possible at FMC, with a target of under 10 min (Class I, Level B); its text specifies interpretation by a qualified emergency medical technician or physician. It also recommends an additional 12-lead ECG with recurrent symptoms or diagnostic uncertainty (Class I, Level C).^[1] ACC/AHA 2025 sets the goal of an ECG obtained and interpreted by a trained clinician within 10 minutes of presentation.^[2]

In NSTE-ACS the ECG may be normal in more than one-third of patients, but characteristic abnormalities are frequent and raise the probability of ACS (ESC 2023).^[1] These include ST depression and T-wave changes, especially biphasic or deep negative T waves (Wellens' sign), which relate to severe proximal LAD stenosis.^[1]

^[2]

ACC/AHA 2025 calls transient ST elevation a high-risk finding.^[2] An initial nondiagnostic ECG does not rule out ACS: compare it with prior ECGs and repeat it during the ED course, guided by symptoms, especially recurrent chest pain.^[2] ST depression in V1–V3 could indicate an evolving posterior STEMI and should prompt a posterior-lead ECG if warranted.^[2]

High-sensitivity troponin and the rapid algorithms

ESC 2023 recommends measuring a marker of cardiomyocyte injury, preferably high-sensitivity cardiac troponin (hs-cTn), in all patients with suspected ACS, immediately after presentation, with results within 60 min of sampling (Class I, Level B).^[1] ESC 2023 states that, if the presentation is compatible with ischaemia, a rise and/or fall in troponin above the 99th percentile of healthy individuals points to MI under the fourth universal definition, a framing now dated.^[1] The current Fifth UDMI (2026) defines myocardial injury by the sex-specific 99th percentile URL.^[29] Under it, the triage thresholds of accelerated diagnostic pathways are often not based on the 99th percentile, are the same for both sexes, and stratify risk rather than define injury or confirm MI.^[29] The Fifth UDMI rates primary MI likely when acute myocardial injury arises spontaneously, with no alternative acute condition or cardiac procedure as the trigger, plus one or more of ischaemic symptoms, new or presumed new ischaemic ECG changes, or pathological Q waves.^[29] Imaging confirms it by showing an acute coronary pathology, or a new or presumed new regional wall motion abnormality or absent viable myocardium in an ischaemic pattern.^[29] ACC/AHA 2025 states that AMI shows a rising and/or falling troponin pattern with at least one value above the 99th percentile upper reference limit, believed to be caused by ischaemia.^[2]

In MI, troponin usually rises within 1 h of symptom onset with high-sensitivity assays and stays elevated for a variable period, usually several days (ESC 2023).^[1] hs-cTn T and hs-cTn I assays appear to give comparable diagnostic accuracy.^[1] ESC 2023 asks for "non-elevated" and "elevated", relative to the 99th percentile, in place of "normal" and "abnormal".^[1]

High-sensitivity assays are recommended over lower-sensitivity assays (ESC 2023).^[1] Most point-of-care tests are not high-sensitivity assays; their shorter turnaround is offset by lower sensitivity, diagnostic accuracy and NPV.^[1]

The ESC 0 h/1 h and 0 h/2 h algorithms

ESC 2023 recommends an algorithmic approach with serial hs-cTn, 0 h/1 h (best option) or 0 h/2 h (second-best option), to rule in and rule out NSTEMI (Class I, Level B).^[1] They apply to ED patients with suspected NSTEMI who have no indication for immediate invasive angiography.^[1] Rule-out thresholds were chosen for a sensitivity and NPV of at least 99%, and rule-in thresholds for a PPV of at least 70%.^[1] The cut-offs are assay specific and are listed in the ESC Supplementary Table S4, which this page does not hold.^[1]

The algorithms rest on two ideas. First, hs-cTn is a continuous variable, and the probability of MI rises with the value.^[1] Second, early absolute changes over 1 h or 2 h can stand in for changes over 3 h or 6 h.^[1]

^[1]

After rule-out, elective imaging may still be appropriate, and an alternative diagnosis should be identified (ESC 2023).^[1] In the observe zone, most patients with high clinical suspicion (for example a relevant troponin rise by 3 h) are suitable for invasive angiography. Most with low to intermediate likelihood are suitable for non-invasive imaging on the ward.^[1] In the appropriate clinical context, if an alternative cause explains the troponin, such as AF with a rapid ventricular rate, marked anaemia or a hypertensive emergency, angiography may not be required.^[1]

The algorithms should always be integrated with a detailed clinical assessment and a 12-lead ECG, and repeat sampling is mandatory with ongoing or recurrent chest pain (ESC 2023).^[1] Draw the 0 h and 1 h samples regardless of other clinical details and pending results.^[1] The ESC 0 h/3 h algorithm is an alternative for cases where the 0 h/1 h or 0 h/2 h algorithms are not available. Three recent large diagnostic studies suggested that it appears to balance efficacy and safety less well than more rapid protocols using lower rule-out concentrations, including the ESC 0 h/1 h algorithm.^[1]

What shifts troponin besides MI

ESC 2023 names four variables that change hs-cTn concentrations in suspected NSTE-ACS: age (up to 300%), renal dysfunction (up to 300%), time from chest pain onset (over 300%) and, to a lesser extent, sex (about 40%).^[1] Sex-specific values have not shown a clear clinical benefit, so uniform cut-offs remain the ESC standard of care until automated tools (risk calculators) incorporating age, eGFR, time from chest pain onset and sex are available.^[1]

ACC/AHA 2025: clinical decision pathways

^{[1] [2]}

ACC/AHA 2025 adds that changes within the normal range, below the 99th percentile, can signal ischaemia and may warrant further evaluation.^[2] Clinicians should know the analytical characteristics of their own assay: limit of quantification, 99th percentile upper reference limit and criteria for significant change.^[2] hs-cTn assays are preferred over conventional assays because their sensitivity and negative predictive values are greater (ACC/AHA 2025).^[2]

ESC 2023 does not recommend biomarkers other than troponin for diagnosis unless troponin is unavailable.^[1] Among other biomarkers, only CK-MB, myosin-binding protein C and copeptin may have clinical relevance alongside standard troponin, and their incremental value is limited in most situations.^[1]

Imaging in suspected NSTE-ACS

^[1]

Why not CCTA for everyone? In RAPID-CTCA, a default early CCTA approach in suspected NSTE-ACS did not improve 1-year outcomes and was associated with a modest increase in the length and cost of stay (ESC 2023).^[1] The utility of CCTA may be limited with tachycardia, established CAD, previous stents or extensive calcification.^[1] It can risk stratify selected low-risk NSTEMI patients; those with normal arteries, non-obstructive disease or distal obstructive disease may then not need invasive angiography.^[1]

ACC/AHA 2025 indicates urgent echocardiography, which may begin with point-of-care ultrasound by trained clinicians, for cardiogenic shock, haemodynamic instability or suspected mechanical complications.^[2]

Differential diagnosis: troponin is not specific for MI

Several cardiac and non-cardiac conditions can mimic ACS, and troponin rises in conditions other than type 1 MI (ESC 2023; dated Fourth UDMI term).^[1] The table groups the alternatives ESC 2023 names in its type 2 MI, myocardial injury, imaging and observe-zone sections.

