Chest pain evaluation

Overview and definitions

Chest pain is a symptom, so the evaluation starts with words. AHA/ACC 2021 calls chest pain acute when it is new onset or involves a change in pattern, intensity or duration compared with previous episodes in a patient with recurrent symptoms.^[1] It calls chest pain stable when symptoms are chronic and associated with consistent precipitants such as exertion or emotional stress.^[1]

Patients often report pressure, tightness, squeezing, heaviness or burning, and because they may not use the descriptor pain, AHA/ACC 2021 calls chest discomfort a more appropriate term.^[1] The discomfort may be felt somewhere other than the chest, including the shoulder, arm, neck, back, upper abdomen or jaw.^[1] ESC 2023 describes acute chest discomfort, which may be described as pain, pressure, tightness, heaviness or burning, as the leading presenting symptom prompting consideration of the clinical diagnosis of ACS and the initiation of testing aligned with specific diagnostic algorithms.^[2]

Why drop atypical? AHA/ACC 2021 explains that the term was meant to indicate angina without typical chest symptoms but is more often used to say the symptom is noncardiac, so it discourages the term.^[1] AHA/ACC 2021 recommends an initial assessment of chest pain to triage patients effectively on the basis of the likelihood that symptoms may be attributable to myocardial ischaemia (COR 1, LOE B-NR).^[1]

Classification: two clinical pathways

The AHA/ACC 2021 acute and stable definitions above lead to two different sets of recommendations, one for acute chest pain and one for stable chest pain.^[1] In suspected ACS, ESC 2023 says patients can be differentiated by the initial ECG into two working diagnoses.^[2]

^[1]

ESC 2023: working diagnoses from the initial ECG in suspected ACS

^[2]

ESC 2023 notes that patients initially assigned a working diagnosis of STEMI or NSTE-ACS may eventually receive a final non-ACS diagnosis, and that MI will not be the final diagnosis in all patients with a working diagnosis of STEMI.^[2] In NSTE-ACS, ESC 2023 says the majority who then show a typical rise and fall in cardiac troponin (fulfilling the fourth universal definition of MI) receive a final diagnosis of NSTEMI.^[2] ESC 2023 adds that in the others the troponin stays below the 99th centile and the final diagnosis is unstable angina (UA).^[2] It notes that with high-sensitivity assays this diagnosis has become less common.^[2] That ESC 2023 framing rests on the fourth universal definition of MI, which is now dated.^[2][9] The current definition is the Fifth Universal Definition of MI (UDMI), issued in 2026 on behalf of the joint ESC/ACC/AHA/WHF task force.^[9] Under the Fifth UDMI, unstable angina should be considered if serial cardiac troponin measurements exclude MI in patients with new ischaemic symptoms at rest or on minimal exertion (class 3 or 4 angina), or sudden intensification of chronic coronary syndrome symptoms.^[9] The Fifth UDMI agrees that the diagnosis has become progressively less common over time, because of high-sensitivity troponin assays and lower diagnostic thresholds for MI.^[9]

^{[1] [2]}

Epidemiology and risk factors

^{[1] [5]}

After injuries, chest pain is the second most common reason for adults to present to the ED in the United States (AHA/ACC 2021).^[1] Although there are several life-threatening causes, chest pain usually reflects a more benign condition, and its cause is often noncardiac.^[1] In patients older than 75 years, alternative diagnoses are still more common than a cardiac cause of chest pain at presentation.^[1]

Most patients who present to the ED with chest pain are women, particularly among those 65 years or older (AHA/ACC 2021).^[1] Increased age is a significant risk factor for ACS, but it is also a risk factor for comorbidities associated with alternative diagnoses, so AHA/ACC 2021 says a more extensive diagnostic workup is required in older patients.^[1]

The 2025 NHFA/CSANZ guideline summary describes marked inequities in both the management and outcomes of ACS in Australia, including:^[5]

• women experiencing longer symptom duration, under or delayed diagnosis and treatment, including lower rates of secondary prevention medication use;^[5]
• First Nations peoples experiencing coronary events at twice the rate of non-Indigenous Australians, exacerbated by ongoing inequities in healthcare access and outcomes;^[5]
• older adults, having a higher rate of ACS presentations, often complicated by atypical symptoms and age-related comorbidities;^[5]
• people living in regional and remote areas, with higher rates of missed diagnosis and poorer outcomes.^[5]

Pathophysiology

The 2025 ACC/AHA ACS guideline describes ACS as typically caused by disruption (rupture or erosion) of an unstable coronary atherosclerotic plaque.^[4] Associated partial or complete coronary thrombosis and/or microemboli diminish blood flow to the myocardium, with subsequent myocardial ischaemia.^[4] It places unstable angina, NSTEMI and STEMI along a continuum of severity.^[4]

Why is ischaemic pain hard to pin down? AHA/ACC 2021 explains that, like most visceral discomfort, it is characteristically deep, difficult to localise and usually diffuse.^[1] Point tenderness renders ischaemia less likely.^[1]

How does a stress test provoke ischaemia? ESC 2024 describes stress echocardiography as relying on inducing myocardial ischaemia by increasing myocardial oxygen demand beyond the myocardial blood supply.^[3] Because ischaemia starts in the subendocardium, which contributes to more than 50% of systolic myocardial wall thickening, stress testing will precipitate wall-thickening abnormalities in the perfusion territory of narrowed coronary arteries.^[3]

Troponin: what the number means

AHA/ACC 2021 states that cTn is organ-specific but not disease-specific: numerous ischaemic, noncoronary cardiac and noncardiac causes of cardiomyocyte injury can raise it.^[1] AHA/ACC 2021 adds that a cTn concentration above the 99th percentile upper reference limit, which is assay-dependent, indicates myocardial injury.^[1] In the same guideline, detection of injury possibly indicative of AMI depends on a rise or fall of the biomarker in blood.^[1] The current Fifth UDMI (2026) specifies sex-specific thresholds: sex-specific 99th percentile upper reference limits for cTn define myocardial injury.^[9] This avoids systematic bias and the under-recognition of both MI and other cardiac conditions associated with myocardial injury in female patients.^[9]

Timing explains the serial sampling. ESC 2023 states that in MI, cTn levels rise rapidly (usually within 1 h with high-sensitivity assays) after symptom onset and remain elevated for a variable period, usually several days.^[2] High-sensitivity assays can detect cTn in the blood of most healthy individuals, with different sex-specific thresholds (AHA/ACC 2021).^[1]

ESC 2023 defines assays by how many healthy people they detect: sensitive assays detect cTn in about 20–50% of healthy individuals and high-sensitivity assays in about 50–95%.^[2] It asks clinicians to describe hs-cTn levels below and above the 99th percentile as non-elevated and elevated, avoiding the terms normal and abnormal.^[2]

Despite these baseline differences, absolute changes in hs-cTn are still of diagnostic and prognostic value (ESC 2023).^[2] ESC 2023 states that data on sex-specific hs-cTn values in the diagnosis of MI have been controversial and failed to show a clear clinical benefit.^[2] Until automated tools incorporating all four variables are available, it keeps uniform cut-off concentrations as the standard of care for the early diagnosis of MI.^[2]

^{[1] [2] [9]}

Clinical presentation

AHA/ACC 2021 describes angina as retrosternal discomfort that builds gradually in intensity over several minutes.^[1] It is usually precipitated by physical or emotional stress, or occurs at rest as in ACS.^[1] It has characteristic radiation (e.g. left arm, neck, jaw) and associated symptoms (e.g. dyspnoea, nausea, lightheadedness).^[1] Chest pain more likely associated with ischaemia is substernal discomfort provoked by exertion or emotional stress and relieved by rest or nitroglycerin.^[1]

