STEMI: reperfusion strategy, timing and transfer decisions

Overview and definition

STEMI begins as a working diagnosis made from the ECG. ESC 2023 defines it as acute chest pain (or chest pain-equivalent signs or symptoms) with persistent ST-segment elevation, or ST-elevation equivalents, on the ECG.^[1] Most of these patients will sustain myocardial necrosis and a troponin rise, but MI will not be the final diagnosis in all of them.^[1] The priority is reperfusion therapy as soon as possible.^[1]

Why is the ECG so different from NSTEMI? ACC/AHA 2025 explains that ACS is usually caused by rupture or erosion of an unstable plaque with partial or complete thrombosis. In NSTEMI the artery may be partially occluded, causing subendocardial ischaemia. In STEMI the vessel is typically completely occluded, causing transmural ischaemia and infarction.^[2]

ACC/AHA 2025 states that, under the Universal Definition it cites (the Fourth UDMI, 2018, whose numbered types are now dated), such MI events would be classified as type 1 MI.^[2][37] ACC/AHA 2025 Table 4 describes type 1 MI as caused by acute coronary atherothrombosis, usually precipitated by plaque disruption, and type 2 MI as caused by a supply–demand imbalance unrelated to acute atherothrombosis.^[2] The current Fifth UDMI (2026) has replaced the numbered classification with three clinical types: primary, secondary and procedure-related MI.^[37] Primary MI is a spontaneous presentation due to a primary acute coronary pathology, secondary MI results from an oxygen supply–demand imbalance due to another acute condition, and procedure-related MI is a complication of a percutaneous or surgical cardiac procedure.^[37] Primary MI includes all acute coronary pathologies, not only atherothrombosis, and can present with or without ST-segment elevation.^[37] ESC 2023 notes that the incidence of STEMI is decreasing, while NSTEMI is increasing. It adds that women and men receive equal benefit from invasive and non-invasive strategies and should, in general, be managed similarly.^[1]

The vocabulary of timing (ESC 2023 Table 3)

^[1]

Pathophysiology: why minutes matter

The chain of events is the reason every delay counts. Plaque rupture exposes plaque contents to the circulation, which may activate the coagulation cascade and cause thrombosis. The thrombus may compromise flow to the myocardium, leading to ischaemia and eventual myonecrosis.^[2]

^[2]

Rapid reperfusion of the infarct-related artery is associated with improved myocardial salvage and improved survival.^[2] ACC/AHA 2025 quantifies the cost of waiting: in primary PCI, each 30 minutes of delay is associated with a 7.5% increase in the relative risk of 1-year mortality.^[2]

Opening the artery is not the end of the injury. ESC 2023 describes the damage of acute MI as the result of ischaemia and subsequent reperfusion (ischaemia/reperfusion injury).^[1] In patient-level pooled analyses, infarct size and microvascular obstruction (MVO) independently predict long-term mortality and HF in STEMI survivors, and strategies to reduce MVO remain an unmet need.^[1] ACC/AHA 2025 adds that suboptimal myocardial perfusion from distal embolisation or MVO is frequent despite a patent artery. It is associated with larger infarcts, LV impairment and excess mortality.^[2]

The early hours are also electrically dangerous. ESC 2023: a suspected STEMI carries a higher risk of immediate, life-threatening complications such as ventricular fibrillation (VF).^[1] ESC 2023 recommends continuous ECG monitoring and defibrillator availability as soon as possible in all patients with suspected STEMI (Class I, Level B), to detect arrhythmias and allow prompt defibrillation if indicated.^[1]

Right-ventricular infarction. RV infarction most often occurs with inferior MI and complicates about 25% (range 20%–60%) of inferior infarcts.^[30] When the RV fails, LV filling depends entirely on preload. A significant fall in preload is likely to cause hypotension, which nitroglycerin and morphine may worsen.^[30] Treatment, alongside standard STEMI care, relies largely on enhancing preload with intravenous fluid and judicious use of vasodilators.^[30] Inferior STEMI with RV infarction carries a markedly worse prognosis than isolated inferior STEMI.^[30]

ECG diagnosis

Get the ECG early and read it fast. 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.^[1] ACC/AHA 2025 sets the same 10-minute goal from presentation.^[2]

ST-elevation thresholds

ST elevation is measured at the J-point and must be present in at least two contiguous leads.^[1][2]

^{[1] [2]}

Extra leads and repeat ECGs

• Inferior STEMI: ESC 2023 recommends additional leads (V3R, V4R and V7–V9) in inferior STEMI, or if total vessel occlusion is suspected and standard leads are inconclusive (Class I, Level B).^[1] ACC/AHA 2025 also asks for right-sided leads to look for RV involvement.^[2]
• Posterior STEMI: that ESC 2023 Class I recommendation includes V7–V9; its text adds that they can be recorded to look for posterior STEMI, particularly with ongoing symptoms and an inconclusive standard 12-lead ECG.^[1] ACC/AHA 2025 says V7–V9 should be obtained when left circumflex occlusion is suspected, particularly with isolated ST depression of 0.5 mm or more in V1–V3.^[2]
• Serial ECGs: a second or third ECG during EMS transport may find up to 15% more STEMIs.^[2] In the NCDR ACTION registry, 11% of patients ultimately diagnosed with STEMI had a non-diagnostic first ECG. Of these, 72.4% had a diagnostic ECG within 90 minutes of their initial ECG.^[2]

Occlusion without classic ST elevation

ESC 2023 accepts that occlusion can be hard to see. Some cases may warrant triage for immediate reperfusion despite the absence of ST elevation.^[1] When the following findings are present, ESC 2023 indicates prompt triage for immediate reperfusion:^[1]

• Posterior occlusion: ST depression in V1–V3 (especially with a positive terminal T wave) and/or ST elevation in V7–V9. These are highly suggestive of posterior occlusion, often of the left circumflex.^[1]
• RV ischaemia: ST elevation in V3R and V4R is highly suggestive of ongoing RV ischaemia.^[1]
• Left main or multivessel ischaemia: ST depression of 1 mm or more in six or more leads, with ST elevation in aVR and/or V1. This is particularly significant with haemodynamic compromise.^[1]

ACC/AHA 2025 makes the posterior point too: ST depression in V1–V3 may be an evolving posterior STEMI and needs a posterior-lead ECG if warranted.^[2]

Two further patterns come from publications outside the guidelines:

• Persistent precordial "hyperacute" T-wave pattern (Verouden 2009): J-point ST depression of at least 1 mm in the precordial leads, with upsloping ST segments running into tall, symmetrical T waves. The authors concluded that, in patients presenting with chest pain, it signifies proximal LAD occlusion.^[29] It was found in 35 of 1890 patients (2%) having primary PCI of the LAD. The authors advise triage for immediate reperfusion.^[29]
• RV infarction pattern (Moye 2005): inferior STEMI with hypotension and jugular venous distension. ST elevation is greatest in lead III (compared with II and aVF), with ST elevation in V1 and/or the right chest leads.^[30]

Bundle branch block and paced rhythm: ESC and ACC/AHA differ

^{[1] [2]}

The modified Sgarbossa rule (Smith 2012). The study used admission ECGs of patients with an acutely occluded coronary artery and LBBB. ECGs of emergency department patients with chest pain or dyspnoea and LBBB, but without coronary occlusion, were the controls. It replaced the third Sgarbossa component (discordant ST elevation of 5 mm or more with a negative QRS) with a proportional ST/S ratio.^[28] The revised rule is unweighted: any 1 of 3 criteria is positive.^[28] With ST elevation at an ST/S ratio of −0.25 or less replacing the 5 mm criterion, sensitivity was 91%, against 52% (weighted) and 67% (unweighted) for the original rule.^[28] Specificity was 90%, lower than the weighted rule.^[28] The derivation was small: 33 occlusion ECGs and 129 control ECGs.^[28]