^{[1] [29]}

Older patients need extra care: hs-cTn performs excellently, but its specificity is lower than in younger patients, and elevated values are more commonly associated with conditions other than ACS (ESC 2023).^[1] ESC 2023 notes that a diagnosis of myocardial injury can change if later tests show MI criteria are met.^[1] The current Fifth UDMI (2026) counts ischaemic as well as non-ischaemic troponin rises as acute myocardial injury, and classifies injury shown to be ischaemic as MI when the MI criteria are met.^[29]

Risk stratification

Timing follows the risk tier. ESC 2023 sorts NSTE-ACS by very high-risk and high-risk features, while ACC/AHA 2025 speaks of unstable patients and of low, intermediate and high ischaemic risk.^[1][2]

^[1]

ACC/AHA 2025 gives examples of unstable NSTE-ACS: refractory or recurrent angina despite optimal medical therapy, haemodynamic or electrical instability, acute pulmonary oedema or heart failure, or worsening mitral regurgitation.^[2] These patients were largely excluded from RCTs.^[2]

GRACE and TIMI scores

ESC 2023 says established risk scores such as GRACE should be considered for prognosis estimation (Class IIa, Level B); its text adds that GRACE offers the best discriminative performance among the clinical risk scores.^[1] ACC/AHA 2025 calls GRACE and TIMI well validated and possibly useful for guiding some therapeutic decisions. They are not diagnostic tools.^[2]

^[2]

Killip class depends on evidence of heart failure, such as pulmonary congestion on examination, chest X-ray or lung ultrasound, and it feeds into the GRACE score (ACC/AHA 2025).^[2] GRACE has outperformed subjective physician assessment for predicting death or MI in STEMI or intermediate-risk NSTE-ACS.^[2] There is insufficient evidence that routine use of risk scores reduces cardiovascular events.^[2]

ESC 2023 lists clinical markers of high risk: older age, tachycardia, hypotension, Killip class above I, anterior MI, previous MI, elevated initial serum creatinine, history of heart failure, peripheral arterial disease or anaemia.^[1] Creatinine and eGFR should be measured in all ACS patients because they affect prognosis and are key elements of GRACE.^[1]

Biomarkers for prognosis

Initial troponin adds prognostic information to clinical and ECG variables, and the higher the hs-cTn, the greater the risk of death (ESC 2023).^[1] hs-cTn T has slightly greater prognostic accuracy for mortality than hs-cTn I.^[1] Natriuretic peptides add prognostic information on death, acute heart failure and AF (ESC 2023), and may identify patients at increased risk of death, heart failure and recurrent MACE (ACC/AHA 2025).^[1][2]

Bleeding risk

Major bleeding is associated with increased mortality in ACS, so ischaemic and bleeding risks are weighed together (ESC 2023).^[1] Both guidelines use the Academic Research Consortium definition of high bleeding risk (HBR): one major or two minor ARC-HBR criteria, assessed at the time of PCI in the ACC/AHA 2025 wording.^[1][2]

^[2]

Invasive strategy and its timing

ESC 2023 defines three invasive strategies (Table 3): immediate (emergency angiography as soon as possible), early (angiography under 24 h from diagnosis of ACS) and selective (angiography guided by clinical assessment and/or non-invasive testing).^[1]

ESC 2023

^[1]

Patients without very high- or high-risk features are generally those with suspected NSTE-ACS and non-elevated troponin, or elevated troponin not meeting MI criteria; their strategy can be tailored to the degree of clinical suspicion (ESC 2023).^[1] A selective approach is also appropriate for NSTEMI or UA patients who are poorly suited to angiography.^[1]

What does routine invasive management buy? ESC 2023 concludes that it does not reduce all-cause mortality in the overall NSTE-ACS population but reduces composite ischaemic endpoints, particularly in high-risk patients.^[1] It can increase peri-procedural complications and bleeding.^[1] Most of the evidence comes from old RCTs that predate radial access, modern drug-eluting stents, complete functional revascularisation, better adjunctive drugs and contemporary biomarker assays.^[1]

On timing, meta-analyses found no superiority of early over routine invasive strategies for death or non-fatal MI, though early strategies were associated with less recurrent or refractory ischaemia and a shorter stay (ESC 2023).^[1] A collaborative meta-analysis using a modified individual-patient-data approach found no overall mortality difference but a survival benefit in high-risk patients, including GRACE over 140 and positive troponin; tests for interaction were inconclusive.^[1] The largest meta-analysis (17 RCTs, over 10 000 patients) found that, in all-comers, early angiography only significantly reduced recurrent ischaemia and length of stay.^[1] Studies testing GRACE over 140 to guide timing in the hs-cTn era are lacking.^[1]

ACC/AHA 2025 (US)

For unstable NSTE-ACS, an immediate invasive strategy (under 2 hours from hospital admission) with intent to revascularise is recommended.^[2] If patients with ACS and unstable features are at a non-PCI-capable hospital, they should be transferred immediately to a PCI-capable facility, aiming for immediate angiography (ACC/AHA 2025).^[2]

At intermediate or high ischaemic risk, an invasive approach with intent to revascularise is recommended during hospitalisation. At low ischaemic risk, a routine invasive or a selective invasive approach with further risk stratification is recommended.^[2] Some low-risk patients, particularly those with normal biomarkers in whom ACS is questioned, should be considered for a selective invasive approach because they derive less benefit from routine invasive management. They should still have stress testing or coronary CT angiography before discharge.^[2] Patients at prohibitively high angiographic risk, or whose anatomy or preferences preclude revascularisation, may be managed noninvasively.^[2] The strategy trials predate hs-cTn assays, routine radial access, newer drug-eluting stents and contemporary antiplatelet therapy.^[2]

Once angiography is planned, the benefit of early versus delayed timing is unclear (ACC/AHA 2025).^[2] In a meta-analysis of early versus delayed angiography, no mortality benefit was seen, but GRACE over 140, diabetes, age over 75 years and elevated biomarkers favoured earlier angiography, although formal interaction tests were not significant.^[2] In TIMACS and VERDICT there was no significant MACE difference overall, but a signal toward fewer MACE with earlier vs delayed angiography when GRACE exceeded 140.^[2] A continuing steep biomarker rise despite optimised medical therapy may also favour earlier angiography, though this has not been tested in RCTs.^[2] In intermediate- or low-risk patients, a delayed strategy within 48 to 72 hours does not appear to increase future MACE.^[2]

^[2]