AHA/ACC 2021 Table 3: chest pain characteristics and corresponding causes (all rows)

^[1]

What about the response to nitroglycerin? AHA/ACC 2021 says relief with nitroglycerin should not be used as a diagnostic criterion, especially because other entities such as oesophageal spasm respond comparably.^[1] In suspected chronic coronary syndrome, ESC 2024 states that rapid relief of chest discomfort within 1 or 2 min after sublingual nitroglycerine increases the likelihood of CCS.^[3]

ESC 2023 says chest pain-equivalent symptoms include dyspnoea, epigastric pain, and pain in the left or right arm or neck/jaw.^[2] In suspected chronic coronary syndrome, ESC 2024 states that symptoms such as chest pain triggered by emotional stress; dyspnoea or dizziness on exertion; pain in the arms, jaw, neck or upper back; or fatigue should be considered as potential angina equivalents (Class IIa, Level B).^[3]

Presentations that are easy to miss

• Women. AHA/ACC 2021: women who present with chest pain are at risk for underdiagnosis, and potential cardiac causes should always be considered (COR 1, LOE B-NR). In women presenting with chest pain, a history that emphasises accompanying symptoms more common in women with ACS is recommended (COR 1, LOE B-NR).^[1]
• Women may present with accompanying symptoms (e.g. nausea, fatigue and shortness of breath) more often than men, and traditional risk score tools and physician assessments often underestimate their risk and misclassify them as having nonischaemic chest pain (AHA/ACC 2021).^[1]
• Older patients. AHA/ACC 2021: in patients with chest pain who are older than 75 years, ACS should be considered when accompanying symptoms such as shortness of breath, syncope or acute delirium are present, or when an unexplained fall has occurred (COR 1, LOE C-LD).^[1]
• Diabetes, women and elderly patients. Symptoms on the left or right side of the chest, stabbing, sharp pain, or discomfort in the throat or abdomen may occur in these groups (AHA/ACC 2021 Table 3).^[1]

Differential diagnosis

For patients presenting to the ED with nontraumatic chest pain, AHA/ACC 2021 sets two priorities.^[1] First, rapid initiation of optimal management in life-threatening conditions such as ACS, aortic dissection and pulmonary embolism (PE), as well as nonvascular syndromes (e.g. oesophageal rupture, tension pneumothorax).^[1] Second, deliberate therapy for those with less critical illness.^[1]

ESC 2023 asks that several cardiac and non-cardiac conditions that may mimic ACS be considered in the differential diagnosis of acute chest pain as part of the clinical assessment.^[2] The ECG itself may point elsewhere: AHA/ACC 2021 notes that an ECG may identify other nonischaemic causes of chest pain (e.g. pericarditis, myocarditis, arrhythmia, electrolyte abnormalities, paced rhythm, hypertrophic cardiomyopathy, pulmonary hypertension, congenital long QT or a normal variant).^[1]

Examination clues by syndrome: AHA/ACC 2021 Table 4 (all rows)

^[1]

When the heart is not the cause, AHA/ACC 2021 describes the differential diagnosis for noncardiac causes of acute chest pain as broad: respiratory, musculoskeletal, gastrointestinal, psychological and other causes.^[1] Of these noncardiac causes of acute chest pain, musculoskeletal causes are the most common.^[1] Respiratory causes are less frequent but potentially more serious and include PE, pneumonia and pneumothorax.^[1]

AHA/ACC 2021 Table 9: differential diagnosis of noncardiac chest pain (all rows)

^[1]

AHA/ACC 2021 recommends evaluating patients with acute chest pain for noncardiac causes if they have persistent or recurring symptoms despite a negative stress test or anatomic cardiac evaluation, or a low-risk designation by a CDP (COR 1, LOE C-EO).^[1]

Clinical and bedside assessment

• History (AHA/ACC 2021): in patients with chest pain, a focused history that includes characteristics and duration of symptoms relative to presentation as well as associated features, and cardiovascular risk factor assessment, should be obtained (COR 1, LOE C-LD).^[1]
• History (ESC 2024, suspected chronic coronary syndrome): in individuals reporting symptoms of suspected myocardial ischaemic origin, a detailed assessment of cardiovascular risk factors, medical history and symptom characteristics (including onset, duration, type, location, triggers, relieving factors, time of day) is recommended (Class I, Level C).^[3]
• Examination (AHA/ACC 2021): in patients presenting with chest pain, a focused cardiovascular examination should be performed initially to aid in the diagnosis of ACS or other potentially serious causes of chest pain (e.g. aortic dissection, PE or oesophageal rupture) and to identify complications (COR 1, LOE C-EO).^[1]
• Examination (ESC 2023, suspected ACS): focused physical examination should include checking for the presence of all major pulses, measurement of blood pressure in both arms, auscultation of the heart and lungs, and assessing for signs of heart failure or circulatory compromise.^[2]

AHA/ACC 2021 stresses that the chest pain characteristics should be taken directly from the patient for optimal interpretation.^[1] In patients with suspected ACS, ESC 2023 recommends prompt assessment of vital signs at first medical contact, at the same time as acquisition of an initial ECG.^[2]

With an uncomplicated AMI the examination may be negative (AHA/ACC 2021).^[1] Chest tenderness on palpation or pain with inspiration markedly reduces the probability of ACS.^[1] Sudden severe chest or back pain with a limb pulse differential suggests aortic dissection, but the sensitivity of the pulse differential alone was only 30%.^[1]

^[2]

Investigations

The ECG

The resting 12-lead ECG is the first-line diagnostic tool in suspected ACS (ESC 2023).^[2] The rows below set the timing and the response to a nondiagnostic first trace.

• AHA/ACC 2021: in all patients who present with acute chest pain regardless of the setting, an ECG should be acquired and reviewed for STEMI within 10 minutes of arrival (COR 1, LOE C-LD).^[1]
• AHA/ACC 2021: unless a noncardiac cause is evident, an ECG should be performed for patients seen in the office setting with stable chest pain; if an ECG is unavailable the patient should be referred to the ED so one can be obtained (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: in patients with chest pain in which an initial ECG is nondiagnostic, serial ECGs to detect potential ischaemic changes should be performed, especially when clinical suspicion of ACS is high, symptoms are persistent, or the clinical condition deteriorates (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021: patients with chest pain in whom the initial ECG is consistent with an ACS should be treated according to STEMI and NSTE-ACS guidelines (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021: in patients with chest pain and intermediate-to-high clinical suspicion for ACS in whom the initial ECG is nondiagnostic, supplemental leads V7 to V9 are reasonable to rule out posterior MI (COR 2a, LOE B-NR).^[1]
• ESC 2023, suspected ACS: twelve-lead ECG recording and interpretation is recommended as soon as possible at the point of FMC, with a target of under 10 min (Class I, Level B).^[2]
• ESC 2023: additional ECG leads (V3R, V4R and V7–V9) are recommended in cases of inferior STEMI or if total vessel occlusion is suspected and standard leads are inconclusive (Class I, Level B).^[2]
• ESC 2023, suspected ACS: an additional 12-lead ECG is recommended in cases with recurrent symptoms or diagnostic uncertainty (Class I, Level C).^[2]
• ESC 2023: continuous ECG monitoring and the availability of defibrillator capacity is recommended as soon as possible in all patients with suspected STEMI, in suspected ACS with other ECG changes or ongoing chest pain, and once the diagnosis of MI is made (Class I, Level B).^[2]
• ACC/AHA 2025 (ACS guideline): in patients with suspected ACS, acquisition and interpretation of an ECG within 10 minutes is recommended to help guide patient management (COR 1, LOE B-NR); in those whose initial ECG is nondiagnostic, serial 12-lead ECGs should be performed to detect potential ischaemic changes, especially when clinical suspicion of ACS is high, symptoms are persistent, or clinical condition deteriorates (COR 1, LOE B-NR). The table marks both rows as adapted from the 2021 chest pain guideline.^[4]
• ESC 2024 (suspected chronic coronary syndrome): a resting 12-lead ECG is recommended in all individuals reporting chest pain (unless an obvious non-cardiac cause is identified), particularly during, or immediately after, an episode suggestive of myocardial ischaemia (Class I, Level C).^[3]

