Guide détaillé à venir
Nous préparons un guide éducatif complet pour le HEART Score for Chest Pain. Revenez bientôt pour des explications étape par étape, des formules, des exemples concrets et des conseils d'experts.
The HEART Score is a validated 5-component clinical decision tool used in emergency departments worldwide to rapidly risk-stratify adult patients presenting with acute chest pain and suspected acute coronary syndrome (ACS). It was developed by Six and colleagues at the University Medical Centre Groningen, Netherlands, and first published in 2008 with subsequent large-scale prospective validation studies following. The acronym covers five domains: History (the clinical characteristics of the chest pain), ECG findings, Age, Risk factors for coronary artery disease, and Troponin level. Each component is scored 0, 1, or 2, giving a total score of 0–10. The score stratifies patients into low (0–3), moderate (4–6), and high (7–10) risk categories for major adverse cardiac events (MACE) — defined as myocardial infarction, percutaneous coronary intervention, coronary artery bypass grafting, or death — within 6 weeks. The primary power of the HEART Score lies in its low-risk category: patients scoring 0–3 have a MACE rate of approximately 1.7% at 6 weeks in the original derivation cohort and < 2% in most validation studies, supporting safe early discharge or observation without immediate invasive workup. This has been shown to reduce unnecessary hospital admissions by up to 20% and is endorsed in the 2022 AHA/ACC Chest Pain guidelines as a Class IIa recommendation.
HEART Score = H + E + A + R + T (total 0–10) • H — History (clinical characteristics of pain): 2 = Highly suspicious: typical chest pressure, radiation to arm/jaw, diaphoresis, onset with exertion, relief with nitrates 1 = Moderately suspicious: combination of typical and atypical features 0 = Slightly suspicious: mostly atypical features — sharp/positional/pleuritic pain, reproduced by palpation • E — ECG: 2 = Significant ST deviation: new ST depression ≥ 1 mm, new ST elevation, new LBBB, new T-wave inversion 1 = Non-specific repolarisation disturbance: LBBB or LVH not known to be new, early repolarisation, paced rhythm, non-specific ST/T changes 0 = Normal ECG: no significant abnormality • A — Age: 2 = Age ≥ 65 years 1 = Age 45–64 years 0 = Age < 45 years • R — Risk factors (CAD risk factors: known hypertension, hypercholesterolaemia, diabetes, obesity BMI > 30, smoking, family history of CAD): 2 = ≥ 3 risk factors OR history of atherosclerotic disease (prior MI, PCI, CABG, stroke, peripheral artery disease) 1 = 1–2 risk factors 0 = No known risk factors • T — Troponin (relative to assay-specific upper limit of normal, ULN): 2 = > 3× ULN 1 = 1–3× ULN 0 = ≤ ULN (within normal limits)
- 1Obtain a focused history and characterise the chest pain: location, quality, radiation, onset (exertion vs rest), duration, associated symptoms (diaphoresis, dyspnoea, nausea), and response to nitrates or antacids. Assign H = 2 for classic ACS presentation, H = 1 for mixed features, H = 0 for atypical pain (sharp, positional, pleuritic, reproduced by palpation).
- 2Review the 12-lead ECG, comparing with previous tracings if available. Assign E = 2 for significant new ST changes, T-wave inversion, or new LBBB. Assign E = 1 for non-specific changes, known LBBB, LVH, or paced rhythm. Assign E = 0 for a completely normal ECG.
- 3Record the patient's age. Assign A = 0 for age < 45, A = 1 for age 45–64, A = 2 for age ≥ 65. Age is the strongest fixed risk modifier in the score.
- 4Systematically assess cardiovascular risk factors: hypertension, hypercholesterolaemia, diabetes, obesity (BMI > 30), active or recent smoking, and family history of premature CAD (first-degree relative with MI or revascularisation before age 65 in women / 55 in men). Also ask about prior MI, PCI, CABG, stroke, or peripheral artery disease. Assign R = 2 for ≥ 3 factors or established atherosclerotic disease; R = 1 for 1–2 factors; R = 0 for none.
- 5Obtain troponin using the local high-sensitivity or conventional assay. Assign T = 0 for troponin ≤ ULN, T = 1 for troponin 1–3× ULN, T = 2 for troponin > 3× ULN. Note: high-sensitivity troponin allows earlier rule-out; the T component may be scored on initial or serial troponin.
- 6Sum all five components. Interpret: 0–3 = low risk (~1.7% MACE, consider discharge with outpatient follow-up); 4–6 = moderate risk (12–16.6% MACE, observation and serial troponin); 7–10 = high risk (50–65% MACE, early invasive strategy, cardiology referral).
- 7Document the HEART score in the patient record alongside the clinical rationale for each component. Low-risk patients (0–3) discharged from the ED should have clear return precautions and arranged outpatient cardiology or stress testing follow-up within 72 hours.