The trials behind invasive management

• FRISC II (1999): 2457 patients with unstable coronary-artery disease, early invasive vs non-invasive strategy, with placebo-controlled dalteparin for 3 months.^[3]
• FRISC II result: at 6 months the composite of death or MI was 9.4% with the invasive vs 12.1% with the non-invasive strategy (risk ratio 0.78). The abstract also reports, without restating the timepoint, MI alone at 7.8% vs 10.1% (significant) and mortality at 1.9% vs 2.9% (non-significant). The greatest advantages were in high-risk patients.^[3]
• TACTICS-TIMI 18 (2001): 2220 patients with UA or NSTEMI and ST-T changes, elevated cardiac markers or known CAD, all given aspirin, heparin and tirofiban.^[4]
• TACTICS-TIMI 18 compared routine catheterisation at 4 to 48 hours with a conservative (selectively invasive) strategy. The composite of death, nonfatal MI or rehospitalisation for ACS at 6 months was 15.9% with the early invasive vs 19.4% with the conservative strategy (OR 0.78).^[4]
• RITA 3 (2002): 1810 patients with NSTE-ACS, interventional vs conservative strategy. The composite of death, MI or refractory angina at 4 months was 9.6% with the interventional vs 14.5% with the conservative strategy (RR 0.66), mainly from halving refractory angina; death or MI at 1 year was similar (7.6% vs 8.3%).^[5]
• ICTUS (2005): 1200 patients with NSTE-ACS, chest pain, troponin T of 0.03 µg/L or more, and either ischaemic ECG changes at admission or documented coronary disease. The composite of death, nonfatal MI or rehospitalisation for anginal symptoms within 1 year was 22.7% with the early invasive vs 21.2% with the selectively invasive strategy (RR 1.07, P=0.33).^[6][26]
• In ICTUS, MI was significantly more frequent with early invasive than with selectively invasive management (15.0% vs 10.0%), and rehospitalisation was less frequent (7.4% vs 10.9%); the abstract does not restate the timepoint for these components.^[6][26]
• TIMACS (2009): 3031 patients randomised to angiography within 24 h or at 36 h or later (median 14 vs 50 hours). The composite of death, MI or stroke at 6 months was 9.6% with early vs 11.3% with delayed intervention (HR 0.85, P=0.15).^[7]
• The prespecified secondary composite of death, MI or refractory ischaemia at 6 months was 9.5% with early vs 12.9% with delayed intervention (a 28% relative reduction, HR 0.72), and early intervention improved the primary outcome in the highest-risk third (HR 0.65).^[7]
• VERDICT (2018): 2147 patients randomised to angiography within 12 hours or within 48 to 72 hours (median 4.7 vs 61.6 hours).^[8]
• VERDICT: over a median 4.3 years, the composite of all-cause death, nonfatal recurrent MI, hospital admission for refractory ischaemia or for heart failure was 27.5% with very early vs 29.5% with standard-timing angiography (HR 0.92). With GRACE over 140, the HR was 0.81 (P for interaction 0.023).^[8]
• In a collaborative meta-analysis of RCTs cited by ACC/AHA 2025, a routine invasive approach reduced death or MI by 18% (OR 0.82) and MI by 25% (OR 0.75) compared with a selective approach. Benefit was more apparent in higher-risk patients with elevated biomarkers.^[2]

Initial medical therapy

^{[1] [2]}

ESC 2023 says relief of chest pain after nitroglycerine can mislead and is not recommended as a diagnostic manoeuvre.^[1] ACC/AHA 2025 times the nitrate–PDE5 interval by drug: 12 h for avanafil, 24 h for sildenafil or vardenafil, and 48 h for tadalafil.^[2] ESC 2023 recommends continuous ECG monitoring and defibrillator availability as soon as possible in suspected STEMI, in suspected ACS with other ECG changes or ongoing chest pain, and once MI is diagnosed (Class I, Level B).^[1]

Antiplatelet therapy

Doses

^{[1] [2]}

ESC 2023 starts aspirin with a loading dose as soon as possible; aspirin is recommended for all patients without contraindications (Class I, Level A).^[1] ACC/AHA 2025 gives aspirin, in patients without an absolute contraindication, as soon as possible on presentation, irrespective of the final management strategy. If the patient cannot take oral medication, rectal or intravenous (where available) administration are options.^[2] With ticagrelor, aspirin should always be 100 mg daily or less (ACC/AHA 2025).^[2] After cangrelor, ESC 2023 gives the thienopyridine loading dose immediately after stopping the infusion; prasugrel may also be given 30 min before the infusion stops.^[1]

Which P2Y12 inhibitor?

^{[1] [2]}

• CURE (2001): 12 562 NSTE-ACS patients presenting within 24 hours received clopidogrel (300 mg, then 75 mg daily) or placebo on top of aspirin for 3 to 12 months. The abstract gives no single timepoint for these results. The first primary composite of cardiovascular death, nonfatal MI or stroke was 9.3% with clopidogrel vs 11.4% with placebo (RR 0.80); major bleeding was 3.7% vs 2.7%.^[9]
• TRITON-TIMI 38 (2007): 13 608 moderate-to-high-risk ACS patients scheduled for PCI. Prasugrel (60 mg loading, then 10 mg daily maintenance) vs clopidogrel (300 mg loading, then 75 mg daily maintenance) for 6 to 15 months: the primary composite of cardiovascular death, nonfatal MI or nonfatal stroke was 9.9% vs 12.1% (HR 0.81).^[11]
• TRITON-TIMI 38, prasugrel vs clopidogrel: stent thrombosis 1.1% vs 2.4%; major bleeding 2.4% vs 1.8% (HR 1.32) and fatal bleeding 0.4% vs 0.1%. In NSTE-ACS, the drug was given only once anatomy was deemed suitable for PCI (ACC/AHA 2025).^[11][2]
• PLATO (2009): 18 624 ACS patients with or without ST elevation. Ticagrelor (180 mg loading, then 90 mg twice daily) vs clopidogrel (300–600 mg loading, then 75 mg daily): the primary composite of vascular death, MI or stroke at 12 months was 9.8% vs 11.7% (HR 0.84); death from any cause was also lower with ticagrelor (4.5% vs 5.9%).^[10]
• PLATO bleeding, ticagrelor vs clopidogrel: overall major bleeding did not differ significantly (11.6% vs 11.2%), but ticagrelor was associated with a higher rate of non-CABG major bleeding (4.5% vs 3.8%). In the 28% managed without a planned invasive strategy, a 15% reduction in the primary endpoint of vascular death, MI or stroke was observed with ticagrelor versus clopidogrel (ACC/AHA 2025).^[10][2]
• ISAR-REACT 5 (2019): open-label, 4018 ACS patients with planned invasive evaluation. The primary composite of death, MI or stroke at 1 year was 9.3% with ticagrelor vs 6.9% with prasugrel (HR 1.36); BARC major bleeding was 5.4% vs 4.8% (P=0.46).^[12]
• In ISAR-REACT 5, NSTE-ACS patients received prasugrel after the coronary anatomy was delineated, whereas ticagrelor was loaded as soon as possible after randomisation (ESC 2023). ESC lists the open-label design and limited data in medically managed or CABG patients as limitations.^[1]
• POPular AGE (2020): 1002 NSTE-ACS patients aged 70 or older; 95% of the potent-drug arm received ticagrelor. Over 12 months of follow-up, PLATO major or minor bleeding was 18% with clopidogrel vs 24% with ticagrelor (HR 0.71). The co-primary net clinical benefit composite (all-cause death, MI, stroke, PLATO major and minor bleeding) was non-inferior for clopidogrel vs ticagrelor (28% vs 32%).^[15]

Pretreatment: wait for the anatomy?