AHA/ACC 2021 states that an initial normal ECG does not exclude ACS, and a patient with an initial normal ECG should have a repeat ECG, if symptoms are ongoing, until other testing rules out ACS.^[1] Up to 6% of patients with evolving ACS are discharged from the ED with a normal ECG.^[1] ESC 2023 adds that the ECG in NSTE-ACS may be normal in more than one-third of patients.^[2]

AHA/ACC 2021: a nondiagnostic ECG should be compared with previous ECGs, if available.^[1] Left ventricular hypertrophy, bundle branch blocks and ventricular pacing may mask signs of ischaemia or injury.^[1] A normal ECG may accompany left circumflex or right coronary artery occlusions and posterior wall ischaemia, which is often electrically silent.^[1]

ESC 2023 treats ST-segment depression in V1–V3 (especially with a positive terminal T wave) and/or ST-segment elevation in V7–V9 as highly suggestive of posterior coronary occlusion, often of the left circumflex artery.^[2] In patients with a high clinical suspicion of ongoing myocardial ischaemia, LBBB, RBBB or a paced rhythm precludes accurate assessment of ST-segment elevation.^[2] Such patients with signs or symptoms highly suspicious for ongoing ischaemia are managed like those with clear ST-segment elevation, regardless of whether the bundle branch block was previously known.^[2]

In suspected chronic coronary syndrome, ESC 2024 does not recommend using ST-segment deviations during supraventricular tachyarrhythmias, particularly during re-entrant atrioventricular tachycardias, per se, as reliable evidence of obstructive CAD (Class III, Level B).^[3]

Chest radiograph

AHA/ACC 2021: in patients presenting with acute chest pain, a chest radiograph is useful to evaluate for other potential cardiac, pulmonary and thoracic causes of symptoms (COR 1, LOE C-EO).^[1] It may show a widened mediastinum in aortic dissection, but it is not sensitive enough in this setting to rule out the diagnosis.^[1]

In suspected chronic coronary syndrome, ESC 2024 states that the chest X-ray does not yield specific information for accurate diagnosis or risk stratification.^[3] In suspected chronic coronary syndrome, it says a chest X-ray should be considered for individuals with signs and symptoms suggestive of heart failure, suspected acute pulmonary disease, or suspected aortic, non-coronary cardiac or other thoracic causes of chest pain (Class IIa, Level C).^[3]

Cardiac troponin

• AHA/ACC 2021: in all patients presenting to the ED with acute chest pain and suspected ACS, cTn should be measured as soon as possible after presentation (COR 1, LOE C-LD).^[1]
• AHA/ACC 2021: in patients presenting with acute chest pain, serial cTn I or T levels are useful to identify abnormal values and a rising or falling pattern indicative of acute myocardial injury (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: in patients presenting with acute chest pain, high-sensitivity cTn is the preferred biomarker because it enables more rapid detection or exclusion of myocardial injury and increases diagnostic accuracy (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: clinicians should be familiar with the analytical performance and the 99th percentile upper reference limit that defines myocardial injury for the cTn assay used at their institution (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021: with availability of cTn, CK-MB isoenzyme and myoglobin are not useful for diagnosis of acute myocardial injury (COR 3: No Benefit, LOE B-NR).^[1]
• ESC 2023, suspected ACS: it is recommended to measure cardiac troponins with high-sensitivity assays immediately after presentation and to obtain the results within 60 min of blood sampling (Class I, Level B).^[2]
• ESC 2023: it is recommended to base the diagnosis and initial short-term risk stratification of ACS on a combination of clinical history, symptoms, vital signs, other physical findings, ECG and hs-cTn (Class I, Level B).^[2]
• ACC/AHA 2025 (ACS guideline): in patients with suspected ACS, cTn should be measured as soon as possible, preferably using a high-sensitivity cTn (hs-cTn) assay (COR 1, LOE B-NR). The table marks this row as adapted from the 2021 chest pain guideline.^[4]
• NHFA/CSANZ 2025 (summary): in people with suspected ACS, evaluation with hs-cTn assays is recommended (strong recommendation, high certainty of evidence).^[5]

ESC 2023 recommends high-sensitivity assays over lower-sensitivity assays because they provide higher diagnostic accuracy at an identical low cost.^[2] The majority of currently used point-of-care (POC) tests cannot be considered high-sensitivity assays; their shorter turnaround is counterbalanced by lower sensitivity, lower diagnostic accuracy and lower negative predictive value (NPV).^[2] Automated assays have been more thoroughly evaluated than POC tests and are currently preferred.^[2]

AHA/ACC 2021 notes that the level of detection, 99th percentile upper reference limit, analytical precision and criteria for a significant delta are assay-specific, even among manufacturers of the same analyte (e.g. hs-cTnI).^[1] The coefficient of variation at the 99th percentile upper reference limit for each assay should be 10% or less.^[1] ESC 2023 does not recommend biomarkers other than cTn for the diagnosis of ACS (unless cTn is not available).^[2]

AHA/ACC 2021 states that multiple other cardiovascular biomarkers, including some in common clinical use such as natriuretic peptides, have been associated with the risk of adverse cardiovascular outcomes in patients with chest pain.^[1] None has sufficient diagnostic accuracy for myocardial injury to be recommended for that purpose.^[1]

Troponin thresholds: how the three bodies differ

^{[2] [5] [1] [4]}

The current Fifth UDMI (2026) separates two uses of these thresholds.^[9] It notes that the thresholds used to triage patients in accelerated diagnostic pathways, such as the ESC 0/1- and 0/2-hour pathways, are often not based on the 99th percentile.^[9] They are uniform rather than sex-specific and are used for risk stratification.^[9] A final diagnosis of MI rests instead on a rise and/or fall in cTn with at least one value above the 99th percentile URL, where sex-specific thresholds apply.^[9]

High-sensitivity troponin pathways and thresholds

Because of their higher sensitivity and diagnostic accuracy for detecting MI at presentation, ESC 2023 says the time interval to the second cTn assessment can be shortened with hs-cTn assays.^[2] It recommends the 0 h/1 h algorithm as the best option and the 0 h/2 h algorithm as the second-best option.^[2]

• ESC 2023: it is recommended to use an ESC algorithmic approach with serial hs-cTn measurements (0 h/1 h or 0 h/2 h) to rule in and rule out NSTEMI (Class I, Level B).^[2]
• ESC 2023: additional testing after 3 h is recommended if the first two hs-cTn measurements of the 0 h/1 h algorithm are inconclusive and no alternative diagnoses explaining the condition have been made (Class I, Level B).^[2]
• AHA/ACC 2021: in the evaluation of patients presenting with acute chest pain and suspected ACS for whom serial troponins are indicated to exclude myocardial injury, the recommended time intervals after the initial sample (time zero) for repeat measurements are 1 to 3 hours for high-sensitivity troponin and 3 to 6 hours for conventional troponin assays (COR 1, LOE B-NR).^[1]
• ACC/AHA 2025 (ACS guideline, newer): in patients with suspected ACS with an initial hs-cTn or cTn that is nondiagnostic, the recommended time intervals for repeat measurements after the initial sample (time zero) are 1 to 2 hours for hs-cTn and 3 to 6 hours for conventional cTn assays (COR 1, LOE B-NR). The table marks this row as adapted from the 2021 chest pain guideline.^[4]
• AHA/ACC 2021: for patients with acute chest pain, a normal ECG, and symptoms suggestive of ACS that began at least 3 hours before ED arrival, a single hs-cTn concentration below the limit of detection on initial measurement (time zero) is reasonable to exclude myocardial injury (COR 2a, LOE B-NR).^[1]
• AHA/ACC 2021: to standardise the detection and differentiation of myocardial injury in patients presenting with acute chest pain and suspected ACS, institutions should implement a CDP that includes a protocol for troponin sampling based on their particular assay (COR 1, LOE C-LD).^[1]
• NHFA/CSANZ 2025 (summary): a high-sensitivity troponin-based CDP is recommended, using the 0/1- or 0/2-h strategy, or the High-sensitivity troponin in the evaluation of patients with acute coronary syndrome (High-STEACS) algorithm (consensus).^[5]