Mimics and traps

• Conditions that can produce ST changes: acute pericarditis, LV hypertrophy, LBBB, Brugada syndrome, RV pacing, Takotsubo syndrome and early repolarisation (ACC/AHA 2025).^[2]
• Nitrate response (ESC 2023): relief after nitroglycerine can be misleading and is not recommended as a diagnostic manoeuvre. If pain settles with an ongoing STEMI pattern, repeat the 12-lead ECG. Complete normalisation of ST elevation with symptom relief suggests coronary spasm, with or without MI.^[1]
• Troponin and other tests: with ECG evidence of STEMI, reperfusion should not wait for biomarker results.^[2] Further diagnostic testing should not delay catheterisation or fibrinolysis unless it would immediately change management (ACC/AHA 2025).^[2]
• Diabetes (ESC 2023): ACS patients with diabetes may more commonly present with non-specific symptoms, which can delay diagnosis and treatment.^[1]

Choosing the reperfusion strategy

^{[1] [2]}

ESC 2023

ESC 2023 recommends that patients with suspected STEMI are immediately triaged for an emergency reperfusion strategy (Class I, Level A).^[1] Reperfusion therapy is recommended in all patients with a working diagnosis of STEMI and ischaemic symptoms of 12 h or less (Class I, Level A).^[1] A primary PCI strategy is recommended over fibrinolysis if the anticipated time from diagnosis to PCI is under 120 min (Class I, Level A); its text says primary PCI is preferred if it can be performed within 120 min of the ECG-based diagnosis.^[1] If the delay to treatment is similar, RCTs show primary PCI is superior to fibrinolysis for mortality, non-fatal reinfarction and stroke.^[1] If timely primary PCI (under 120 min) cannot be performed in a working diagnosis of STEMI, fibrinolytic therapy is recommended within 12 h of symptom onset in patients without contraindications (Class I, Level A), as part of a pharmaco-invasive strategy.^[1]

Why 120 minutes? Contemporary data on the delay at which PCI loses its advantage are lacking.^[1] For simplicity, ESC 2023 chose an absolute 120 min from STEMI diagnosis to wire crossing of the IRA, rather than a relative PCI-related delay.^[1] Allowing 10 min from diagnosis to a lytic bolus, this equals a relative PCI-related delay of 110–120 min.^[1]

At a non-PCI centre, transfer immediately to a PCI-capable centre.^[1] If primary PCI is not feasible within 120 min, give fibrinolysis immediately and transfer without waiting for signs of reperfusion.^[1] After 12 h from symptom onset, ESC 2023 prefers a primary PCI strategy over fibrinolysis in all cases; whether to intervene depends on timing and symptoms (see Late presenters).^[1]

The system matters as much as the decision:

• ESC 2023 recommends that EMS transfer patients with suspected STEMI to a PCI-capable centre, bypassing non-PCI centres (Class I, Level C), and that PCI-capable centres deliver a 24/7 service (Class I, Level B).^[1]
• When the EMS makes a prehospital STEMI diagnosis and the patient is triaged for emergency invasive management, the patient should bypass the ED and go straight to the catheterisation laboratory.^[1] It is recommended that patients transferred for primary PCI bypass the ED and CCU/ICU and go directly to the catheterisation laboratory (Class I, Level B).^[1] Immediate prehospital activation of the catheterisation team reduces treatment delays and mortality.^[1]
• Door-in to door-out time runs from arrival at the non-PCI hospital to departure in an ambulance to the PCI centre. ESC 2023 recommends 30 min or less.^[1]
• System delay predicts mortality after primary PCI.^[1] It is amenable to organisational change, whereas patient delay is multifactorial.^[1] People who recognise ischaemic symptoms should contact the EMS rather than self-present to an ED or primary care clinician.^[1]
• Where transfer times make it impossible to routinely achieve the maximal allowable delays, networks should develop rapid fibrinolysis protocols at the place of diagnosis, aiming for treatment within 10 min of FMC, then immediate transfer to a 24/7 PCI centre.^[1]

ACC/AHA 2025 (US)

In patients with STEMI and ischaemic symptoms, primary PCI is the recommended reperfusion therapy when FMC-to-first-device is 90 minutes or less, or 120 minutes or less in transfer patients.^[2] The goal is first device activation within 90 minutes of FMC for walk-in and EMS patients, and 120 minutes for transfers from non-PCI-capable facilities.^[2]

At a non-PCI-capable hospital, transfer is recommended if device activation can reasonably be predicted within 2 hours of FMC.^[2] If not, and symptoms began less than 12 hours earlier, fibrinolytic therapy is recommended.^[2] Transfer delays are frequently underestimated, so the delay should be calculated case by case before transfer.^[2] The benefit of primary PCI over fibrinolysis diminishes with increasing PCI-related delay; beyond 120 minutes it is no longer as clear compared with more timely fibrinolysis.^[2]

Prehospital STEMI should go preferentially to a PCI-capable hospital with a goal of 90 minutes or less to first device.^[2] Where that is not feasible, direct transport may be considered if the anticipated FMC-to-device time is 120 minutes or less.^[2] Within a regional system, care processes shown to reduce reperfusion times include prehospital catheterisation laboratory activation, a single-call transfer protocol, and ED bypass.^[2]

With a similar treatment delay, primary PCI gives higher IRA patency and TIMI 3 flow than fibrinolytic therapy. It also gives lower rates of recurrent ischaemia, reinfarction, emergency repeat revascularisation, intracranial haemorrhage and death.^[2]

Time targets side by side

^{[1] [2]}

Trials behind the transfer-or-lyse decision

• DANAMI-2 (2003): 1572 patients with acute MI randomised to angioplasty or accelerated alteplase.^[3] The primary end point was death, reinfarction or disabling stroke at 30 days.^[3] At referral hospitals it occurred in 8.5% after transfer for angioplasty vs 14.2% with on-site fibrinolysis.^[3] Across all patients, the benefit came mainly from less reinfarction (1.6% vs 6.3%). Across all patients, death (6.6% vs 7.8%) and stroke did not differ significantly.^[3] Ninety-six percent of transfers took two hours or less, and the authors' conclusion is conditional on that.^[3]
• PRAGUE-2 (2003): 850 patients with acute STEMI presenting within 12 h to the nearest community hospital without a catheter laboratory, randomised to thrombolysis there or immediate transport for primary PCI.^[4] Thirty-day mortality was 10.0% with on-site thrombolysis vs 6.8% with transport for PCI (P=0.12, intention to treat).^[4] Beyond 3 h from onset, mortality was 15.3% vs 6%. Within 3 h it was 7.4% vs 7.3%.^[4]
• STREAM (2013): 1892 patients with STEMI presenting within 3 h of symptom onset who could not have primary PCI within 1 hour, randomised to either primary PCI or fibrinolytic therapy with bolus tenecteplase.^[5] The lytic arm received bolus tenecteplase (amended to half dose at 75 years or older), clopidogrel and enoxaparin before transport.^[5] Angiography followed 6–24 h after randomisation unless lysis failed.^[5] Death, shock, heart failure or reinfarction at 30 days occurred in 12.4% with fibrinolysis vs 14.3% with primary PCI (RR 0.86, P=0.21).^[5] In the fibrinolysis group, emergency angiography was needed in 36.3%; the rest were catheterised at a median of 17 h after randomisation.^[5] Intracranial haemorrhage was 1.0% with fibrinolysis vs 0.2% with primary PCI (P=0.04), and 0.5% vs 0.3% after the amendment (P=0.45, not significant).^[5]
• STREAM-2 (2023): patients aged 60 or older with 2 mm or more ST elevation in 2 contiguous leads, unable to have primary PCI within 1 hour, randomised 2:1 to half-dose tenecteplase then angiography 6–24 h after randomisation, or primary PCI.^[6] The 30-day composite was 12.8% vs 13.3% (RR 0.96, 95% CI 0.62–1.48).^[6] Six intracranial haemorrhages (1.5%) occurred with half-dose tenecteplase, 3 of them protocol violations. None occurred with primary PCI.^[6] The authors call this a reasonable alternative if timely PCI is unavailable, provided fibrinolysis contraindications are observed and excess anticoagulation is avoided.^[6]

Fibrinolysis

Who benefits

ESC 2023: fibrinolysis is an important strategy within 12 h of symptom onset when primary PCI cannot be timely. It prevents 30 early deaths per 1000 patients treated within 6 h.^[1] The largest absolute benefit is in the highest-risk patients, including the elderly.^[1]