Pretreatment means giving an antiplatelet drug, usually a P2Y12 inhibitor, before angiography and so before the anatomy is known (ESC 2023).^[1] Large randomised trials supporting routine pretreatment are lacking, and caution may be particularly relevant at high bleeding risk, for example in patients on an oral anticoagulant.^[1]

^{[1] [2]}

• ACCOAST (2013): 4033 NSTE-ACS patients with positive troponin scheduled for angiography within 2 to 48 hours received a prasugrel 30 mg loading dose before angiography or placebo; at PCI, the pretreatment group received 30 mg more and the control group 60 mg.^[13]
• ACCOAST, pretreatment vs placebo: the primary composite through day 7 (cardiovascular death, MI, stroke, urgent revascularisation or GP IIb/IIIa bailout) did not differ significantly (HR 1.02), while TIMI major bleeding through day 7 increased with pretreatment (HR 1.90). The median time from loading to angiography was 4.4 h (ESC 2023).^[13][1]
• DUBIUS (2020): 1449 NSTE-ACS patients randomised to upstream ticagrelor or no pretreatment (downstream ticagrelor or prasugrel at PCI). The trial stopped for futility; its primary composite through day 30 (vascular death, nonfatal MI or stroke, and BARC 3–5 bleeding) did not differ significantly between the downstream and upstream groups.^[14]
• ISAR-REACT 5 also showed a ticagrelor strategy with routine pretreatment to be inferior to prasugrel with a deferred loading dose in NSTE-ACS (ESC 2023).^[1]

Intravenous antiplatelet drugs

ESC 2023 finds no strong evidence for routine GP IIb/IIIa inhibitors in ACS patients scheduled for angiography, but their bailout use should be considered for no-reflow or a thrombotic complication during PCI (Class IIa, Level C). Pretreatment with a GP IIb/IIIa inhibitor is not recommended (Class III, Level A).^[1] Another potential use is high-risk PCI in patients not pretreated with a P2Y12 inhibitor.^[1] In EARLY-ACS, 9492 NSTE-ACS patients assigned to an invasive strategy received early eptifibatide 12 hours or more before angiography, or placebo with provisional eptifibatide after angiography. The primary composite at 96 hours (death, MI, recurrent ischaemia needing urgent revascularisation, or thrombotic bailout) was 9.3% vs 10.0%, so early use was not superior. Within 120 hours, early eptifibatide had higher rates of bleeding and red-cell transfusion.^[16][27]

Cangrelor may be considered case by case in P2Y12-naïve ACS patients undergoing PCI, including those in whom oral drugs may not be feasible in the setting of emergent PCI (eg, cardiogenic shock or mechanical ventilation) (ESC 2023; Class IIb, Level A).^[1] Its benefit on major ischaemic endpoints was offset by more minor bleeding in a meta-analysis.^[1]

Parenteral anticoagulation

Both guidelines recommend parenteral anticoagulation for all ACS patients: ESC 2023 at the time of diagnosis (Class I, Level A), ACC/AHA 2025 irrespective of the initial treatment strategy.^[1][2] ACC/AHA 2025 starts it upstream, at diagnosis and before angiography, and continues it to support PCI.^[2]

ESC 2023: choose by the expected timing of angiography

^[1]

ESC 2023 says crossover between anticoagulants, especially UFH and LMWH, should in general be avoided, with the exception of adding UFH to fondaparinux at PCI.^[1] Discontinuation of parenteral anticoagulation should be considered immediately after an invasive procedure (Class IIa, Level C); ESC 2023 names exceptions such as a confirmed LV aneurysm with thrombus or AF requiring anticoagulation.^[1] ESC 2023 also asks that the OASIS-5 effect be read against contemporary practice: radial access, early catheterisation and infrequent GP IIb/IIIa use.^[1]

Doses

^{[1] [2]}

ACC/AHA 2025 notes that UFH has traditionally been preferred in NSTE-ACS, both for treatment and to support PCI, with ACT guiding PCI dosing.^[2] A meta-analysis of 12 171 patients across 5 trials found no significant difference between LMWH and UFH in death or MI (2.2% vs 2.3%) or major bleeding.^[2] Anticoagulation is indicated until revascularisation; premature discontinuation is associated with a transient rebound in thrombin activity and reactivation of ischaemic events, with the greatest reinfarction risk in the first 4 to 8 hours (ACC/AHA 2025).^[2]

For heparin-induced thrombocytopenia, ESC 2023 recommends bivalirudin as the alternative to UFH, and ACC/AHA 2025 accepts argatroban or bivalirudin.^[1][2] In ACUITY, bivalirudin was noninferior to UFH plus a GP IIb/IIIa inhibitor for MACE and superior for 30-day major bleeding, using a trial-specific bleeding definition (ACC/AHA 2025).^[2]

• OASIS-5 (2006): 20 078 ACS patients received fondaparinux 2.5 mg daily or enoxaparin 1 mg/kg twice daily for a mean of six days. The primary composite of death, MI or refractory ischaemia at 9 days was 5.8% with fondaparinux vs 5.7% with enoxaparin (noninferior).^[17]
• OASIS-5: major bleeding at 9 days was 2.2% with fondaparinux vs 4.1% with enoxaparin (HR 0.52), and fondaparinux was associated with fewer deaths at 30 days (295 vs 352) and 180 days (574 vs 638).^[17]

Revascularisation: PCI or CABG, culprit or complete

In very high-risk NSTE-ACS needing immediate revascularisation, ESC 2023 says PCI is usually preferred for timeliness, unless mechanical complications favour surgery.^[1] Otherwise the modality follows the number of diseased vessels and general revascularisation principles.^[1]

ESC 2023 recommends radial access as the standard approach unless there are overriding procedural considerations, and drug-eluting stents in preference to bare metal stents (both Class I, Level A). Intravascular imaging should be considered to guide PCI (Class IIa, Level A) and, preferably as optical coherence tomography, may be considered for ambiguous culprit lesions (Class IIb, Level C).^[1] Culprit ambiguity can occur in more than 30% of suspected NSTE-ACS, and over 10% may have multiple culprits.^[1] ACC/AHA 2025 recommends intracoronary imaging to guide PCI in ACS with complex lesions.^[2]

Multivessel disease

^{[1] [2] [2]}

In FIRE, 1445 elderly patients (median age 80) with multivessel disease and ACS (about 65% NSTE-ACS), multivessel PCI reduced MACE and all-cause and cardiovascular mortality; the guideline sentence does not name the comparator or follow-up (ACC/AHA 2025).^[2] In SMILE (584 patients with multivessel NSTEMI), multivessel PCI in a single procedure, compared with a multistaged approach, reduced 1-year MACE, driven by less revascularisation.^[2] For patients appropriate for CABG, surgery could be considered before discharge, and observational data suggest early CABG (within 3 days) may have outcomes similar to delayed CABG.^[2]

With cardiogenic shock and multivessel disease, multivessel PCI at the index procedure is associated with worse outcomes (ACC/AHA 2025).^[2]

Stopping P2Y12 inhibitors before CABG (ACC/AHA 2025 Table 8)

^[2]

Resume the P2Y12 inhibitor after surgery when bleeding risk is not excessive, typically at 24–72 hours.^[2]