How were the ESC cut-offs chosen? Rule-out thresholds were selected to allow a sensitivity and NPV of at least 99%, and rule-in thresholds to allow a positive predictive value (PPV) of at least 70%.^[2] The algorithms rest on two concepts: hs-cTn is a continuous variable, with MI probability rising as values rise, and early absolute changes within 1 h or 2 h act as surrogates for changes over 3 h or 6 h.^[2] The cut-off concentrations within the 0 h/1 h and 0 h/2 h algorithms are assay specific (ESC 2023).^[2]

ESC 2023 Figure 6: 0 h/1 h or 0 h/2 h algorithm in patients presenting to the ED with suspected NSTEMI and without an indication for immediate invasive angiography

^[2]

ESC 2023 states that the 0 h/1 h and 0 h/2 h algorithms should always be integrated with a detailed clinical assessment and a 12-lead ECG, and repeat blood sampling is mandatory in cases where there is ongoing or recurrent chest pain.^[2] To make the 0 h/1 h algorithm safe and feasible, blood for hs-cTn at 0 h and 1 h should be taken irrespective of other clinical details and pending results, and the 0 h sample immediately after admission to the ED.^[2]

In the observe zone, most patients with a high degree of clinical suspicion of ACS (e.g. a relevant increase in cTn from presentation to 3 h) are candidates for ICA (ESC 2023).^[2] Most patients with a low to intermediate likelihood of ACS by clinical judgement are candidates for non-invasive imaging after transfer from the ED to the ward.^[2] If an alternative condition explains the cTn values in the appropriate clinical context (i.e. rapid ventricular rate response to atrial fibrillation [AF], marked anaemia, or a hypertensive emergency), further diagnostic testing (i.e. ICA) may not be required.^[2]

ESC 2023 says its 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.^[2] ESC 2023 explains that three recent large diagnostic studies suggested it appears to balance efficacy and safety less well than more rapid protocols using lower rule-out concentrations, including the 0 h/1 h algorithm.^[2]

The single-sample rule depends on timing. AHA/ACC 2021 limits CDPs that rule out on a single hs-cTn below the limit of detection to patients whose symptoms started at least 3 hours before ED arrival.^[1] Unlike high-sensitivity assays, clinical decision-making based on a single conventional cTn measurement has not been validated.^[1] The 2025 ACC/AHA ACS guideline likewise says that when a CDP incorporates a single hs-cTn value, the troponin should be obtained at least 3 hours after symptom onset.^[4]

The 2025 NHFA/CSANZ summary adds that in most hospitals standard laboratory turnaround times make the 0/1-h strategy impractical, so a 0/2-h strategy is currently the most pragmatic option in most settings.^[5] Its Figure 4 shows the 0/2-h testing recommendations (with time frames changed accordingly if a 0/1-h strategy is used), and its footnote states that all people with symptom onset less than 2 h before need serial testing.^[5]

^{[2] [1] [4] [5] [2]}

Risk scores and clinical decision pathways

AHA/ACC 2021 calls these structured chest pain protocols, whatever their local name, clinical decision pathways (CDPs).^[1] Compared with unstructured assessment, CDPs have been shown to decrease unnecessary testing and reduce admissions while keeping high sensitivity for acute myocardial injury and 30-day MACE.^[1] Chest pain risk scores combine clinical information such as age, ST changes, symptoms, CAD risk factors and cTn to estimate the probability of ACS or the risk of 30-day major adverse cardiovascular events (MACE).^[1]

• AHA/ACC 2021: in patients presenting with acute chest pain and suspected ACS, CDPs should categorise patients into low-, intermediate- and high-risk strata to facilitate disposition and subsequent diagnostic evaluation (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: in patients with acute chest pain and suspected ACS, previous testing when available should be considered and incorporated into CDPs (COR 1, LOE C-LD).^[1]
• ESC 2023, suspected ACS: the use of established risk scores (e.g. GRACE risk score) for prognosis estimation should be considered (Class IIa, Level B).^[2]
• NHFA/CSANZ 2025 (summary): people presenting with acute chest pain or other symptoms suggestive of ACS should receive care guided by an evidence-based CDP that includes assay-specific troponin results to categorise people as high, intermediate or low risk (consensus); when contemporary troponin assays are used, a CDP incorporating formal clinical score-based risk stratification is recommended (consensus).^[5]

AHA/ACC 2021 states that risk scores are essential when conventional cTn assays are used, and that the hs-cTn result may be more predictive than the other clinical components of the score.^[1] CDPs that include risk scores perform with 99% to 100% sensitivity for index-visit AMI and 30-day MACE, but because sex-specific considerations are not in all scores, their effectiveness in men and women may not be equal.^[1]

AHA/ACC 2021 Table 8: definitions used for low-risk patients with chest pain (all rows)

^[1]

HEART stands for history, ECG, age, risk factors and troponin.^[1] In the AHA/ACC 2021 sample pathways (Table 6), the HEART Pathway intermediate-risk row is a HEART score of 4–6, and its high-risk row a HEART score of 7–10.^[1] Table 8 defines HEART Pathway low risk as a HEART score of 3 or less with initial and serial cTn/hs-cTn below the assay 99th percentile.^[1]

AHA/ACC 2021 says the warranty period of prior cardiac testing should be considered when symptoms are unchanged: 2 years for a normal coronary angiogram or a CCTA with no stenosis or plaque, and 1 year for a normal stress test given adequate stress (Table 7).^[1] In patients with acute chest pain once ACS has been ruled out, previous test results should always be considered.^[1] Among them, in those with recent cardiac testing and normal findings who do not have biomarker evidence of acute myocardial injury, further testing is of limited value, with provisos.^[1] The provisos are that adequate exercise levels were achieved or pharmacologic stress was performed, imaging was of sufficient quality, and there are no changes in symptom frequency or stability at the new visit.^[1]

ACC/AHA 2025 Table 5 (selected columns): risk tools in established ACS

^[4]

The 2025 ACC/AHA ACS guideline presents these scores as tools to help assess short- and long-term risk in established ACS.^[4] It says the GRACE and TIMI risk scores for NSTE-ACS and STEMI are well validated and may be useful for helping to guide some therapeutic decisions.^[4]

Imaging in acute chest pain

• ESC 2023: emergency transthoracic echocardiography (TTE) is recommended in patients with suspected ACS presenting with cardiogenic shock or suspected mechanical complications (Class I, Level C).^[2]
• ESC 2023, suspected ACS: emergency TTE should be considered at triage in cases of diagnostic uncertainty, but this should not result in delays in transfer to the cardiac catheterisation laboratory if there is suspicion of an acute coronary artery occlusion (Class IIa, Level C).^[2]
• ESC 2023: in patients with suspected ACS, non-elevated (or uncertain) hs-cTn levels, no ECG changes and no recurrence of pain, incorporating CCTA or a non-invasive stress imaging test as part of the initial workup should be considered (Class IIa, Level A).^[2]
• ESC 2023: routine, early CCTA in patients with suspected ACS is not recommended (Class III, Level B).^[2]