ACC/AHA 2025: the benefit is time-dependent over the first 12 hours. It is greatest in early presenters with low bleeding risk and large anterior infarcts.^[2] Benefit beyond 12 hours is not established.^[2] ACC/AHA 2025 says consideration should be given to a fibrinolytic after 12 hours in symptomatic patients with a large area at risk or haemodynamic instability, if PCI is unavailable.^[2] ACC/AHA 2025 also states there is no benefit, and potential harm, from fibrinolysis in patients with symptoms of ACS who have neither ST elevation nor suspected true posterior STEMI.^[2]

Aspirin and lysis work together. In ISIS-2 (17,187 patients with suspected acute MI, up to 24 hours from onset), the effects of streptokinase and aspirin on vascular death were additive: 8.0% with both vs 13.2% with neither.^[9][1]

Prehospital lysis (ESC 2023)

• When fibrinolysis is the reperfusion strategy, ESC 2023 recommends starting it as soon as possible after diagnosis in the prehospital setting, aiming for under 10 min to the lytic bolus (Class I, Level A).^[1] Its text adds that this applies if trained staff can interpret the ECG on site or transmit it.^[1]
• A fibrin-specific agent (tenecteplase, alteplase or reteplase) is recommended (Class I, Level B).^[1]
• Aim to start within 10 min of the STEMI diagnosis, and do not wait for biomarker results.^[1]
• In a meta-analysis of six trials, prehospital lysis reduced early mortality by 17% compared with in-hospital lysis, particularly within 2 h of symptom onset.^[1]
• In STREAM, prehospital lysis then early PCI gave outcomes similar to transfer for primary PCI, with a slight excess of intracranial bleeding. The patients presented within 3 h of onset and could not have primary PCI within 1 h of FMC.^[1] ESC 2023 notes this excess was blunted by halving the tenecteplase dose in patients older than 75. (The STREAM protocol amendment applied from age 75.)^[1][5] ESC 2023: a half-dose of tenecteplase should be considered in patients over 75 years (Class IIa, Level B).^[1]

Agents and doses

ESC 2023 places fibrinolytic doses in its Supplementary data, which is not held here.^[1] The doses below are from ACC/AHA 2025 Table 13.^[2]

^[2]

Streptokinase is no longer available in the United States.^[2] Fibrin-specific agents are preferred over non-fibrin-specific agents because of better patency and less immunogenicity.^[2] In a large trial, tenecteplase and alteplase gave equivalent 30-day mortality, with less non-cerebral bleeding on tenecteplase.^[2] Reteplase has not been shown to improve survival over alteplase, but its bolus dosing is easier.^[2]

Contraindications (ACC/AHA 2025 Table 14)

ESC 2023 also places its contraindication list in the Supplementary data.^[1] ACC/AHA 2025 lists:

^[2]

ACC/AHA 2025 calls the table advisory; it may not be all-inclusive or definitive.^[2]

Antithrombotics given with fibrinolysis

^{[1] [2]}

ESC 2023 notes that in ExTRACT–TIMI 25 (Enoxaparin and Thrombolysis Reperfusion for Acute myocardial infarction Treatment; 20,506 patients), enoxaparin reduced death and reinfarction at 30 days compared with weight-adjusted UFH, at the cost of more non-cerebral bleeding.^[1] Bivalirudin has not been studied with fibrin-specific agents.^[1] The most extensively studied pharmaco-invasive regimen is weight-adjusted IV tenecteplase, low-dose aspirin, oral clopidogrel, and enoxaparin IV then SC until PCI.^[1]

ACC/AHA 2025 recommends parenteral anticoagulation before and after fibrinolysis.^[2] If no invasive approach is planned, enoxaparin is preferred over UFH, and fondaparinux is an alternative.^[2] The clopidogrel age split is printed two ways. Table 7 uses 75 or younger vs older than 75. The supporting text gives a 300 mg load under 75 and 75 mg daily without a load at 75 or older.^[2] For PCI after a fibrinolytic, clopidogrel or ticagrelor (age under 75, within 24 hours of lysis) or prasugrel (more than 24 hours after lysis) are alternatives.^[2]

The pharmaco-invasive strategy

ESC 2023 recommends transfer to a PCI-capable centre in all patients immediately after fibrinolysis (Class I, Level A).^[1] Rescue PCI is recommended for failed fibrinolysis, defined as ST resolution under 50% within 60–90 min of the lytic, or with haemodynamic or electrical instability, worsening ischaemia or persistent chest pain (Class I, Level A).^[1] Emergency angiography and PCI of the IRA, if indicated, are recommended for new-onset or persistent heart failure or shock after fibrinolysis (Class I, Level A).^[1] Re-occlusion or reinfarction with recurrent ST elevation also needs immediate angiography and rescue PCI. Repeat fibrinolysis is discouraged.^[1]

Even when lysis appears to work, angiography and PCI of the IRA, if indicated, are recommended between 2 and 24 h after successful fibrinolysis (Class I, Level A). Signs of success include ST resolution over 50% at 60–90 min, a typical reperfusion arrhythmia and disappearance of chest pain.^[1] Routine early angiography reduced reinfarction and recurrent ischaemia compared with watchful waiting.^[1] In a network meta-analysis, the pharmaco-invasive strategy ranked second after primary PCI, with an OR for death of 0.79 (95% CI 0.59–1.08) against conventional fibrinolysis.^[1] Trial delays from lysis to PCI ranged from a median of 1.3 to 17 h.^[1]

ACC/AHA 2025 transfers every lysed patient to a PCI-capable hospital.^[2] It lists different signs that can indicate incomplete reperfusion: lack of improvement in ischaemic symptoms, persistent ST elevation (under 50% resolution in anterior leads or under 70% in inferior leads), or haemodynamic or electrical instability.^[2] Failed reperfusion means immediate angiography and rescue PCI. Everyone else has routine angiography within 2 to 24 hours.^[2] In a network meta-analysis, a pharmaco-invasive strategy (lysis-to-PCI interval of 2 hours or more) was associated with 21% lower mortality than lysis alone.^[2] In the rescue PCI trials, done in the femoral-access era, bleeding and stroke were higher with rescue PCI. With radial access and no routine GP IIb/IIIa inhibitors, ACC/AHA 2025 judges that the benefit outweighs the risk for most patients with failed reperfusion.^[2]

• REACT (2005): 427 patients with STEMI in whom reperfusion failed (less than 50% ST resolution) within 90 minutes after thrombolysis.^[7] The end point was death, reinfarction, stroke or severe HF at 6 months.^[7] Event-free survival was 84.6% with rescue PCI, 70.1% with conservative care and 68.7% with repeat thrombolysis.^[7] All-cause mortality did not differ, and non-fatal bleeding (mostly at the sheath site) was more common with rescue PCI.^[7]
• TRANSFER-AMI (2009): 1059 high-risk patients with STEMI receiving fibrinolysis at hospitals without PCI capability, randomised to immediate transfer to another hospital and PCI within 6 hours after fibrinolysis, or to standard treatment (including rescue PCI if required, or delayed angiography).^[8] The 30-day composite of death, reinfarction, recurrent ischaemia, new or worsening HF or shock was 11.0% vs 17.2% (RR 0.64).^[8] Major bleeding did not differ.^[8]

Antithrombotic therapy with primary PCI

Doses

^{[1] [2]}

Which P2Y12 inhibitor?