DAPT duration, shortening and de-escalation

ESC 2023: after PCI, aspirin plus prasugrel or ticagrelor is generally recommended for 12 months, irrespective of stent type, unless there are contraindications; its formal row recommends a P2Y12 inhibitor in addition to aspirin for 12 months unless there is HBR (Class I, Level A).^[1] By default, DAPT with a potent P2Y12 inhibitor is recommended for at least 12 months after ACS, except when surgery is urgently needed, oral anticoagulation is indicated, or bleeding risk is otherwise too high.^[1] ACC/AHA 2025: DAPT with aspirin and an oral P2Y12 inhibitor is indicated for at least 12 months as the default in ACS patients who are not at high bleeding risk.^[2]

Both guidelines offer bleeding-driven alternatives, but ESC 2023 says these should not be the default in the wider ACS population. They suit a specific motivation, such as HBR or other concerns about 12-month potent DAPT.^[1][2] Much of the evidence comes from non-inferiority trials powered for bleeding and not for ischaemic outcomes.^[1]

^{[1] [2]}

Why not drop the P2Y12 inhibitor instead? ACC/AHA 2025 cites a trial in which stopping the P2Y12 inhibitor and continuing aspirin alone after 6 months resulted in excess thrombotic risk, albeit with less bleeding.^[2] Data with clopidogrel have been mixed, but concerns have been raised that clopidogrel monotherapy started 1 to 2 months after PCI may increase MACE in ACS compared with longer DAPT.^[2] ESC 2023 adds that, in a network meta-analysis, 3-month but not 6-month DAPT was associated with more MI or stent thrombosis in ACS.^[1]

• TWILIGHT (2019): in patients at high risk for bleeding or an ischaemic event who had undergone PCI, after 3 months of ticagrelor plus aspirin without major bleeding or ischaemic events, 7119 patients took ticagrelor with aspirin or placebo for 1 year. Between randomisation and 1 year, BARC 2, 3 or 5 bleeding was 4.0% with ticagrelor plus placebo vs 7.1% with ticagrelor plus aspirin (HR 0.56); the composite of death, nonfatal MI or nonfatal stroke was 3.9% in both groups.^[18]
• TWILIGHT excluded STEMI; ESC 2023 cites a subgroup analysis suggesting consistent findings in 4614 patients with NSTEMI/UA.^[1]
• TICO (2020): 3056 Korean ACS patients with drug-eluting stents. Compared with ticagrelor-based 12-month DAPT, ticagrelor monotherapy after 3 months of DAPT resulted in a modest but statistically significant reduction in the 1-year net adverse clinical event composite of major bleeding and cardiovascular events (3.9% vs 5.9%, HR 0.66).^[19]
• STOPDAPT-2 ACS (2022): 4169 Japanese ACS patients. For the 12-month primary composite of cardiovascular (cardiovascular death, MI, stroke, definite stent thrombosis) or bleeding (TIMI major or minor) events, clopidogrel monotherapy after 1 to 2 months of DAPT was not noninferior to 12 months of aspirin plus clopidogrel (3.2% vs 2.8%). The cardiovascular component was numerically higher (2.8% vs 1.9%).^[20]
• ESC 2023 says this suggests that systematic very short DAPT (under 3 months) followed by clopidogrel monotherapy is not a useful strategy in ACS.^[1]
• TROPICAL-ACS (2017): 2610 biomarker-positive ACS patients after PCI. Platelet-function-guided step-down from prasugrel to clopidogrel from day 14 was non-inferior to 12 months of prasugrel for the 1-year net clinical benefit composite of cardiovascular death, MI, stroke or BARC 2 or higher bleeding (7% vs 9%).^[21][1]
• TOPIC (2017): 645 event-free ACS patients switched from a potent P2Y12 inhibitor to clopidogrel at 1 month or continued. At 1 year the primary composite (cardiovascular death, urgent revascularisation, stroke and BARC 2 or higher bleeding) occurred in 13.4% with switching vs 26.3% with unchanged DAPT (HR 0.48); BARC 2 or higher bleeding was 4.0% vs 14.9%, with no significant ischaemic difference.^[22][1]

ACC/AHA 2025 summarises de-escalation as switching from ticagrelor or prasugrel to clopidogrel. Randomised trials have suggested that unguided de-escalation 1 month after PCI reduces bleeding without incurring ischaemic risk, compared with longer-term prasugrel- or ticagrelor-based DAPT.^[2] Guided de-escalation trials showed a lack of benefit, noninferiority, or reductions in minor bleeding versus conventional DAPT.^[2]

Gastric protection and the clopidogrel–PPI question

ESC 2023 recommends a PPI with DAPT in patients at high risk of gastrointestinal bleeding (Class I, Level A). Its Table 7 lists higher-than-average risk as prior ulcer or haemorrhage, anticoagulant therapy, or chronic NSAID or corticosteroid use.^[1] It also means two or more of: age 65 or older, dyspepsia, reflux disease, Helicobacter pylori infection or chronic alcohol use.^[1] ACC/AHA 2025 recommends a PPI for ACS patients at elevated bleeding risk on DAPT or an oral anticoagulant.^[2]

PPIs that inhibit CYP2C19, particularly omeprazole and esomeprazole, may reduce the response to clopidogrel, but there is no strong evidence of more ischaemic events or stent thrombosis (ESC 2023).^[1] No interaction has been observed between PPIs and aspirin, prasugrel or ticagrelor.^[1] ACC/AHA 2025 cites a placebo-controlled trial in patients on clopidogrel that showed no significant difference in ischaemic events between omeprazole and placebo, while PPI use markedly decreased the risk of gastrointestinal bleeding.^[2]

Patients who need oral anticoagulation

In 6–8% of patients undergoing PCI, long-term oral anticoagulation is indicated and should be continued during the procedure (ESC 2023).^[1] Continue it only if a compelling indication exists. The ESC 2023 default below is for AF with a CHA2DS2-VASc score of 1 or more in men or 2 or more in women. Its text adds that in AF without mechanical prosthetic valves or moderate–severe mitral stenosis, NOACs are preferred over VKAs because they reduce bleeding risk.^[1]

^{[1] [2]}

ESC 2023 NOAC doses for both triple and dual therapy are apixaban 5 mg twice daily, dabigatran 110 or 150 mg twice daily, edoxaban 60 mg once daily, and rivaroxaban 15 or 20 mg once daily.^[1] Dose reductions follow drug-specific criteria, including renal function, body weight, concomitant drugs and age.^[1] At PCI on a NOAC, add low-dose parenteral anticoagulation (eg, enoxaparin 0.5 mg/kg IV or UFH 60 IU/kg); on a VKA, give no UFH if the INR is over 2.5; avoid P2Y12 pretreatment (ESC 2023 Table 7). A UFH bolus during PCI is recommended if the patient is on a NOAC or the INR is under 2.5 on a VKA (Class I, Level C).^[1]

Medically managed (non-revascularised) NSTE-ACS

Patients with a final ACS diagnosis who do not undergo reperfusion should receive a P2Y12 inhibitor in addition to aspirin, maintained over 12 months unless there is HBR (ESC 2023).^[1] Aspirin plus ticagrelor for up to 12 months has shown benefit over aspirin plus clopidogrel in this group. Aspirin plus prasugrel can be justified in preference to aspirin plus clopidogrel if angiography has confirmed CAD.^[1] CURE, in patients primarily managed medically, showed consistent early and late reductions in ischaemic events with clopidogrel over aspirin alone (ACC/AHA 2025).^[2]