ESC 2023 explains the last row. A default approach using CCTA as the first-line imaging investigation in suspected NSTE-ACS is not recommended.^[2] CCTA may add value in certain settings (i.e. the observe zone when cTn and ECG results remain inconclusive), and a normal CCTA ruling out both obstructive and non-obstructive plaque has a high NPV to exclude ACS with excellent clinical outcomes.^[2] Its utility may be limited by tachycardia, established coronary artery disease, previous stents or extensive coronary calcification.^[2]

On clinical presentation, CT is often the diagnostic tool of choice for ruling out alternative potentially life-threatening differential diagnoses of ACS, like PE or aortic dissection (ESC 2023).^[2] This should be an ECG-gated contrast CT angiogram with full coverage of the thoracic aorta and the proximal head and neck vessels.^[2] Generally, CT has no role in suspected ongoing acute coronary occlusion, for whom emergency ICA is the priority.^[2]

^{[2] [8]}

ESC 2023 summarises RAPID-CTCA as showing that a default approach using early non-invasive CCTA in suspected NSTE-ACS did not improve clinical outcomes at 1 year and was associated with a modest increase in the duration and cost of the hospital stay.^[2] The trial authors concluded that their findings do not support routine early CT coronary angiography in intermediate-risk patients with acute chest pain and suspected acute coronary syndrome.^[8]

Immediate management: the first minutes

The goals in acute chest pain, in the ED or the office, are to identify life-threatening causes, determine clinical stability, and assess the need for hospitalisation versus the safety of outpatient evaluation and management (AHA/ACC 2021).^[1]

• ESC 2023: patients with suspected STEMI should be immediately triaged for an emergency reperfusion strategy (Class I, Level A).^[2]
• ESC 2023: hs-cTn results are not required for the initial stratification of ACS, and the initial emergency management (i.e. for a working diagnosis of STEMI or very high-risk NSTE-ACS) should not be delayed based on them. The 2025 ACC/AHA ACS guideline likewise states that, with electrocardiographic evidence of STEMI, reperfusion therapy should not be delayed pending biomarker results.^[2][4]
• NHFA/CSANZ 2025 (summary): people with symptoms and ECG changes consistent with acute coronary occlusion MI (ACOMI) require urgent reperfusion, and a CDP should not be used (strong recommendation, very low certainty).^[5]
• NHFA/CSANZ 2025 (summary): in people with ongoing ischaemic symptoms or haemodynamic compromise or new ischaemic findings on ECG, continuous cardiac monitoring and defibrillator availability is recommended while assessment for ACOMI continues (strong recommendation, low certainty).^[5]
• AHA/ACC 2021: in patients with acute chest pain, it is recommended that 9-1-1 be activated by patients or bystanders to initiate transport to the closest ED by emergency medical services (EMS) (COR 1, LOE C-LD).^[1]
• AHA/ACC 2021: patients with clinical evidence of ACS or other life-threatening causes of acute chest pain seen in the office setting should be transported urgently to the ED, ideally by EMS (COR 1, LOE C-LD).^[1]
• AHA/ACC 2021: for patients with acute chest pain and suspected ACS initially evaluated in the office setting, delayed transfer to the ED for cTn or other diagnostic testing should be avoided (COR 3: Harm, LOE C-LD).^[1]
• NHFA/CSANZ 2025 (summary): for people with suspected ACS initially evaluated in the primary care setting, prompt transfer to a facility where definitive risk assessment can occur (e.g. ED) is recommended (consensus).^[5]

Why EMS? AHA/ACC 2021 recommends transfer by EMS from the office setting for acute chest pain with suspected ACS or other life-threatening conditions.^[1] It cites the advantages EMS provides, including a prehospital ECG that can facilitate reperfusion if ST elevation is present, trained personnel who can treat chest pain and arrhythmias and defibrillate en route, and shorter travel time to the ED.^[1] It also reports that 1 in 300 patients with chest pain transported to the ED by private vehicle suffers a cardiac arrest en route.^[1]

AHA/ACC 2021 recommends that patients with chest pain whose initial ECG is consistent with an ACS be treated according to STEMI and NSTE-ACS guidelines (COR 1, LOE C-EO).^[1] Those treatments are covered in the STEMI and NSTE-ACS topics.

Acute chest pain: the risk-stratified pathway

Once STEMI has been excluded, AHA/ACC 2021 stratifies suspected ACS into low-risk versus intermediate- or high-risk groups to guide subsequent management.^[1] AHA/ACC 2021 adds that although most high-risk patients identified by CDPs should undergo cardiac catheterisation, these patients still require a clinical assessment to determine if invasive evaluation is appropriate.^[1]

Low risk

• AHA/ACC 2021: patients with acute chest pain and a 30-day risk of death or MACE below 1% should be designated as low risk (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: in patients with acute chest pain and suspected ACS who are deemed low risk (below 1% 30-day risk of death or MACE), it is reasonable to discharge home without admission or urgent cardiac testing (COR 2a, LOE B-R).^[1]
• AHA/ACC 2021: for patients with acute chest pain and suspected ACS who are deemed low risk by a CDP, patient decision aids are beneficial to improve understanding and effectively facilitate risk communication (COR 1, LOE B-R).^[1]
• NHFA/CSANZ 2025 (summary): in people at low risk who remain symptom-free, further cardiac testing for CAD is not routinely required (consensus).^[5]

Every method for declaring low risk, says AHA/ACC 2021, involves an appropriate history and examination and an ECG that is normal, nonischaemic or unchanged from the previous ECG.^[1] It also needs cTn measured at a single point in time (if presentation is more than 3 hours from symptom onset and a high-sensitivity assay is used) or serially, with a chest pain risk score in the CDP if a conventional assay is used.^[1] With conventional cTn assays this requires risk scores in the CDP, but hs-cTn results can be used on their own.^[1] This approach was validated in 15 studies including more than 9,600 patients, with a negative predictive value for MI or death at 30 days of 99.8%.^[1] These findings reflect studies of hs-cTnI and hs-cTnT using serial measurement algorithms or a single hs-cTn, provided the final measurement was performed at least 3 hours after symptom onset, without risk scores.^[1]

For this low-risk subset there is no evidence that stress testing or cardiac imaging within 30 days of the index ED visit improves outcomes; AHA/ACC 2021 notes this is a change from previous guidelines that broadly recommended stress testing within 72 hours.^[1] Many of these patients have baseline cardiac risk factors that need managing, so pathways to facilitate outpatient follow-up for further evaluation and guideline-directed management of cardiac risk factors should be considered.^[1]

Intermediate risk

AHA/ACC 2021 places ED patients without high-risk features who are not classified as low risk by a CDP in the intermediate-risk group.^[1] They have no troponin evidence of acute myocardial injury but remain candidates for additional cardiac testing, and some may have chronic or minor troponin elevations.^[1]