ESC 2023 recommends DAPT with aspirin plus prasugrel or ticagrelor as the default for ACS.^[1] Aspirin is recommended for all patients without contraindications, and a P2Y12 inhibitor is added for 12 months unless there is high bleeding risk (both Class I, Level A). Prasugrel is recommended in P2Y12-naïve patients proceeding to PCI, and ticagrelor irrespective of treatment strategy (both Class I, Level B).^[1] Clopidogrel is recommended when prasugrel or ticagrelor are not available, cannot be tolerated, or are contraindicated (Class I, Level C).^[1] In older ACS patients, especially with high bleeding risk, clopidogrel may be considered (Class IIb, Level B); ESC 2023 notes that trial definitions of older patients range from 70 to 80 years, and that frailty and comorbidities should also be taken into consideration.^[1] Prasugrel should be considered in preference to ticagrelor for ACS patients who proceed to PCI (Class IIa, Level B).^[1] ESC 2023 cites ISAR-REACT 5, the largest head-to-head comparison of the two drugs.^[1] It randomised 4018 ACS patients, with or without ST elevation, for whom invasive evaluation was planned. Death, MI or stroke at 1 year occurred in 9.3% with ticagrelor vs 6.9% with prasugrel (HR 1.36). Major bleeding did not differ significantly.^[26]

ACC/AHA 2025 recommends ticagrelor or prasugrel in preference to clopidogrel for ACS patients undergoing PCI.^[2] TRITON-TIMI 38 randomised 13,608 patients with moderate-to-high-risk ACS with scheduled PCI to prasugrel or clopidogrel.^[35] In its subgroup analyses, cited by ACC/AHA 2025, prasugrel showed net harm versus clopidogrel in patients with previous TIA or stroke.^[2] With ticagrelor, aspirin should always be 100 mg daily or less.^[2] Clopidogrel is an alternative for STEMI patients at high bleeding risk or with contraindications to the potent agents.^[2]

Pretreatment. ATLANTIC randomised 1862 patients with ongoing STEMI of less than 6 hours' duration to prehospital (in the ambulance) or in-hospital (in the catheterisation laboratory) ticagrelor.^[25] The strategies differed by a median of 31 minutes.^[25] The co-primary reperfusion end points did not differ.^[25] Definite stent thrombosis was lower with prehospital loading (0.2% vs 1.2% at 30 days).^[25] ESC 2023 says pre-treatment with a P2Y12 inhibitor may be considered in patients undergoing a primary PCI strategy (Class IIb, Level B). If not pretreated, a loading dose is recommended at the time of PCI.^[1]

Intravenous antiplatelets. ESC 2023 finds no strong evidence for routine GP IIb/IIIa inhibitors. GP IIb/IIIa antagonists should be considered if there is evidence of no-reflow or a thrombotic complication during PCI (Class IIa, Level C).^[1] In P2Y12-naïve patients undergoing PCI, cangrelor may be considered (Class IIb, Level A); its text adds case by case, including patients who cannot take oral drugs, such as those in shock or on ventilation.^[1] ACC/AHA 2025 notes that, in current practice, GP IIb/IIIa inhibitors are largely limited to large thrombus burden or bailout for no-reflow or persistent or recurring thrombus. They can be considered at the time of PCI for a large thrombus burden, or no-reflow or slow flow attributed to distal thrombus embolisation.^[2]

Anticoagulation

ESC 2023 recommends parenteral anticoagulation for all ACS patients at the time of diagnosis (Class I, Level A).^[1] For primary PCI, UFH is the default choice at present; routine use of a weight-adjusted UFH bolus of 70–100 IU/kg during PCI is recommended in patients undergoing PCI (Class I, Level C).^[1] In STEMI undergoing primary PCI, enoxaparin (Class IIa, Level A) and bivalirudin with a full-dose post-PCI infusion (Class IIa, Level A) should be considered as alternatives to UFH, and fondaparinux is not recommended (Class III, Level B).^[1] ESC 2023 adds that further studies are required to confirm the bivalirudin findings in non-East Asian populations.^[1] It is the recommended alternative to UFH after heparin-induced thrombocytopenia.^[1] In general, avoid crossover between anticoagulants, especially between UFH and LMWH.^[1] Discontinuation of parenteral anticoagulation should be considered immediately after an invasive procedure (Class IIa, Level C): anticoagulants should generally be stopped immediately after PCI, except in specific settings such as confirmed LV aneurysm with thrombus or AF requiring anticoagulation.^[1]

ACC/AHA 2025 recommends a full-dose bivalirudin infusion for 2–4 hours after the procedure whenever bivalirudin supports primary PCI.^[2] Fondaparinux should not be used to support PCI because of the risk of catheter thrombosis.^[2]

• BRIGHT-4 (2022): 6016 patients in China with STEMI undergoing primary PCI with radial artery access (used in 93.1%) within 48 h of symptom onset, without prior fibrinolysis, anticoagulants or GP IIb/IIIa inhibitors. Bivalirudin plus a 2–4 h high-dose post-PCI infusion was compared with UFH monotherapy.^[23] Death or BARC 3–5 bleeding at 30 days fell from 4.39% to 3.06% (HR 0.69).^[23] Stent thrombosis within 30 days was 0.37% with bivalirudin vs 1.10% with heparin.^[23]
• ATOLL (2011): 910 patients with STEMI randomised to IV enoxaparin 0.5 mg/kg or UFH before primary PCI. Patients who had received any anticoagulant before randomisation were excluded.^[24] The primary end point was 28% vs 34% (RR 0.83, P=0.06), and the main secondary end point 7% vs 11% (RR 0.59).^[24]

After PCI, ESC 2023 generally recommends DAPT with prasugrel or ticagrelor plus aspirin for 12 months, unless contraindicated (formal row: P2Y12 inhibitor for 12 months unless there is high bleeding risk, Class I, Level A).^[1]

Initial medical therapy

^{[1] [2]}

• Oxygen: DETO2X-AMI randomised 6629 patients with suspected MI and SpO2 of 90% or more to oxygen at 6 L/min for 6–12 hours or ambient air. One-year death was 5.0% vs 5.1% (HR 0.97, P=0.80, not significant).^[27]
• Opioids: morphine may slow gut absorption of oral medicines, which may delay the action of oral antiplatelets.^[1] ESC 2023 notes that current data show no increase in adverse clinical outcomes from this interaction.^[1] ACC/AHA 2025 says opiates should not be used just to mask ongoing ischaemia. Non-aspirin NSAIDs should be avoided for ischaemic pain whenever possible.^[2]
• Beta-blockers: ESC 2023 links early IV metoprolol with less VF and less MVO, and finds it safe without acute HF.^[1] ACC/AHA 2025 cites COMMIT (Clopidogrel and Metoprolol in Myocardial Infarction; 45,852 patients, about 50% given fibrinolytic therapy and 7% with NSTE-ACS): high-dose metoprolol reduced reinfarction and VF but increased cardiogenic shock, mainly in the first 24 hours.^[2] ACC/AHA 2025 recommends early beta-blocker therapy (within 24 hours) in most ACS patients without contraindications, for example Killip II–IV HF, a low-output state or risk of shock, second- or third-degree block without a pacemaker, severe bradycardia or active bronchospasm.^[2]

Primary PCI technique

Access. ESC 2023 recommends radial access as the standard approach, unless there are overriding procedural considerations (Class I, Level A). Femoral access may still be chosen selectively.^[1] MATRIX randomised 8404 patients with ACS, with or without STEMI, who were about to undergo coronary angiography and PCI; 47.6% had STEMI.^[1][13] Net adverse clinical events fell from 11.7% to 9.8% with radial access (RR 0.83).^[13] The benefit came from less BARC major bleeding unrelated to CABG (1.6% vs 2.3%) and lower all-cause mortality (1.6% vs 2.2%). MACE alone was not significant at the trial's α of 0.025.^[13] In ACC/AHA 2025, a 7-trial individual-patient meta-analysis linked radial access with 24% less all-cause death and 51% less major bleeding.^[2] Femoral access, preferably ultrasound-guided, should be considered when temporary MCS is planned.^[2]

Thrombus aspiration. Routine thrombus aspiration is not recommended by ESC 2023 (Class III, Level A). Its text adds that aspiration may be considered for large residual thrombus after opening the vessel.^[1] TASTE (7244 patients with STEMI undergoing PCI) found no significant difference in 30-day death: 2.8% vs 3.0% (HR 0.94, P=0.63).^[11] TOTAL randomised 10,732 patients with STEMI undergoing primary PCI to routine upfront manual thrombectomy or PCI alone. Thrombectomy did not reduce the primary outcome of cardiovascular death, recurrent MI, cardiogenic shock or NYHA class IV HF within 180 days (6.9% vs 7.0%), and stroke within 30 days was higher with thrombectomy (0.7% vs 0.3%).^[12] ACC/AHA 2025 allows bailout aspiration for thrombus that persists after balloon or stent, particularly with no-reflow. An estimated 4% to 7% of primary PCI patients need it.^[2]