MINOCA (ESC 2023)

In ESC 2023, MINOCA (myocardial infarction with non-obstructive coronary arteries) is symptoms suggestive of ACS, troponin elevation, and non-obstructive coronary arteries at angiography, defined as stenosis under 50% in any major epicardial vessel.^[1] The current Fifth UDMI (2026) renames it myocardial injury with non-obstructive coronary arteries (no stenosis of 50% or more), a working diagnosis.^[29] If MI is subsequently confirmed, the Fifth UDMI classifies it as primary, secondary or procedure-related MI.^[29] ESC 2023 reports a MINOCA prevalence ranging from around 1% to 14% of ACS patients undergoing angiography.^[1] ESC 2023 calls MINOCA an umbrella term covering coronary and non-coronary causes, the latter both cardiac and extra-cardiac.^[1]

ESC 2023 calls MINOCA a working diagnosis, not a final one, when angiography has not established the cause.^[1] Failing to find the cause may lead to inadequate or inappropriate therapy.^[1] ESC 2023 recommends following a diagnostic algorithm to find the underlying cause in every patient with this working diagnosis (Class I, Level C).^[1]

^[1]

Takotsubo management is not informed by prospective RCTs and is largely supportive and empiric (ESC 2023).^[1]

Type 2 MI and myocardial injury (ESC 2023)

In ESC 2023, type 2 MI is common and carries a prognosis similar to type 1 MI.^[1] ESC 2023 says type 2 and type 1 MI require diagnostic distinction, best achieved with an algorithmic approach.^[1] Once the patient is stabilised and any precipitating illness treated, targeted echocardiography and/or coronary angiography (invasive or CCTA) can identify contributory cardiac conditions and guide long-term treatment.^[1] The current Fifth UDMI (2026) splits the ESC 2023 (Fourth UDMI) type 2 group: spontaneous coronary artery dissection, embolism and vasospasm become primary MI.^[29] Supply–demand ischaemia from another acute condition is secondary MI only when the Fifth UDMI criteria are met, and most such patients instead have acute myocardial injury.^[29]

Special populations

Older adults. ESC 2023 says older adults should, in general, undergo the same diagnostic and treatment strategies as younger patients, including invasive angiography and revascularisation (Class I, Level B).^[1] In the absence of robust trial evidence, decisions should be individualised. ESC 2023 lists ischaemic and bleeding risks, estimated life expectancy, comorbidities, the need for non-cardiac surgery, quality of life, frailty, cognitive and functional impairment, patient values and preferences, and the estimated risks and benefits of an invasive strategy. ESC 2023 recommends adapting the choice and dose of antithrombotic and secondary prevention drugs to renal function, co-medications, comorbidities, frailty, cognitive function and specific contraindications (Class I, Level B). For frail older patients with comorbidities, it recommends a holistic approach to individualise treatment (Class I, Level B).^[1] Routine frailty (eg, Rockwood) and comorbidity (eg, Charlson) assessment is recommended.^[1]

In After Eighty, patients aged 80 or older with NSTEMI or UA were randomised to invasive (229) or conservative (228) management. Over a median 1.53 years, the composite of MI, urgent revascularisation, stroke and death was 40.6% with invasive vs 61.4% with conservative management (HR 0.53).^[23] Death alone did not differ (HR 0.89), benefit diluted with increasing age, and bleeding did not differ.^[23] Frail patients with NSTE-ACS receive less pharmacotherapy and invasive assessment and are at higher risk of death (ESC 2023).^[1]

Chronic kidney disease. ESC 2023 recommends the same diagnostic and therapeutic strategies as for normal kidney function, although dose adjustment may be necessary (Class I, Level C).^[1] Observational data suggest a better prognosis with early revascularisation than medical therapy alone in moderate to severe CKD.^[1] Antithrombotic type and dose and contrast volume should be considered based on kidney function; low- or iso-osmolar contrast at the lowest possible volume is recommended (Class I, Level A). Hydration during and after angiography should be considered in patients at risk of contrast-induced nephropathy, especially with acute kidney injury and/or CKD with eGFR under 30 mL/min/1.73 m² (Class IIa, Level B).^[1]

Diabetes. Symptoms may be non-specific, and ESC 2023 recommends assessing glycaemic status at initial evaluation in all patients with ACS (Class I, Level B).^[1]

Women. ACC/AHA 2025 notes women are less likely to have obstructive disease but derive as much benefit from a routine invasive approach with similar risk predictors.^[2]

Cancer. ESC 2023 recommends an invasive strategy in cancer patients presenting with high-risk ACS and expected survival of 6 months or more (Class I, Level B); its text extends invasive management to unstable patients regardless of prognosis.^[1] A conservative non-invasive strategy should be considered with poor cancer prognosis (expected survival under 6 months) and/or very high bleeding risk (Class IIa, Level C).^[1] Active cancer diagnosed in the past 12 months counts as HBR, and clopidogrel is the preferred P2Y12 inhibitor.^[1]

Pregnancy. Diagnostic criteria are unchanged in pregnancy, and SCAD is the most common cause of AMI in pregnancy (ESC 2023).^[1] A multidisciplinary team plans care, and delivery should ideally be postponed for at least 2 weeks after ACS.^[1]

Anaemia and transfusion

^{[1] [2]}

• REALITY (2021): 668 patients with MI and haemoglobin 7–10 g/dL; restrictive (trigger 8 or less) vs liberal (trigger 10 or less). 30-day MACE (all-cause death, stroke, recurrent MI or emergency revascularisation) was 11.0% with the restrictive vs 14.0% with the liberal strategy (RR 0.79), meeting non-inferiority, though the CI included a possibly important harm.^[1][25]
• MINT (2023): 3504 patients with MI and haemoglobin under 10 g/dL; restrictive (7 or 8 g/dL) vs liberal (under 10 g/dL). The composite of MI or death at 30 days was 16.9% with the restrictive vs 14.5% with the liberal strategy (RR 1.15, P=0.07); harms of a restrictive strategy cannot be excluded.^[24]

Complications and pitfalls

Bleeding is associated with a poor prognosis in ACS and is a major driver of unplanned DAPT discontinuation (ESC 2023).^[1] ESC 2023 recommends radial access as the standard approach unless there are overriding procedural considerations (Class I, Level A).^[1] An individual-patient meta-analysis of 7 RCTs (48.6% NSTE-ACS) associated radial access with relative risk reductions of 24% in all-cause death and 51% in major bleeding compared with femoral access (ACC/AHA 2025).^[2]

Mechanical complications are rarer after NSTEMI than STEMI: 0.06% vs 0.27% of cases in an ESC-cited series of almost 9 million ACS patients, with in-hospital mortality of 18% vs 42.4%.^[1] Sudden hypotension, recurrent chest pain, a new murmur, pulmonary congestion or raised venous pressure should raise suspicion, and immediate echocardiography is indicated.^[1]