• AHA/ACC 2021: for intermediate-risk patients with acute chest pain, TTE is recommended as a rapid, bedside test to establish baseline ventricular and valvular function, evaluate for wall motion abnormalities, and to assess for pericardial effusion (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021: for intermediate-risk patients with acute chest pain, management in an observation unit is reasonable to shorten length of stay and lower cost relative to an inpatient admission (COR 2a, LOE A).^[1]
• AHA/ACC 2021, no known CAD: for intermediate-risk patients with acute chest pain and no known CAD eligible for diagnostic testing after a negative or inconclusive evaluation for ACS, CCTA is useful for exclusion of atherosclerotic plaque and obstructive CAD (COR 1, LOE A).^[1]
• AHA/ACC 2021, no known CAD: for intermediate-risk patients with acute chest pain and no known CAD who are eligible for cardiac testing, either exercise ECG, stress echocardiography, stress PET/SPECT MPI or stress CMR is useful for the diagnosis of myocardial ischaemia (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: for intermediate-risk patients with acute chest pain, moderate-severe ischaemia on current or prior (1 year or less) stress testing, and no known CAD established by prior anatomic testing, ICA is recommended (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021, no known CAD: for intermediate-risk patients with acute chest pain and no known CAD, with a coronary artery stenosis of 40% to 90% in a proximal or middle coronary artery on CCTA, FFR-CT (fractional flow reserve with CT) can be useful for the diagnosis of vessel-specific ischaemia and to guide decision-making regarding the use of coronary revascularisation (COR 2a, LOE B-NR).^[1]
• AHA/ACC 2021, known CAD: for intermediate-risk patients with acute chest pain who have known CAD and present with new onset or worsening symptoms, guideline-directed medical therapy (GDMT) should be optimised before additional cardiac testing is performed (COR 1, LOE A).^[1]
• AHA/ACC 2021, known CAD: for intermediate-risk patients with acute chest pain who have worsening frequency of symptoms with significant left main, proximal left anterior descending stenosis, or multivessel CAD on prior anatomic testing or history of prior coronary revascularisation, ICA is recommended (COR 1, LOE A).^[1]
• AHA/ACC 2021: for patients with acute chest pain and suspected ACS who are deemed intermediate risk by a CDP, shared decision-making between the clinician and patient regarding the need for admission, for observation, discharge, or further evaluation in an outpatient setting is recommended for improving patient understanding and reducing low-value testing (COR 1, LOE B-R).^[1]

AHA/ACC 2021 widened the meaning of known CAD to include patients with prior anatomic testing (invasive angiography or CCTA) that identified nonobstructive atherosclerotic plaque as well as obstructive CAD.^[1]

The 2025 NHFA/CSANZ summary splits the intermediate group by troponin.^[5] In people at intermediate risk (as defined by a validated CDP) with elevated troponin concentrations (above the 99th percentile), inpatient investigation is recommended (strong recommendation, moderate certainty).^[5] The summary cites a relatively high 30-day cardiac event rate (2%–22%) for people at intermediate risk with hs-cTn above the sex-specific 99th percentile.^[5] If non-invasive testing is selected in these people, it supports CTCA as a first-line investigation for those without previously known CAD presenting with intermediate-risk ACS, if no contraindications exist.^[5] In people at intermediate risk without elevated troponin concentrations, the summary says to consider outpatient investigation with non-invasive testing (consensus).^[5] The summary adds that outpatient non-invasive testing can be considered (ideally within 30 days) for people at intermediate risk with serial troponin values at or below the 99th percentile, as the 30-day event rate is lower (below 2%).^[5]

High risk

• AHA/ACC 2021 (recommendations for high-risk patients, including those with high-risk findings on CCTA or stress testing): patients with acute chest pain and suspected ACS who have new ischaemic changes on electrocardiography, troponin-confirmed acute myocardial injury, new-onset left ventricular systolic dysfunction (ejection fraction below 40%), newly diagnosed moderate-severe ischaemia on stress testing, haemodynamic instability, and/or a high CDP risk score should be designated as high risk for short-term MACE (COR 1, LOE B-NR).^[1]
• AHA/ACC 2021: for patients with acute chest pain and suspected ACS who are designated as high risk, ICA is recommended (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021: for high-risk patients with acute chest pain who are troponin positive in whom obstructive CAD has been excluded by CCTA or ICA, CMR or echocardiography can be effective in establishing alternative diagnoses (COR 2a, LOE B-NR).^[1]
• NHFA/CSANZ 2025 (summary): people stratified as high risk of MACE using hs-cTn-based CDPs should be admitted for further evaluation.^[5]

AHA/ACC 2021 reports that about 6% to 15% of troponin-positive ACS occurs without obstructive CAD.^[1] ESC 2023 describes CMR as of particular clinical value in establishing a diagnosis of AMI where there is diagnostic uncertainty.^[2]

^[1]

Stable chest pain: likelihood first, then the test

AHA/ACC 2021 describes stable chest pain as a symptom of myocardial ischaemia characterised by chest pain provoked with physical or emotional stress.^[1] In suspected chronic coronary syndrome (CCS), ESC 2024 bases the diagnosis on interpreting the symptoms, balancing age, sex, risk factors and comorbidities, and choosing the most appropriate test.^[3]

ESC 2024: estimating the likelihood of obstructive CAD

The older Diamond–Forrester tables used sex, age and angina symptoms but needed repeated updating because the prevalence of obstructive CAD at invasive angiography has declined in contemporary Western cohorts (ESC 2024).^[3] The Risk Factor-weighted Clinical Likelihood (RF-CL) model includes sex, age, angina symptoms and the number of risk factors, without losing diagnostic accuracy compared with more advanced models requiring computed calculation.^[3] Compared with the ESC pretest probability model, RF-CL triples the share classed as very low (5% or less) likelihood, 38% versus 12%.^[3]

In the RF-CL figure, family history means one or more first-degree relatives with early signs of CAD (men under 55 and women under 65 years); smoking means current or past smoking; and dyslipidaemia, hypertension and diabetes count if present at diagnosis.^[3] ESC 2024 adds that individual adjustment may be necessary for severe single risk factors or comorbidities associated with an increased prevalence of obstructive CAD that are not reflected in RF-CL, e.g. familial hypercholesterolaemia, severe kidney dysfunction, rheumatic or inflammatory diseases and peripheral artery disease.^[3]

• ESC 2024: it is recommended to estimate the pre-test likelihood of obstructive epicardial CAD using the Risk Factor-weighted Clinical Likelihood model (Class I, Level B).^[3]
• ESC 2024: it is recommended to use additional clinical data (e.g. examination of peripheral arteries, resting ECG, resting echocardiography, presence of vascular calcifications on previously performed imaging tests) to adjust the estimate yielded by the RF-CL model (Class I, Level C).^[3]
• ESC 2024: in individuals with a very low (5% or less) pre-test likelihood of obstructive CAD, deferral of further diagnostic tests should be considered (Class IIa, Level B).^[3]
• ESC 2024: in individuals with a low (more than 5% to 15%) pre-test likelihood of obstructive CAD, CACS should be considered to reclassify subjects and to identify more individuals with very low (5% or less) CACS-weighted clinical likelihood (Class IIa, Level B).^[3]
• ESC 2024: in individuals with an initially low (more than 5% to 15%) likelihood of obstructive CAD, exercise ECG and detection of atherosclerotic disease in non-coronary arteries may be considered to adjust the pre-test likelihood estimate (Class IIb, Level C).^[3]

What do the likelihood bands mean in practice? ESC 2024 states that individuals with a very low (5% or less) likelihood generally do not require further testing unless symptoms persist and non-cardiac causes have been excluded.^[3] With low likelihood (more than 5% to 15%), the benefit of testing is uncertain but it may be done if symptoms are limiting and need clarification.^[3] Moderate (more than 15% to 50%), high (more than 50% to 85%) and very high (more than 85%) likelihood groups are encouraged to undergo further testing.^[3]

The absence of coronary artery calcium (CACS = 0) has a very high negative predictive value (more than 95%) for obstructive CAD (ESC 2024).^[3] Of note, in younger patients obstructive CAD is rare, but when present a higher percentage have a CACS of 0 (58% of those younger than 40 years, compared with 9% of older patients with obstructive CAD aged 60 to 69 years).^[3] The likelihood models do not include the probability of ANOCA/INOCA, which always needs to be considered if symptoms persist after deferral of testing or after testing that excludes obstructive CAD.^[3]