Stents, physiology and imaging. ESC 2023 recommends stent deployment in the IRA during the index procedure in patients undergoing primary PCI (Class I, Level A), and drug-eluting stents in preference to bare metal stents in all cases (Class I, Level A).^[1] ESC 2023: new-generation drug-eluting stents are associated with superior safety and efficacy compared with bare metal stents.^[1] Intravascular imaging should be considered to guide PCI (Class IIa, Level A).^[1] ESC 2023: PCI of the IRA should not be deferred on the basis of invasive epicardial functional assessment.^[1] ACC/AHA 2025 recommends intracoronary imaging to guide PCI of complex lesions.^[2]

When surgery comes first. ESC 2023: CABG should be considered with an occluded IRA when primary PCI is not feasible or unsuccessful and a large area of myocardium is in jeopardy (Class IIa, Level C).^[1] Its text adds that emergency CABG should be considered with a patent IRA but anatomy unsuitable for PCI and a large area at jeopardy; in cardiogenic shock, emergency CABG is recommended if PCI of the IRA is not feasible or unsuccessful (Class I, Level B).^[1] CABG is recommended at the time of repair of a mechanical complication if revascularisation is needed.^[1] ACC/AHA 2025: CABG can be the primary reperfusion strategy when PCI is impossible for anatomical reasons or complex disease.^[2] After failed primary PCI, emergency CABG is not recommended without a large area at risk, or if surgical targets are poor and surgery may carry a higher risk of death and adverse outcomes than non-surgical management.^[2]

^[1]

Multivessel disease and complete revascularisation

About half of primary PCI patients have multivessel disease, which carries an adverse prognosis.^[1] In a systematic review of 10 randomised trials (7030 patients with STEMI and multivessel disease), complete revascularisation was associated with reduced cardiovascular mortality compared with IRA-only PCI. All-cause mortality was similar.^[1] ESC 2023 recommends basing the revascularisation strategy (IRA PCI, multivessel PCI or CABG) on clinical status, comorbidities and disease complexity (Class I, Level B).^[1]

^{[1] [2]}

ESC 2023 makes no recommendation for immediate vs staged non-IRA PCI. In COMPLETE (patients with STEMI and multivessel disease), non-IRA PCI happened in hospital in 67% and after discharge in 33%, always within 45 days.^[1][14] ACC/AHA 2025 describes the patients best suited to immediate complete revascularisation. They have uncomplicated IRA PCI, low-complexity non-IRA disease, stable haemodynamics, normal LV filling pressures and normal renal function.^[2]

• COMPLETE (2019): patients with STEMI and multivessel disease, after successful culprit PCI, were randomised to PCI of angiographically significant non-culprit lesions or no further revascularisation.^[14] Cardiovascular death or MI at a median of 3 years was 7.8% vs 10.5% (HR 0.74).^[14] The benefit was consistent regardless of intended timing.^[14]
• MULTISTARS AMI (2023): 840 haemodynamically stable patients with STEMI and multivessel disease randomised to immediate multivessel PCI or staged PCI 19–45 days after the index procedure.^[15] The 1-year composite (death, MI, stroke, unplanned ischaemia-driven revascularisation or HF hospitalisation) was 8.5% vs 16.3% (RR 0.52). Immediate PCI was noninferior and superior.^[15] Death, stroke and HF hospitalisation appeared similar.^[15]
• FLOWER-MI (2021): in patients with STEMI and multivessel disease after successful PCI of the IRA, FFR-guided vs angiography-guided complete revascularisation gave death, MI or urgent revascularisation at 1 year in 5.5% vs 4.2% (HR 1.32, P=0.31, not significant). The authors say the wide confidence intervals prevent a conclusive interpretation.^[16] ACC/AHA 2025 cites a meta-analysis that did not suggest a differential benefit for FFR-guided over angiography-guided complete revascularisation.^[2]

Late presenters

ESC 2023: beyond 12 h from symptom onset, a primary PCI strategy is recommended with ongoing symptoms suggestive of ischaemia, haemodynamic instability or life-threatening arrhythmias (Class I, Level C).^[1] A routine primary PCI strategy should be considered in patients presenting late, 12–48 h after symptom onset (Class IIa, Level B); the text notes its value is less well established than within 12 h.^[1] A small RCT of 347 patients at 12–48 h without persistent symptoms found that a routine primary PCI strategy improved myocardial salvage and long-term survival compared with conservative treatment.^[1] Routine PCI of an occluded IRA is not recommended in STEMI presenting more than 48 h after symptom onset without persistent symptoms (Class III, Level A).^[1]

ACC/AHA 2025: the benefit of primary PCI begins to diminish after 12 hours but appears to continue to about 24 hours.^[2] Transfer for PCI at 12–24 hours is reasonable.^[2] Revascularisation should be considered for late presenters with ongoing ischaemia, cardiogenic shock, acute severe HF, persistent angina or life-threatening arrhythmias.^[2] PCI is not recommended for an occluded IRA in an asymptomatic patient with a completed infarct.^[2] At a non-PCI hospital, a patient presenting at 12 hours or later should be transferred when possible. If they are haemodynamically unstable or have a large territory at risk and timely PCI is impossible, fibrinolysis likely outweighs prolonged transfer.^[2]

OAT (2006): 2166 stable patients with total occlusion of the IRA 3 to 28 days after MI who met a high-risk criterion (EF under 50% or proximal occlusion).^[10] The 4-year rate of death, reinfarction or class IV HF was 17.2% with PCI vs 15.6% with medical therapy (HR 1.16, P=0.20, not significant).^[10]

Cardiogenic shock

^{[1] [2]}

ESC 2023: shock is more frequent with complete coronary occlusion. Ischaemic HF, acute severe mitral regurgitation and mechanical complications are the main precipitants.^[1] Patients should go as soon as possible to a tertiary shock centre with a Shock Team.^[1] Immediate coronary angiography and PCI of the IRA (if indicated) is recommended in cardiogenic shock complicating ACS (Class I, Level B). Emergency CABG is recommended for ACS-related shock if PCI of the IRA is not feasible or unsuccessful (Class I, Level B).^[1] In cardiogenic shock with multivessel disease, IRA-only PCI during the index procedure is recommended (Class I, Level B); staged PCI of non-IRA should be considered (Class IIa, Level C), based on ischaemia, symptoms, comorbidities and clinical condition.^[1] Fibrinolysis should be considered in STEMI with shock if a primary PCI strategy is not available within 120 min of diagnosis and mechanical complications have been ruled out (Class IIa, Level C).^[1] The routine use of an IABP in shock without mechanical complications is not recommended (Class III, Level B). In severe or refractory shock, short-term MCS may be considered (Class IIb, Level C); the text advises caution until more randomised data are available.^[1] A benefit of LV assist devices in ACS has not been demonstrated in ESC 2023, which was published before DanGer Shock (2024).^[1][2][22]

ACC/AHA 2025: primary PCI in STEMI with shock should be done as soon as possible, ideally within 90 minutes.^[2] In an observational analysis from FITT-STEMI, for patients with cardiogenic shock, each 10-minute delay in primary PCI after 60 minutes from FMC added 3 to 4 deaths per 100, with mortality over 80% beyond 6 hours of delay from FMC.^[2] Culprit-vessel-only PCI is recommended, and emergency CABG when PCI is not feasible.^[2] CULPRIT-SHOCK enrolled only patients with an identifiable culprit lesion.^[2]

Mechanical support: the guidelines differ. ACC/AHA 2025 finds a microaxial flow pump reasonable to reduce mortality in STEMI shock patients who resemble the DanGer Shock population.^[2] The trial enrolled STEMI with cardiogenic shock of under 24 hours: SBP under 100 mm Hg or vasopressor support, end-organ hypoperfusion (lactate 2.5 mmol/L or more and/or SvO2 under 55% with a normal PaO2), and LVEF under 45%.^[2] Patients comatose (GCS under 8) after out-of-hospital cardiac arrest and those with overt RV failure were excluded.^[2] ACC/AHA 2025 considers pump use reasonable in patients in SCAI shock stages C, D or E who are non-comatose, with vessels that can take large-bore access.^[2] Bleeding, limb ischaemia and renal failure are higher with the pump.^[2] Neither guideline supports routine IABP in infarct-related shock (ESC 2023: in the absence of mechanical complications); ACC/AHA 2025 also does not support routine VA-ECMO.^[1][2]