Hospital care, prognosis and discharge

ESC 2023 text recommends admitting high-risk ACS patients to a coronary or intensive cardiac care unit after reperfusion. Its formal rows recommend ECG monitoring for at least 24 h in high-risk patients, including very high-risk NSTE-ACS (Class I, Level C). After successful reperfusion and an uncomplicated course, they recommend keeping these patients in the CCU/ICCU for at least 24 h whenever possible (Class I, Level C).^[1] LVEF should be measured before discharge in all ACS patients, and routine echocardiography during hospitalisation is recommended to assess LV function, detect mechanical complications and exclude LV thrombus (Class I, Level C).^[1] Discharge of high-risk ACS patients within 48–72 h should be considered in selected patients if early rehabilitation and adequate follow-up are arranged (Class IIa, Level A).^[1]

Secondary prevention should be offered to every patient and start as early as possible after the index event (ESC 2023).^[1] ACC/AHA 2025 recommends high-intensity statin therapy for all ACS patients and referral to cardiac rehabilitation, with home-based options for patients unable or unwilling to attend in person.^[2] Patients remain at increased risk of recurrent events for months to years after ACS (ACC/AHA 2025).^[2]

ESC 2023 and ACC/AHA 2025: where they differ

^{[1] [2]}

Australia and New Zealand (NHFA/CSANZ 2025)

The 2025 Australian clinical guideline for diagnosing and managing acute coronary syndromes, from the National Heart Foundation of Australia and the Cardiac Society of Australia and New Zealand (NHFA/CSANZ), replaces the 2016 guideline.^[28] This section draws on the summary of its recommendations published in the Medical Journal of Australia in 2026; the full guideline was published in Heart, Lung & Circulation.^[28] Each GRADE recommendation in the summary tables carries a strength of recommendation and a certainty of evidence, which the tables below reproduce as printed.^[28] Consensus recommendations are informed by expert opinion when there is indirect supporting evidence and the GRADE approach is not applicable; the summary tables give them no certainty of evidence (shown as —).^[28]

NHFA/CSANZ 2025 adopted the term acute coronary occlusion myocardial infarction (ACOMI), which incorporates both STEMI and STEMI equivalents.^[28] The term acknowledges ECG patterns beyond traditional ST-elevation criteria that reflect acute coronary occlusion without ST elevation, such as posterior MI and De Winter T waves.^[28] Certain high-risk ECG patterns, such as Wellens T waves, diffuse ST-segment depression in multiple leads with ST elevation in aVR, and hyperacute T waves, are associated with potential progression to ACOMI.^[28] Recognition of these patterns should prompt urgent, continuous cardiac monitoring and consideration for coronary angiography.^[28]

ECG and troponin

^[28]

Clinical decision pathways and risk stratification

^[28]

In most hospitals, standard turnaround times for laboratory-based assay results render the 0/1-h strategy impractical, so the NHFA/CSANZ 2025 summary calls a 0/2-h strategy currently the most pragmatic option in most settings.^[28]

Invasive management and timing

^[28]

Given variability in trial populations and lesion complexity, the NHFA/CSANZ 2025 summary says the recommendation for intracoronary imaging over angiography alone should not be applied to every PCI procedure.^[28]

Initial and antithrombotic therapy

^[28]

In people with NSTEACS, the NHFA/CSANZ 2025 summary says P2Y12 inhibitors can be withheld until the coronary anatomy is known and if angiography can be performed within the recommended time frames.^[28] CHA2DS2VA is a scoring system that evaluates the risk of stroke in people with atrial fibrillation (NHFA/CSANZ 2025 Table 3 footnote).^[28]

Multivessel disease, cardiogenic shock, cardiac arrest and SCAD

^[28]

There are currently no dedicated trials comparing complete versus culprit-only PCI in people with NSTEACS, although observational data suggest long-term benefit with multivessel revascularisation (NHFA/CSANZ 2025 summary).^[28] Physiology-guided PCI of the non-IRA using FFR has not shown clear benefit over an angiography-guided approach in STEMI/ACOMI, but may be reasonable to use in people with NSTEACS and older adults.^[28] The NHFA/CSANZ 2025 summary says routine intra-aortic balloon pump and venoarterial extracorporeal membrane oxygenation are not recommended in ACS with cardiogenic shock because of increased risk of bleeding and vascular complications with no survival benefits, though they may be considered in select cases.^[28] In people with ACOMI or NSTEACS due to SCAD who are haemodynamically stable, the NHFA/CSANZ 2025 summary says routine revascularisation is not recommended because of its association with PCI-related complications and limited evidence of benefit.^[28] In cases of haemodynamic compromise or significant ongoing ischaemia, urgent revascularisation with PCI or CABG may be required.^[28]

Systems of care

^[28]

Marked inequities in both the management and outcomes of ACS persist in Australia, and the NHFA/CSANZ 2025 summary describes them in these groups:^[28]

• Women: longer symptom duration, under or delayed diagnosis and treatment, including lower rates of secondary prevention medication use.^[28]
• First Nations peoples: coronary events at twice the rate of non-Indigenous Australians, exacerbated by ongoing inequities in healthcare access and outcomes.^[28]
• Older adults: a higher rate of ACS presentations, often complicated by atypical symptoms and age-related comorbidities.^[28]
• People living in regional and remote areas: higher rates of missed diagnosis and poorer outcomes.^[28]

Where NHFA/CSANZ 2025 and ESC 2023 differ in NSTE-ACS

^{[28] [1]}

ACC/AHA 2025 recommends an immediate invasive strategy (under 2 hours from hospital admission) with intent to revascularise for unstable NSTE-ACS.^[2]

High-yield summary

• Working diagnosis (ESC 2023): acute chest pain or a chest pain-equivalent without persistent ST elevation or equivalents; troponin then separates NSTEMI from UA, although not every patient ends with either diagnosis.^[1]
• Algorithms: ESC 0 h/1 h best, 0 h/2 h second-best; rule-out NPV over 99%, rule-in PPV about 70–75%; observe zone gets a 3 h troponin.^[1]
• Very high risk (ESC 2023): haemodynamic instability or cardiogenic shock, recurrent or refractory chest pain despite medical treatment, in-hospital life-threatening arrhythmias, mechanical complications of MI, acute heart failure presumed secondary to ongoing myocardial ischaemia, recurrent dynamic ST-T changes (particularly intermittent ST elevation): immediate invasive strategy (Class I, Level C).^[1]
• High risk (ESC 2023): confirmed NSTEMI, dynamic ST-T changes, transient ST elevation, GRACE over 140: early (within 24 h) angiography should be considered (Class IIa, Level A).^[1]
• ACC/AHA 2025: unstable patients, immediate invasive strategy under 2 hours with intent to revascularise; intermediate or high risk, invasive approach with intent to revascularise during admission.^[2]
• P2Y12: prasugrel or ticagrelor over clopidogrel; ESC 2023 says prasugrel should be considered over ticagrelor for PCI (ISAR-REACT 5); no routine pretreatment before early angiography (ACCOAST).^[1][12]
• Anticoagulation (ESC 2023): UFH for early angiography; fondaparinux if angiography is delayed, plus a UFH bolus at PCI (OASIS-5).^[1][17]
• DAPT: 12 months by default; de-escalation not in the first 30 days (ESC 2023); ticagrelor monotherapy 1 month or more after PCI if ticagrelor DAPT was tolerated (ACC/AHA 2025).^[1][2]
• MINOCA: stenosis under 50%; working diagnosis; CMR can identify the cause in up to 87% (ESC 2023); the Fifth UDMI (2026) renames it myocardial injury with non-obstructive coronary arteries.^[1][29]
• Type 2 MI (ESC 2023; dated Fourth UDMI term): no specific drug therapy; treat the precipitant (ESC 2023); the Fifth UDMI (2026) reclassifies spontaneous coronary artery dissection, embolism and vasospasm as primary MI, and supply–demand MI as secondary MI when its criteria are met.^[1][29]