Low risk — consider discharge with outpatient follow-up
Pleuritic, positional chest pain with a normal ECG, no risk factors, and negative troponin in a young patient is highly unlikely to represent ACS. A HEART Score of 0 supports safe discharge. Arrange outpatient review and patient education on return precautions.
Moderate risk — admit for observation and serial troponin
A score of 6 sits at the upper boundary of the moderate category. This patient has several concerning features: classic pain quality with radiation, non-specific ECG changes, middle age, two risk factors, and a mildly elevated troponin. Serial troponin measurement, cardiology review, and likely stress testing or coronary imaging are appropriate.
High risk — activate cath lab, immediate cardiology
Maximum HEART Score of 10 in a patient with classic ACS presentation, ischaemic ECG changes, advanced age, established coronary disease, and significantly elevated troponin. This patient requires immediate activation of the cardiac catheterisation laboratory pathway and should not be discharged without revascularisation assessment.
Low risk but borderline — consider 1-hour serial troponin protocol
A score of 3 is technically low risk, but two clinical features merit attention: the ambiguous history (including dyspnoea and nausea, which are ACS equivalents in women) and the presence of two risk factors in a 48-year-old. Using a 0/1-hour or 0/3-hour high-sensitivity troponin protocol before discharge is appropriate to further reduce residual risk.
Emergency department triage of undifferentiated chest pain to identify patients safe for early discharge (score 0–3), reducing unnecessary hospital admissions by 20% without missing major cardiac events
Combined HEART Pathway protocol using serial high-sensitivity troponin alongside HEART Score to further reduce residual miss rate below 1% before safe discharge
Moderate-risk stratification (score 4–6) guiding whether patients receive CT coronary angiography vs invasive coronary angiography vs observation alone
High-risk identification (score 7–10) triggering immediate cardiology consultation and early cath lab activation for suspected NSTEMI
Quality improvement benchmarking in emergency cardiology — tracking HEART Score distributions and outcomes to audit appropriate use of the score in ED populations
STEMI and haemodynamic instability
HEART Score should NOT be applied to patients with ST-elevation myocardial infarction (STEMI) or haemodynamic instability. These patients require immediate revascularisation regardless of score. HEART Score is a triage tool for stable undifferentiated chest pain, not an ACS treatment guide.
History component in women and diabetics
Women and diabetics are more likely to present with atypical ACS symptoms — dyspnoea, fatigue, nausea, or jaw/back pain without classic chest pressure. Clinicians should recognise ACS equivalents and not reflexively assign H = 0 simply because the pain is atypical in character. The 2022 AHA/ACC guidelines specifically recommend awareness of sex differences in symptom presentation.
Renal failure and troponin elevation
Patients with chronic kidney disease (CKD) frequently have chronically elevated troponin due to reduced renal clearance and myocardial stress, not acute MI. In known CKD patients, serial troponin change (delta troponin) is more informative than absolute level for diagnosing acute injury. A rising troponin (> 20–30% increase over 3 hours) suggests acute myocardial injury regardless of baseline.
Prior CABG or PCI — R criterion
Any history of prior coronary revascularisation (PCI or CABG) automatically scores R = 2, regardless of additional risk factors, because established coronary artery disease carries inherently high risk for recurrent events. This should always be verified in the patient history and from previous medical records.
HEART Score vs HEART Pathway
The 'HEART Pathway' is a clinical protocol that combines HEART Score ≤ 3 with serial high-sensitivity troponin (0h and 3h) to guide disposition decisions. It is more structured than using HEART Score alone and has been prospectively validated in a randomised trial (Mahler et al.) showing 21% absolute reduction in hospital admissions without missing MACE events. Many EDs implement the HEART Pathway rather than HEART Score in isolation.
| Score | Risk Category | MACE at 6 Weeks | Recommended Action |
|---|---|---|---|
| 0–3 | Low | ~1.7% | Consider early discharge + outpatient follow-up within 72h |
| 4–6 | Moderate | ~12–17% | Observation, serial troponin, non-invasive or invasive workup |
| 7–10 | High | ~50–65% | Early invasive strategy, cardiology referral, consider cath lab activation |
What does the HEART Score measure?
The HEART Score measures the probability of a major adverse cardiac event (MACE) within 6 weeks in patients presenting to the emergency department with acute chest pain. MACE is defined as acute myocardial infarction, percutaneous coronary intervention (PCI), coronary artery bypass grafting (CABG), significant stenosis requiring admission, or all-cause death. A score of 0–3 indicates low risk (~1.7% MACE), 4–6 moderate risk (~12–17%), and 7–10 high risk (~50–65%).
Who should the HEART Score be used for?