ESC 2024: selecting and sequencing tests

• It is recommended to select the initial non-invasive diagnostic test based on pre-test likelihood of obstructive CAD, other patient characteristics that influence the performance of non-invasive tests (footnote: characteristics determining ability to exercise, likelihood of good image quality, expected radiation exposure, and risks or contraindications), and local expertise and availability (Class I, Level C).^[3]
• In symptomatic patients in whom the pre-test likelihood of obstructive CAD by clinical assessment is above 5%, CCTA or non-invasive functional imaging for myocardial ischaemia is recommended as the initial diagnostic test (Class I, Level B).^[3]
• To rule out obstructive CAD in individuals with low or moderate (more than 5% to 50%) pre-test likelihood, CCTA is recommended as the preferred diagnostic modality (Class I, Level B).^[3]
• Recommendation Table 8 (CCTA, if available, and supported by local expertise): in individuals with suspected CCS and low or moderate (more than 5% to 50%) pre-test likelihood of obstructive CAD, CCTA is recommended to diagnose obstructive CAD and to estimate the risk of MACE (Class I, Level A).^[3]
• CCTA is recommended in individuals with low or moderate (more than 5% to 50%) pre-test likelihood of obstructive CAD if functional imaging for myocardial ischaemia is not diagnostic (Class I, Level B); functional imaging for myocardial ischaemia is recommended if CCTA has shown CAD of uncertain functional significance or is not diagnostic (Class I, Level B).^[3]
• In patients with a known intermediate coronary artery stenosis (footnote: typically around 40%–90% by visual estimate) in a proximal or mid coronary segment on CCTA, CT-based FFR may be considered (Class IIb, Level B).^[3]
• Invasive coronary angiography with the availability of invasive functional assessments is recommended to confirm or exclude the diagnosis of obstructive CAD or ANOCA/INOCA in individuals with an uncertain diagnosis on non-invasive testing (Class I, Level B).^[3]
• Recommendation Table 8 (CCTA, if available, and supported by local expertise): CCTA is not recommended in patients with severe renal failure (eGFR below 30 mL/min/1.73 m²), decompensated heart failure, extensive coronary calcification, fast irregular heart rate, severe obesity, inability to cooperate with breath-hold commands, or any other conditions that can make obtaining good imaging quality unlikely (Class III, Level C).^[3]

Where does exercise ECG fit? ESC 2024 says exercise ECG testing may modify the likelihood of obstructive CAD and can be used in patients with low (more than 5% to 15%) clinical likelihood.^[3] In them, a negative test allows reclassification to the very low (5% or less) clinical likelihood group with a favourable prognosis.^[3] CCTA as a first-line test can give more accurate information and has been associated with fewer angina symptoms during follow-up than a strategy starting with exercise ECG.^[3]

• ESC 2024: exercise ECG may be considered as an alternative test to rule in and rule out CAD when non-invasive imaging tests are unavailable (Class IIb, Level B).^[3]
• ESC 2024: exercise ECG is not recommended for diagnostic purposes in patients with 0.1 mV or more ST-segment depression on resting ECG, left bundle branch block, or who are being treated with digitalis (Class III, Level C).^[3]
• ESC 2024: in individuals with a low or moderate (more than 5% to 50%) pre-test likelihood of obstructive CAD, an exercise ECG is not recommended to rule out CAD if CCTA or functional imaging tests are available (Class III, Level C).^[3]

^[3]

AHA/ACC 2021: stable chest pain

• Low risk, no known CAD: for patients with stable chest pain and no known CAD presenting to the outpatient clinic, a model to estimate pretest probability of obstructive CAD is effective to identify patients at low risk for obstructive CAD and favourable prognosis in whom additional diagnostic testing can be deferred (COR 1, LOE B-NR).^[1]
• Low risk, no known CAD: for patients with stable chest pain and no known CAD categorised as low risk, CAC testing is reasonable as a first-line test for excluding calcified plaque and identifying patients with a low likelihood of obstructive CAD (COR 2a, LOE B-R).^[1]
• Low risk, no known CAD: for patients with stable chest pain and no known CAD categorised as low risk, exercise testing without imaging is reasonable as a first-line test for excluding myocardial ischaemia and determining functional capacity in patients with an interpretable ECG (COR 2a, LOE B-NR).^[1]
• Intermediate-high risk, no known CAD: for intermediate-high risk patients with stable chest pain and no known CAD, CCTA is effective for diagnosis of CAD, for risk stratification, and for guiding treatment decisions (COR 1, LOE A).^[1]
• Intermediate-high risk, no known CAD: for intermediate-high risk patients with stable chest pain and no known CAD, stress imaging (stress echocardiography, PET/SPECT MPI or CMR) is effective for diagnosis of myocardial ischaemia and for estimating risk of MACE (COR 1, LOE B-R).^[1]
• Intermediate-high risk, no known CAD: for intermediate-high risk patients with stable chest pain and no known CAD with an interpretable ECG and ability to achieve maximal levels of exercise (5 METs or more), exercise electrocardiography is reasonable (COR 2a, LOE B-R).^[1]
• Intermediate-high risk, no known CAD (sequential testing): for intermediate-high risk patients with stable chest pain and known coronary stenosis of 40% to 90% in a proximal or middle coronary segment on CCTA, FFR-CT can be useful for diagnosis of vessel-specific ischaemia and to guide decision-making regarding the use of coronary revascularisation (COR 2a, LOE B-NR).^[1]

AHA/ACC 2021 says low-risk patients may alternatively be those younger than 40 years or with symptoms that have a low likelihood of representing ischaemia.^[1] Its figure of pretest probabilities of obstructive CAD in symptomatic patients (shown for patients with anginal symptoms) marks noninvasive testing as most beneficial where pretest probability is above 15%.^[1] At 15% or less, testing for diagnosis may be considered based on clinical judgment.^[1] Its pathway figure for stable chest pain with no known CAD says test choice should be guided by local availability and expertise.^[1] Its footnote adds that test choice is guided by exercise capacity and resting ECG abnormalities.^[1] CCTA is preferable in those younger than 65 years and not on optimal preventive therapies, and stress testing is favoured in those 65 years or older (with a higher likelihood of ischaemia).^[1]

A CAC score of zero identifies a low-risk cohort of symptomatic patients who may not require additional diagnostic testing; most events occur in those with detectable CAC (e.g. 84% in the PROMISE trial) (AHA/ACC 2021).^[1]

• AHA/ACC 2021, known CAD: for patients with obstructive CAD and stable chest pain, it is recommended to optimise GDMT (COR 1, LOE A); for patients with known nonobstructive CAD and stable chest pain, it is recommended to optimise preventive therapies (COR 1, LOE C-EO).^[1]
• AHA/ACC 2021, suspected INOCA: for patients with persistent stable chest pain and nonobstructive CAD and at least mild myocardial ischaemia on imaging, it is reasonable to consider invasive coronary function testing to improve the diagnosis of coronary microvascular dysfunction and to enhance risk stratification (COR 2a, LOE B-NR).^[1]