References29ShowHide

1. [1]Byrne RA, et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J, 2023.PMID 37622654
2. [2]Rao SV, et al. 2025 ACC/AHA/ACEP/NAEMSP/SCAI Guideline for the Management of Patients With Acute Coronary Syndromes: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. Circulation, 2025.PMID 40014670
3. [3]FRagmin and Fast Revascularisation during InStability in Coronary artery disease Investigators Invasive compared with non-invasive treatment in unstable coronary-artery disease: FRISC II prospective randomised multicentre study. FRagmin and Fast Revascularisation during InStability in Coronary artery disease Investigators. Lancet, 1999.PMID 10475181
4. [4]Cannon CP, et al. Comparison of early invasive and conservative strategies in patients with unstable coronary syndromes treated with the glycoprotein IIb/IIIa inhibitor tirofiban. N Engl J Med, 2001.PMID 11419424
5. [5]Fox KA, et al. Interventional versus conservative treatment for patients with unstable angina or non-ST-elevation myocardial infarction: the British Heart Foundation RITA 3 randomised trial. Randomized Intervention Trial of unstable Angina. Lancet, 2002.PMID 12241831
6. [6]de Winter RJ, et al. Early invasive versus selectively invasive management for acute coronary syndromes. N Engl J Med, 2005.PMID 16162880
7. [7]Mehta SR, et al. Early versus delayed invasive intervention in acute coronary syndromes. N Engl J Med, 2009.PMID 19458363
8. [8]Kofoed KF, et al. Early Versus Standard Care Invasive Examination and Treatment of Patients With Non-ST-Segment Elevation Acute Coronary Syndrome. Circulation, 2018.PMID 30565996
9. [9]Yusuf S, et al. Effects of clopidogrel in addition to aspirin in patients with acute coronary syndromes without ST-segment elevation. N Engl J Med, 2001.PMID 11519503
10. [10]Wallentin L, et al. Ticagrelor versus clopidogrel in patients with acute coronary syndromes. N Engl J Med, 2009.PMID 19717846
11. [11]Wiviott SD, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med, 2007.PMID 17982182
12. [12]Schüpke S, et al. Ticagrelor or Prasugrel in Patients with Acute Coronary Syndromes. N Engl J Med, 2019.PMID 31475799
13. [13]Montalescot G, et al. Pretreatment with prasugrel in non-ST-segment elevation acute coronary syndromes. N Engl J Med, 2013.PMID 23991622
14. [14]Tarantini G, et al. Timing of Oral P2Y(12) Inhibitor Administration in Patients With Non-ST-Segment Elevation Acute Coronary Syndrome. J Am Coll Cardiol, 2020.PMID 32882390
15. [15]Gimbel M, et al. Clopidogrel versus ticagrelor or prasugrel in patients aged 70 years or older with non-ST-elevation acute coronary syndrome (POPular AGE): the randomised, open-label, non-inferiority trial. Lancet, 2020.PMID 32334703
16. [16]Giugliano RP, et al. Early versus delayed, provisional eptifibatide in acute coronary syndromes. N Engl J Med, 2009.PMID 19332455
17. [17]Fifth Organization to Assess Strategies in Acute Ischemic Syndromes Investigators, et al. Comparison of fondaparinux and enoxaparin in acute coronary syndromes. N Engl J Med, 2006.PMID 16537663
18. [18]Mehran R, et al. Ticagrelor with or without Aspirin in High-Risk Patients after PCI. N Engl J Med, 2019.PMID 31556978
19. [19]Kim BK, et al. Effect of Ticagrelor Monotherapy vs Ticagrelor With Aspirin on Major Bleeding and Cardiovascular Events in Patients With Acute Coronary Syndrome: The TICO Randomized Clinical Trial. JAMA, 2020.PMID 32543684
20. [20]Watanabe H, et al. Comparison of Clopidogrel Monotherapy After 1 to 2 Months of Dual Antiplatelet Therapy With 12 Months of Dual Antiplatelet Therapy in Patients With Acute Coronary Syndrome: The STOPDAPT-2 ACS Randomized Clinical Trial. JAMA Cardiol, 2022.PMID 35234821
21. [21]Sibbing D, et al. Guided de-escalation of antiplatelet treatment in patients with acute coronary syndrome undergoing percutaneous coronary intervention (TROPICAL-ACS): a randomised, open-label, multicentre trial. Lancet, 2017.PMID 28855078
22. [22]Cuisset T, et al. Benefit of switching dual antiplatelet therapy after acute coronary syndrome: the TOPIC (timing of platelet inhibition after acute coronary syndrome) randomized study. Eur Heart J, 2017.PMID 28510646
23. [23]Tegn N, et al. Invasive versus conservative strategy in patients aged 80 years or older with non-ST-elevation myocardial infarction or unstable angina pectoris (After Eighty study): an open-label randomised controlled trial. Lancet, 2016.PMID 26794722
24. [24]Carson JL, et al. Restrictive or Liberal Transfusion Strategy in Myocardial Infarction and Anemia. N Engl J Med, 2023.PMID 37952133
25. [25]Ducrocq G, et al. Effect of a Restrictive vs Liberal Blood Transfusion Strategy on Major Cardiovascular Events Among Patients With Acute Myocardial Infarction and Anemia: The REALITY Randomized Clinical Trial. JAMA, 2021.PMID 33560322
26. [26]Damman P, et al. 5-year clinical outcomes in the ICTUS (Invasive versus Conservative Treatment in Unstable coronary Syndromes) trial a randomized comparison of an early invasive versus selective invasive management in patients with non-ST-segment elevation acute coronary syndrome. J Am Coll Cardiol, 2010.PMID 20045278
27. [27]Giugliano RP, et al. The early glycoprotein IIb/IIIa inhibition in non-ST-segment elevation acute coronary syndrome (EARLY ACS) trial: a randomized placebo-controlled trial evaluating the clinical benefits of early front-loaded eptifibatide in the treatment of patients with non-ST-segment elevation acute coronary syndrome--study design and rationale. Am Heart J, 2005.PMID 15976780
28. [28]Brieger DB, et al. National Heart Foundation of Australia and Cardiac Society of Australia and New Zealand: Australian Clinical Guideline for Diagnosing and Managing Acute Coronary Syndromes 2025. Med J Aust, 2026.PMID 41693087
29. [29]Mills NL, Newby LK, Zaman S, et al. Fifth Universal Definition of Myocardial Infarction (2026). Glob Heart, 2026.PMID 42666939