HEART Score is designed for adult patients presenting to the emergency department with undifferentiated, potentially cardiac chest pain. It should NOT be applied to patients with obvious STEMI (who need immediate catheterisation regardless of score), haemodynamic instability, or non-cardiac chest pain that has already been clearly characterised. It works best in the intermediate-risk undifferentiated population where clinical uncertainty is highest.
Can low-risk patients (score 0–3) be safely discharged?
Yes, with important caveats. Multiple prospective studies and the HEART Pathway validation trial show that patients scoring 0–3 with a negative initial troponin have a MACE risk of < 2% at 6 weeks, comparable to the general population background risk. However, discharge should include clear written return-to-ED instructions, arranged outpatient cardiology or exercise testing within 72 hours, and clinical judgment — particularly for patients with atypical presentations in women, diabetics, or those with abnormal vital signs.
How does HEART Score compare to TIMI and GRACE?
HEART Score was specifically designed for undifferentiated ED chest pain and has been shown to outperform TIMI risk score (designed for confirmed NSTEMI/UA patients) and GRACE (designed for confirmed ACS) in predicting MACE in unselected ED presentations. TIMI performs poorly in low-risk identification (many low-risk patients still score TIMI 0–1 but have higher-than-expected event rates). HEART Score is preferred by the 2022 AHA/ACC Chest Pain Guidelines for initial ED risk stratification.
Which troponin threshold should I use for the T component?
Use the 99th percentile upper limit of normal (ULN) for your local assay — this varies between conventional troponin I (typically 0.04 µg/L for most assays), conventional troponin T (0.014 µg/L for Roche assay), and high-sensitivity troponins (assay-specific; e.g., hs-TnI < 16 ng/L for Abbott ARCHITECT). Score T = 0 if ≤ 1× ULN, T = 1 if 1–3× ULN, T = 2 if > 3× ULN. For serial troponin protocols (0h/1h or 0h/3h), use the peak value for scoring.
Is HEART Score validated in women?
Yes, but with nuance. Women have lower troponin ULN values on some assays (sex-specific thresholds are now recommended by ESC guidelines), and may present with more atypical symptoms (dyspnoea, nausea, fatigue as primary complaint rather than chest pressure). Applying sex-specific troponin thresholds improves HEART Score accuracy in women. The H component should also weight 'atypical ACS equivalents' in women appropriately, as pure pressure-like pain is less common.
What happens after a moderate HEART Score (4–6)?
Moderate risk patients (HEART 4–6) should be admitted or observed for serial high-sensitivity troponin measurement (0h/1h or 0h/3h protocol). If serial troponins remain negative and the clinical picture is reassuring, many institutions now use non-invasive imaging (CT coronary angiography or stress testing) rather than automatic invasive angiography. The definitive management pathway is guided by local cardiology protocols, evolving serial troponin results, and the overall clinical gestalt.
Can HEART Score be used for COVID-19 chest pain?
HEART Score was not validated in COVID-19 populations, and myocarditis, MINOCA (MI with non-obstructive coronary arteries), and pulmonary embolism — all elevated in COVID-19 — can elevate troponin and present with chest pain without obstructive CAD. In COVID-19-positive patients with chest pain and troponin elevation, the HEART Score may overestimate ACS probability. Clinical judgment, D-dimer, and ECG should complement the score in this context.
Conseil Pro
Use the HEART Pathway protocol rather than HEART Score alone: combine HEART Score ≤ 3 with serial 0h/1h or 0h/3h high-sensitivity troponin (both measurements ≤ ULN) for safe early discharge. This dual-strategy approach has been validated in the HEART Pathway RCT and reduces admissions by ~21% without missing major events. Always document which troponin assay and ULN threshold your institution uses.
Le saviez-vous?
The HEART Score was derived in just 122 patients in a single Dutch centre in 2008 and was initially met with scepticism. However, subsequent prospective validation across international cohorts totalling over 15,000 patients has consistently confirmed its accuracy, making it one of the most rapidly adopted clinical decision tools in emergency cardiology. The 2022 AHA/ACC Chest Pain Guidelines — the first dedicated chest pain guideline ever published — gave it a Class IIa recommendation, reflecting over a decade of evidence.
Références
- ›Six AJ et al. The HEART score for the assessment of patients with chest pain in the emergency department — Neth Heart J 2010
- ›Backus BE et al. Prospective validation of the HEART score for chest pain patients at the emergency department — Int J Cardiol 2010
- ›Mahler SA et al. The HEART Pathway randomized trial — Circ Cardiovasc Qual Outcomes 2015
- ›Fanaroff AC et al. Does This Patient With Chest Pain Have Acute Coronary Syndrome? — JAMA 2015
- ›Writing Committee et al. 2021 AHA/ACC/ASE/CHEST/SAEM/HRS/SCCM Guideline for the Evaluation and Diagnosis of Chest Pain — J Am Coll Cardiol 2021