• SHOCK (1999): patients with shock due to LV failure complicating MI were randomised to emergency revascularisation (152) or initial medical stabilisation (150).^[17] Thirty-day mortality was 46.7% vs 56.0% (P=0.11), but 6-month mortality was lower with revascularisation (50.3% vs 63.1%).^[17]
• CULPRIT-SHOCK (2017): 706 patients with multivessel disease, acute MI and cardiogenic shock.^[18] Death or renal replacement therapy at 30 days was 45.9% with culprit-only PCI vs 55.4% with immediate multivessel PCI (RR 0.83).^[18] At 1 year, death was 50.0% vs 56.9% (RR 0.88, not significant), and repeat revascularisation was more frequent with culprit-only PCI (32.3% vs 9.4%).^[19]
• IABP-SHOCK II (2012): 600 patients with cardiogenic shock complicating acute MI, all expected to undergo early revascularisation; 30-day mortality 39.7% with IABP vs 41.3% without (RR 0.96, P=0.69, not significant).^[20]
• ECLS-SHOCK (2023): 420 patients with acute MI complicated by cardiogenic shock and planned early revascularisation; 30-day death 47.8% with early ECLS vs 49.0% with usual medical treatment alone (RR 0.98, P=0.81, not significant). Moderate or severe bleeding was 23.4% vs 9.6%.^[21]
• DanGer Shock (2024): 360 patients with STEMI and cardiogenic shock randomised to an Impella CP microaxial flow pump plus standard care or standard care alone.^[22] Death at 180 days was 45.8% vs 58.5% (HR 0.74).^[22] The composite safety end point was 24.0% vs 6.2%, and renal replacement therapy 41.9% vs 26.7%.^[22]

Resuscitated cardiac arrest

ESC 2023 recommends a primary PCI strategy after resuscitated cardiac arrest with persistent ST elevation or an equivalent (Class I, Level B). Routine immediate angiography after resuscitated cardiac arrest is not recommended in haemodynamically stable patients without persistent ST elevation or equivalents (Class III, Level A).^[1] Neurological status and survival probability belong in the decision.^[1] In OHCA with an initial shockable rhythm, without ST elevation or equivalents and without shock, routine immediate angiography was not superior to a delayed invasive strategy (COACT, TOMAHAWK).^[1] Temperature control, with continuous core temperature monitoring and active prevention of fever over 37.7°C, is recommended for adults who remain unresponsive after ROSC (Class I, Level B). Evaluation of neurological prognosis, no earlier than 72 h after admission, is recommended in all comatose survivors (Class I, Level C).^[1]

ACC/AHA 2025: about 10% of EMS-transferred STEMI patients are estimated to have had an out-of-hospital arrest.^[2] Nearly one-third of arrest patients with STEMI have normal neurological status in the ED.^[2] Awake patients with ROSC and STEMI should have primary PCI.^[2] Comatose patients need an individual assessment of survival and futility first.^[2] Poor prognostic features include unwitnessed arrest, no bystander CPR and a non-shockable rhythm. Others are CPR or time to ROSC over 30 minutes, pH under 7.2, lactate over 7 mmol/L, age over 85 and end-stage renal disease on dialysis.^[2]

Complications

Mechanical complications

The most commonly encountered lesions are papillary muscle rupture with acute mitral regurgitation, ventricular septal defect, pseudoaneurysm and free wall rupture.^[32] ACC/AHA 2025 prefers the term "contained rupture" for pseudoaneurysm.^[2] An AHA scientific statement (Damluji 2021) notes that patients with large infarcts, or without timely revascularisation, remain at risk.^[32]

ESC 2023: they may occur in the first days after MI, most commonly after STEMI.^[1] In the National Inpatient Sample database (2003 to September 2015), they complicated 0.27% of STEMI hospitalisations, with in-hospital mortality of 42.4%.^[31] ACC/AHA 2025 describes presentation within the first week: recurrent or refractory chest pain, or a new murmur with disproportionate HF, shock or sudden death.^[2]

^{[1] [2]}

Medical therapy alone carries a high risk of early death.^[2]

Arrhythmias and conduction block

• ESC 2023: 6–8% of patients with STEMI develop haemodynamically significant VT or VF. Urgent reperfusion matters most, because ischaemia is often the trigger.^[1]
• ESC 2023 recommends IV beta-blocker and/or amiodarone for polymorphic VT and/or VF unless contraindicated (Class I, Level B), and prompt, complete revascularisation to treat ischaemia in recurrent VT or VF (Class I, Level C). Lidocaine may be considered for recurrent, haemodynamically relevant VT despite repeated electrical cardioversion, if beta-blockers, amiodarone and overdrive stimulation are not effective or applicable (Class IIb, Level C).^[1]
• ESC 2023: accelerated idioventricular rhythm at reperfusion is frequent, benign and needs no intervention.^[1]
• ESC 2023: sustained VT or VF more than 48 h after reperfusion requires evaluation for a secondary-prevention ICD.^[1]
• ESC 2023: AF is the most frequent supraventricular arrhythmia in ACS.^[1]
• ACC/AHA 2025: STEMI with second- or third-degree AV block has higher in-hospital mortality. It recommends a permanent pacemaker if high-degree block persists beyond 72 hours.^[2] ESC 2023 instead recommends a permanent pacemaker when high-degree AV block does not resolve within a waiting period of at least 5 days after MI (Class I, Level C).^[1]

LV thrombus, pericarditis and heart failure

According to a large meta-analysis, LV thrombus can be present in over 9% of patients after anterior STEMI. ESC 2023: oral anticoagulation (VKA or NOAC) should be considered for 3–6 months in confirmed LV thrombus (Class IIa, Level C), guided by repeat echocardiography or CMR, with consideration of bleeding risk and the need for concomitant antiplatelet therapy.^[1] Early infarct-associated pericarditis occurs from a few hours to 4 days after MI. Dressler syndrome typically occurs at 1–2 weeks.^[1] Acute HF, including shock, complicating ACS requires immediate angiography.^[1]

In-hospital care, risk and the ICD question

• ESC 2023 recommends CCU or ICCU admission for all STEMI patients after reperfusion.^[1] After successful reperfusion with an uncomplicated course, it is recommended that STEMI patients are kept in the CCU or ICCU for at least 24 h whenever possible, after which they may be moved to a step-down monitored bed for a further 24–48 h (Class I, Level C).^[1]
• ECG monitoring for a minimum of 24 h is recommended in all STEMI patients (Class I, Level C).^[1]
• Early ambulation (out of bed on day 1) is recommended for most patients.^[1]
• Routine echocardiography is recommended during hospitalisation to assess LV function, detect mechanical complications and exclude LV thrombus (Class I, Level C), and LVEF should be measured before discharge.^[1] Echo can also be used to exclude early mechanical complications and LV thrombus.^[1]
• High-risk markers include older age, tachycardia, hypotension, Killip class above I, anterior MI, previous MI, raised creatinine, HF history, peripheral arterial disease and anaemia.^[1] ESC 2023 says established risk scores such as GRACE should be considered for prognosis estimation (Class IIa, Level B); its text says GRACE offers the best discriminative performance.^[1]

ICD timing. ESC 2023: primary-prevention ICD within 40 days of MI is generally not indicated. An ICD or a temporary wearable defibrillator may be considered within 40 days in selected patients: incomplete revascularisation, pre-existing LV dysfunction, arrhythmias more than 48 h after STEMI onset, or polymorphic VT or VF (Class IIb, Level C).^[1] If pre-discharge LVEF is 40% or less, repeat LVEF evaluation 6–12 weeks after the ACS (and after complete revascularisation and optimal medical therapy) is recommended to assess the need for a primary-prevention ICD (Class I, Level C).^[1] The trials explain the wait:

• DINAMIT (2004): ICD (332 patients) vs no ICD (342 patients) 6 to 40 days after MI, in patients with LVEF 0.35 or less and impaired cardiac autonomic function. Overall mortality did not differ (HR 1.08, P=0.66).^[33]
• IRIS (2009): 898 patients enrolled 5 to 31 days after MI who had either LVEF 40% or less with a heart rate of 90/min or more on the first ECG, or non-sustained VT (150/min or more) on Holter, or both. The ICD did not reduce overall mortality (HR 1.04).^[34]

ACC/AHA 2025 explains DINAMIT: fewer arrhythmic deaths were offset by more non-arrhythmic deaths.^[2] MADIT II, which reduced all-cause mortality, enrolled patients with LVEF 30% or less, at least 30 days after MI or 3 months after revascularisation, with expected survival over 1 year.^[2] A wearable defibrillator could be considered as a bridge in high-risk patients with reduced LVEF.^[2]

Special populations

Older patients. In the absence of robust RCT data, ESC 2023 says primary PCI should be considered for all patients with STEMI, and fibrinolysis may be reasonable when primary PCI cannot be timely.^[1] Decisions in older patients should be individualised based on patient characteristics. 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.^[1] ESC 2023 recommends the same diagnostic and treatment strategies in older patients as in younger patients (Class I, Level B), and a half-dose of tenecteplase should be considered in patients over 75 years (Class IIa, Level B).^[1] Age changes several doses: prasugrel: use with caution at 75 or older, 5 mg daily if treatment is deemed necessary; clopidogrel 75 mg initial dose with lysis over 75 (ESC 2023).^[1] With lysis at 75 or older, ACC/AHA 2025 omits the enoxaparin bolus.^[2] STREAM-2 tested half-dose tenecteplase in patients aged 60 or older with 2 mm or more ST elevation in 2 contiguous leads who could not have primary PCI within 1 hour.^[6]

Chronic kidney disease. Moderate to severe CKD is present in more than 30% of ACS patients.^[1] ESC 2023 recommends the same diagnostic and therapeutic strategies as for normal kidney function, with dose adjustment where needed (Class I, Level C).^[1] The type and dose of antithrombotic and the amount of contrast should be considered based on kidney function; low- or iso-osmolar contrast at the lowest possible volume is recommended for invasive strategies (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] Examples: bivalirudin infusion 1 mg/kg/h if CrCl is under 30 (ESC 2023); fondaparinux with lysis is contraindicated if CrCl is under 30 (ACC/AHA 2025).^[1][2]

Pregnancy. ESC 2023: pregnant women with STEMI should not be managed differently to non-pregnant women, and primary PCI is the preferred reperfusion therapy.^[1] A multidisciplinary team of cardiologists, obstetricians, anaesthetists and neonatologists should plan care.^[1] Do not delay ACS treatment for delivery; ideally postpone delivery for at least 2 weeks.^[1] SCAD is the most common cause of AMI in pregnancy.^[1] Pregnancy is a relative contraindication to fibrinolysis in ACC/AHA 2025.^[2]

Anticoagulated patients. ACC/AHA 2025 lists oral anticoagulant therapy as a relative contraindication to fibrinolysis.^[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.^[36] 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.^[36] Each GRADE recommendation in the summary tables carries a strength of recommendation and a certainty of evidence, which the tables below reproduce as printed.^[36] 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 —).^[36]

NHFA/CSANZ 2025 writes its STEMI rows as STEMI/ACOMI, where ACOMI is acute coronary occlusion myocardial infarction, a term that incorporates both STEMI and STEMI equivalents.^[36] Its Table 3 footnote says the acute STEMI/ACOMI management recommendations apply to people with STEMI and to people with ECG changes that may not be recognised as STEMI but are indicative of ACOMI.^[36] The footnote names these changes: high lateral MI, posterior MI, right ventricular MI, De Winter T waves, and left bundle branch block with modified Sgarbossa criteria.^[36] The ACOMI patterns beyond traditional ST-elevation criteria also include transient ST elevation, and recognition of these patterns should prompt urgent consultation with cardiology services for consideration of reperfusion strategy.^[36]

Recognition and the first ECG

^[36]

Reperfusion and its timing

^[36]

Timely reperfusion limits the extent of myocardial infarction and reduces mortality by minimising total ischaemic time (NHFA/CSANZ 2025).^[36] The guideline emphasises adopting quality assurance targets to reduce delays to wire crossing if primary PCI is the chosen reperfusion strategy.^[36] Its summary text sets first medical contact to wire crossing at no more than 60 min for people with STEMI/ACOMI presenting to primary PCI-capable centres, and no more than 90 min for people presenting to non-PCI-capable centres.^[36]

Fibrinolysis and what follows

^[36]

Primary PCI technique and late presentation

^[36]

Initial and antithrombotic therapy

^[36]

In people with STEMI/ACOMI undergoing primary PCI, the NHFA/CSANZ 2025 summary says P2Y12 inhibitor pretreatment before angiography has not shown clear mortality or bleeding benefits, though pre-hospital administration may reduce reinfarction.^[36] Pretreatment can be considered when STEMI/ACOMI is confirmed; deferring the P2Y12 inhibitor until the coronary anatomy is known is reasonable if the diagnosis is uncertain or cardiothoracic surgery may be needed.^[36] CHA2DS2VA is a scoring system that evaluates the risk of stroke in people with atrial fibrillation (NHFA/CSANZ 2025 Table 3 footnote).^[36]

Cardiac arrest, cardiogenic shock, multivessel disease, mechanical complications and SCAD

^[36]

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.^[36] It supports left ventricular assist devices in select people with STEMI/ACOMI and cardiogenic shock, given potential survival benefits despite increased bleeding and vascular complication risks.^[36] It cautions that the benefits of complete revascularisation in STEMI/ACOMI with MVD without cardiogenic shock may not apply to complex MVD, where coronary artery bypass grafting (CABG) may be more appropriate.^[36] 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.^[36] 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.^[36] In cases of haemodynamic compromise or significant ongoing ischaemia, urgent revascularisation with PCI or CABG may be required.^[36]

Systems of care

^[36]

Marked inequities in both the management and outcomes of ACS persist in Australia, and the NHFA/CSANZ 2025 summary describes them in these groups:^[36]

• Women: longer symptom duration, under or delayed diagnosis and treatment, including lower rates of secondary prevention medication use.^[36]
• First Nations peoples: coronary events at twice the rate of non-Indigenous Australians, exacerbated by ongoing inequities in healthcare access and outcomes.^[36]
• Older adults: a higher rate of ACS presentations, often complicated by atypical symptoms and age-related comorbidities.^[36]
• People living in regional and remote areas: higher rates of missed diagnosis and poorer outcomes.^[36]

Where NHFA/CSANZ 2025 and ESC 2023 differ

^{[36] [1]}

ACC/AHA 2025 measures its primary PCI limits from first medical contact to first device: 90 minutes or less, or 120 minutes or less in transfer patients.^[2]