^{[7] [1] [6] [1]}

Specific scenarios

• Suspected acute aortic syndrome (AHA/ACC 2021): in patients with acute chest pain where there is clinical concern for aortic dissection, CT angiography of the chest, abdomen and pelvis is recommended for diagnosis and treatment planning (COR 1, LOE C-EO); TEE or CMR should be performed to make the diagnosis if CT is contraindicated or unavailable (COR 1, LOE C-EO).^[1]
• Suspected PE (AHA/ACC 2021): in stable patients with acute chest pain with high clinical suspicion for PE, CTA using a PE protocol is recommended (COR 1, LOE B-NR); for patients with acute chest pain and possible PE, the need for further testing should be guided by pretest probability (COR 1, LOE C-EO).^[1]
• Suspected myopericarditis (AHA/ACC 2021): in patients with acute chest pain and myocardial injury who have nonobstructive coronary arteries on anatomic testing, CMR with gadolinium contrast is effective to distinguish myopericarditis from other causes, including myocardial infarction and nonobstructive coronary arteries (MINOCA) (COR 1, LOE B-NR); in patients with acute chest pain and suspected myopericarditis, TTE is effective to determine the presence of ventricular wall motion abnormalities, pericardial effusion, valvular abnormalities or restrictive physiology (COR 1, LOE C-EO).^[1]
• Valvular heart disease (AHA/ACC 2021): in patients presenting with acute chest pain with suspected or known history of VHD, TTE is useful in determining the presence, severity and cause of VHD (COR 1, LOE C-EO).^[1]
• Sickle cell disease (AHA/ACC 2021): in patients with sickle cell disease who report acute chest pain, emergency transfer by EMS to an acute care setting is recommended (COR 1, LOE B-NR), and ACS should be excluded (COR 1, LOE C-LD).^[1]
• Troponin-positive with nonobstructive arteries (AHA/ACC 2021): for high-risk patients with acute chest pain who are troponin positive in whom obstructive CAD has been excluded by CCTA or ICA, CMR or echocardiography can be effective in establishing alternative diagnoses (COR 2a, LOE B-NR).^[1]

ESC 2023 adds that TTE can also suggest alternative causes of chest pain (i.e. acute aortic disease, or right ventricular signs in pulmonary embolism).^[2]

The MINOCA expansion in the AHA/ACC 2021 myopericarditis row is dated.^[9] The current Fifth UDMI (2026) updates the definition of MINOCA to myocardial injury with non-obstructive coronary arteries.^[9]

Complications and pitfalls

The 2025 ACC/AHA ACS guideline states that although most patients with chest pain will not have an ultimate diagnosis of ACS, a missed ACS event has significant implications for the patient’s morbidity and mortality.^[4]

Prognosis and disposition

^{[1] [2]}

Patients assigned to rule-out by the ESC 0 h/1 h or 0 h/2 h algorithms have a very low rate of clinical events through to 30 days (ESC 2023).^[2] Observe-zone patients are a heterogeneous group with mortality comparable to rule-in patients.^[2]

In chronic coronary syndromes, ESC 2024 defines annual cardiac mortality above 3% as high event risk, 1% to 3% as intermediate, and below 1% as low, based on large registries and historical randomised controlled trials (RCTs).^[3]

Special populations

• Women (AHA/ACC 2021): potential cardiac causes should always be considered, and the history should emphasise accompanying symptoms more common in women with ACS (both COR 1, LOE B-NR).^[1]
• Older than 75 years (AHA/ACC 2021): in patients with chest pain who are older than 75 years, ACS should be considered when accompanying symptoms such as shortness of breath, syncope or acute delirium are present, or when an unexplained fall has occurred (COR 1, LOE C-LD).^[1]
• Diverse racial and ethnic backgrounds (AHA/ACC 2021): cultural competency training is recommended to help achieve the best outcomes (COR 1, LOE C-LD); where English may not be the primary language, formal translation services are recommended (COR 1, LOE C-LD).^[1]
• Renal dysfunction (ESC 2023): in patients presenting with suspected NSTE-ACS, renal dysfunction is one of four clinical variables that affect hs-cTn concentrations beyond the presence or absence of MI, with differences of up to 300% between otherwise healthy patients with very high and very low eGFR.^[2]
• Australian priority groups (NHFA/CSANZ 2025 summary): the full guideline includes practice points for applying hs-cTn strategies in women, older adults, First Nations peoples and people with renal impairment, and test-selection considerations for First Nations peoples and regional and remote areas, including telemedicine support and a lower threshold for CTCA when available.^[5]
• Sickle cell disease (AHA/ACC 2021): emergency EMS transfer to an acute care setting (COR 1, LOE B-NR) and exclusion of ACS (COR 1, LOE C-LD) are recommended in patients with sickle cell disease who report acute chest pain.^[1]

Evidence, guidelines and regional differences

This topic draws on the 2021 AHA/ACC multisociety chest pain guideline and the 2025 ACC/AHA ACS guideline.^[4] The 2025 guideline marks its in-hospital initial-assessment rows (ECG, serial ECGs, troponin and repeat-troponin timing) as adapted from the 2021 guideline.^[4] This topic also uses the 2023 ESC ACS and 2024 ESC chronic coronary syndrome guidelines, and the 2025 NHFA/CSANZ Australian ACS guideline (via its Med J Aust summary), which replaces the 2016 Australian guideline.^[5][2][3] For definitions of myocardial injury and MI, this topic uses the Fifth UDMI (2026) from the joint ESC/ACC/AHA/WHF task force.^[9] It supersedes the fourth universal definition that ESC 2023 uses.^[9][2] The table sets the guidelines’ positions side by side.

How the guidelines compare (each cell from that body’s own text)

^{[2] [4] [5] [1] [3]}

One change to know: the 2025 ACC/AHA ACS guideline, in a row marked as adapted from the 2021 chest pain guideline, sets the repeat hs-cTn interval at 1 to 2 hours, where the 2021 chest pain guideline had 1 to 3 hours.^[4][1] Its supportive text for that row adds that CDPs using hs-cTn with repeat sampling at 1 or 2 hours from ED arrival and a delta calculation can identify very-low-risk patients (e.g. negative predictive value above 99.5%) based on assay-specific thresholds.^[4]

For imaging trials, read each result in its own population. RAPID-CTCA studied intermediate-risk acute chest pain with suspected ACS and found that early CT coronary angiography did not alter one-year clinical outcomes.^[8][2] SCOT-HEART studied patients with stable chest pain referred to a cardiology clinic and found a lower 5-year rate of death from coronary heart disease or nonfatal MI with CTA added to standard care.^[7] PROMISE studied symptomatic patients with suspected CAD who required noninvasive testing and found that initial CTA, compared with functional testing, did not improve clinical outcomes over a median follow-up of 2 years.^[6]

Exam pearls

References9ShowHide

1. [1]Gulati M, et al. 2021 AHA/ACC/ASE/CHEST/SAEM/SCCT/SCMR Guideline for the Evaluation and Diagnosis of Chest Pain: A Report of the American College of Cardiology/American Heart Association Joint Committee on Clinical Practice Guidelines. J Am Coll Cardiol, 2021.PMID 34756653
2. [2]Byrne RA, et al. 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J, 2023.PMID 37622654
3. [3]Vrints C, et al. 2024 ESC Guidelines for the management of chronic coronary syndromes. Eur Heart J, 2024.PMID 39210710
4. [4]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. J Am Coll Cardiol, 2025.PMID 40013746
5. [5]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
6. [6]Douglas PS, et al. Outcomes of anatomical versus functional testing for coronary artery disease. N Engl J Med, 2015.PMID 25773919
7. [7]Newby DE, et al. Coronary CT Angiography and 5-Year Risk of Myocardial Infarction. N Engl J Med, 2018.PMID 30145934
8. [8]Gray AJ, et al. Early computed tomography coronary angiography in patients with suspected acute coronary syndrome: randomised controlled trial. BMJ, 2021.PMID 34588162
9. [9]Mills NL, et al. Fifth Universal Definition of Myocardial Infarction (2026): On behalf of the Joint European Society of Cardiology (ESC)/American College of Cardiology (ACC)/American Heart Association (AHA)/World Heart Federation (WHF) Task Force for the Universal Definition of Myocardial Infarction Endorsed by the European Association for Cardio-Thoracic Surgery (EACTS) and the Society of Thoracic Surgeons (STS) Affirmation of Value by the Society for Cardiovascular Angiography and Interventions (SCAI). Glob Heart, 2026.PMID 42666939