High-yield summary

• ECG criteria: new (ACC/AHA: or presumed new) J-point ST elevation in at least two contiguous leads. In V2–V3: 2.5 mm or more (men under 40), 2 mm or more (men 40 or older), 1.5 mm or more (women of any age). Other leads: 1 mm or more (ESC 2023: in the absence of LV hypertrophy or LBBB).^[1][2]
• Other ECG findings suggesting occlusion (examples in the ESC 2023 text): V1–V3 ST depression (especially with a positive terminal T wave) or V7–V9 elevation (posterior); V3R/V4R elevation (RV); ST depression of 1 mm or more in six or more leads with aVR and/or V1 elevation (left main or multivessel, particularly with haemodynamic compromise).^[1]
• Clocks: ECG within 10 min (ESC 2023 Class I, Level B); primary PCI within 120 min of diagnosis (ESC) or 90/120 minutes from FMC (ACC/AHA); door-in to door-out 30 min or less (ESC); lysis within 10 min (ESC).^[1][2]
• After lysis: transfer all patients immediately; rescue PCI if it fails; otherwise angiography and PCI of the IRA, if indicated, 2–24 h after successful lysis (ESC 2023: Class I, Level A for each).^[1][2]
• Tenecteplase (ACC/AHA 2025): 30 mg under 60 kg; 35 mg at 60–69 kg; 40 mg at 70–79 kg; 45 mg at 80–89 kg; 50 mg at 90 kg or more.^[2]
• P2Y12: prasugrel or ticagrelor over clopidogrel. ESC 2023 says prasugrel should be considered in preference to ticagrelor for ACS patients proceeding to PCI (Class IIa, Level B; ISAR-REACT 5).^[1][26]
• Anticoagulation for primary PCI (ESC 2023): UFH default; enoxaparin or bivalirudin with a full-dose post-PCI infusion should be considered (Class IIa, Level A each); fondaparinux not recommended (Class III, Level B).^[1]
• Cath lab: radial as the standard approach unless there are overriding procedural considerations (Class I, Level A), no routine aspiration (Class III, Level A; TASTE, TOTAL), complete revascularisation, during the index PCI or within 45 days, in haemodynamically stable STEMI with multivessel disease undergoing primary PCI (Class I, Level A; COMPLETE), all ESC 2023.^{[1][11][12][14]}
• Shock: culprit-only PCI during the index procedure (ESC 2023 Class I, Level B; CULPRIT-SHOCK); no routine IABP without mechanical complications (ESC 2023 Class III, Level B; IABP-SHOCK II); ECLS did not reduce 30-day death (ECLS-SHOCK); microaxial pump reduced 180-day death but increased complications (DanGer Shock).^{[1][18][20][21][22]}
• Late presenters: routine PCI of an occluded IRA more than 48 h after symptom onset, without persistent symptoms, is not recommended (ESC 2023 Class III, Level A; OAT).^[1][10]
• ICD: primary-prevention ICD generally not indicated within 40 days of MI; if pre-discharge LVEF is 40% or less, re-check it 6–12 weeks after the ACS, after complete revascularisation and optimal therapy (ESC 2023 Class I, Level C; DINAMIT, IRIS).^[1][33][34]

References37ShowHide

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]Andersen HR, et al. A comparison of coronary angioplasty with fibrinolytic therapy in acute myocardial infarction. N Engl J Med, 2003.PMID 12930925
4. [4]Widimský P, et al. Long distance transport for primary angioplasty vs immediate thrombolysis in acute myocardial infarction. Final results of the randomized national multicentre trial--PRAGUE-2. Eur Heart J, 2003.PMID 12559941
5. [5]Armstrong PW, et al. Fibrinolysis or primary PCI in ST-segment elevation myocardial infarction. N Engl J Med, 2013.PMID 23473396
6. [6]Van de Werf F, et al. STREAM-2: Half-Dose Tenecteplase or Primary Percutaneous Coronary Intervention in Older Patients With ST-Segment-Elevation Myocardial Infarction: A Randomized, Open-Label Trial. Circulation, 2023.PMID 37439219
7. [7]Gershlick AH, et al. Rescue angioplasty after failed thrombolytic therapy for acute myocardial infarction. N Engl J Med, 2005.PMID 16382062
8. [8]Cantor WJ, et al. Routine early angioplasty after fibrinolysis for acute myocardial infarction. N Engl J Med, 2009.PMID 19553646
9. [9]Randomised trial of intravenous streptokinase, oral aspirin, both, or neither among 17,187 cases of suspected acute myocardial infarction: ISIS-2. ISIS-2 (Second International Study of Infarct Survival) Collaborative Group. Lancet, 1988.PMID 2899772
10. [10]Hochman JS, et al. Coronary intervention for persistent occlusion after myocardial infarction. N Engl J Med, 2006.PMID 17105759
11. [11]Fröbert O, et al. Thrombus aspiration during ST-segment elevation myocardial infarction. N Engl J Med, 2013.PMID 23991656
12. [12]Jolly SS, et al. Randomized trial of primary PCI with or without routine manual thrombectomy. N Engl J Med, 2015.PMID 25853743
13. [13]Valgimigli M, et al. Radial versus femoral access in patients with acute coronary syndromes undergoing invasive management: a randomised multicentre trial. Lancet, 2015.PMID 25791214
14. [14]Mehta SR, et al. Complete Revascularization with Multivessel PCI for Myocardial Infarction. N Engl J Med, 2019.PMID 31475795
15. [15]Stähli BE, et al. Timing of Complete Revascularization with Multivessel PCI for Myocardial Infarction. N Engl J Med, 2023.PMID 37634190
16. [16]Puymirat E, et al. Multivessel PCI Guided by FFR or Angiography for Myocardial Infarction. N Engl J Med, 2021.PMID 33999545
17. [17]Hochman JS, et al. Early revascularization in acute myocardial infarction complicated by cardiogenic shock. SHOCK Investigators. Should We Emergently Revascularize Occluded Coronaries for Cardiogenic Shock. N Engl J Med, 1999.PMID 10460813
18. [18]Thiele H, et al. PCI Strategies in Patients with Acute Myocardial Infarction and Cardiogenic Shock. N Engl J Med, 2017.PMID 29083953
19. [19]Thiele H, et al. One-Year Outcomes after PCI Strategies in Cardiogenic Shock. N Engl J Med, 2018.PMID 30145971
20. [20]Thiele H, et al. Intraaortic balloon support for myocardial infarction with cardiogenic shock. N Engl J Med, 2012.PMID 22920912
21. [21]Thiele H, et al. Extracorporeal Life Support in Infarct-Related Cardiogenic Shock. N Engl J Med, 2023.PMID 37634145
22. [22]Møller JE, et al. Microaxial Flow Pump or Standard Care in Infarct-Related Cardiogenic Shock. N Engl J Med, 2024.PMID 38587239
23. [23]Li Y, et al. Bivalirudin plus a high-dose infusion versus heparin monotherapy in patients with ST-segment elevation myocardial infarction undergoing primary percutaneous coronary intervention: a randomised trial. Lancet, 2022.PMID 36351459
24. [24]Montalescot G, et al. Intravenous enoxaparin or unfractionated heparin in primary percutaneous coronary intervention for ST-elevation myocardial infarction: the international randomised open-label ATOLL trial. Lancet, 2011.PMID 21856483
25. [25]Montalescot G, et al. Prehospital ticagrelor in ST-segment elevation myocardial infarction. N Engl J Med, 2014.PMID 25175921
26. [26]Schüpke S, et al. Ticagrelor or Prasugrel in Patients with Acute Coronary Syndromes. N Engl J Med, 2019.PMID 31475799
27. [27]Hofmann R, et al. Oxygen Therapy in Suspected Acute Myocardial Infarction. N Engl J Med, 2017.PMID 28844200
28. [28]Smith SW, et al. Diagnosis of ST-elevation myocardial infarction in the presence of left bundle branch block with the ST-elevation to S-wave ratio in a modified Sgarbossa rule. Ann Emerg Med, 2012.PMID 22939607
29. [29]Verouden NJ, et al. Persistent precordial "hyperacute" T-waves signify proximal left anterior descending artery occlusion. Heart, 2009.PMID 19620137
30. [30]Moye S, et al. The electrocardiogram in right ventricular myocardial infarction. Am J Emerg Med, 2005.PMID 16182990
31. [31]Elbadawi A, et al. Temporal Trends and Outcomes of Mechanical Complications in Patients With Acute Myocardial Infarction. JACC Cardiovasc Interv, 2019.PMID 31537282
32. [32]Damluji AA, et al. Mechanical Complications of Acute Myocardial Infarction: A Scientific Statement From the American Heart Association. Circulation, 2021.PMID 34126755
33. [33]Hohnloser SH, et al. Prophylactic use of an implantable cardioverter-defibrillator after acute myocardial infarction. N Engl J Med, 2004.PMID 15590950
34. [34]Steinbeck G, et al. Defibrillator implantation early after myocardial infarction. N Engl J Med, 2009.PMID 19812399
35. [35]Wiviott SD, et al. Prasugrel versus clopidogrel in patients with acute coronary syndromes. N Engl J Med, 2007.PMID 17982182
36. [36]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
37. [37]Mills NL, Newby LK, Zaman S, et al. Fifth Universal Definition of Myocardial Infarction (2026). Glob Heart, 2026.PMID 42666939