GRACE ACS Risk Score Calculator
GRACE ACS Risk Score Calculator
Score all eight GRACE variables from the admission observations and see the registry’s own in-hospital risk band, with the distance from the ESC’s 140-point early-invasive threshold.
GRACE Risk Score
8 variables → 1–372 points67-year-old with a non-ST-elevation ACS. Heart rate 96, systolic blood pressure 128 mmHg, creatinine 1.1 mg/dL, Killip class I, no cardiac arrest at admission, ST-segment deviation present on the admission ECG, troponin raised
Eight variables, added
- Version
- the original 8-variable in-hospital-death model of Granger et al, 2003. Not GRACE 2.0 — see below
- Age
- the largest single contribution: 0 points below 30, 100 points at 90 or over
- Systolic BP
- scored in reverse — 58 points below 80 mmHg, 0 points at 200 or above
- Creatinine
- in mg/dL. Never 0 points: the lowest row still scores 1
- Maximum
- 372 points, which is also the upper bound of the registry’s own high-risk band
- ESC threshold
- a GRACE score above 140 is one of the 2023 ESC high-risk criteria for early invasive angiography within 24 hours
Worked example
67-year-old with a non-ST-elevation ACS. Heart rate 96, systolic blood pressure 128 mmHg, creatinine 1.1 mg/dL, Killip class I, no cardiac arrest at admission, ST-segment deviation present on the admission ECG, troponin raised
Age 67 → the 60–69 row → 58 points, which is more than a third of the total on its own
Heart rate 96 → 90–109 → 15; systolic 128 → 120–139 → 34
Creatinine 1.1 mg/dL → 0.80–1.19 → 7
Killip class I → 0; no cardiac arrest at admission → 0
ST-segment deviation → 28; elevated cardiac markers → 14
58 + 15 + 34 + 7 + 0 + 0 + 28 + 14 = 156 points
156 is in the registry's high-risk band (141–372, over 3% in-hospital death) and is 16 points above the ESC's 140-point threshold for considering invasive angiography within 24 hours
The GRACE point table — all eight variables
| Variable | Category | Points |
|---|---|---|
| Age (years) | <30 | 0 |
| Age (years) | 30 – 39 | 8 |
| Age (years) | 40 – 49 | 25 |
| Age (years) | 50 – 59 | 41 |
| Age (years) | 60 – 69 | 58 |
| Age (years) | 70 – 79 | 75 |
| Age (years) | 80 – 89 | 91 |
| Age (years) | ≥90 | 100 |
| Heart rate (per minute) | <50 | 0 |
| Heart rate (per minute) | 50 – 69 | 3 |
| Heart rate (per minute) | 70 – 89 | 9 |
| Heart rate (per minute) | 90 – 109 | 15 |
| Heart rate (per minute) | 110 – 149 | 24 |
| Heart rate (per minute) | 150 – 199 | 38 |
| Heart rate (per minute) | ≥200 | 46 |
| Systolic BP (mmHg) | <80 | 58 |
| Systolic BP (mmHg) | 80 – 99 | 53 |
| Systolic BP (mmHg) | 100 – 119 | 43 |
| Systolic BP (mmHg) | 120 – 139 | 34 |
| Systolic BP (mmHg) | 140 – 159 | 24 |
| Systolic BP (mmHg) | 160 – 199 | 10 |
| Systolic BP (mmHg) | ≥200 | 0 |
| Creatinine (mg/dL) | <0.40 | 1 |
| Creatinine (mg/dL) | 0.40 – 0.79 | 4 |
| Creatinine (mg/dL) | 0.80 – 1.19 | 7 |
| Creatinine (mg/dL) | 1.20 – 1.59 | 10 |
| Creatinine (mg/dL) | 1.60 – 1.99 | 13 |
| Creatinine (mg/dL) | 2.00 – 3.99 | 21 |
| Creatinine (mg/dL) | ≥4.00 | 28 |
| Killip class | I — no heart failure | 0 |
| Killip class | II — basal crackles or raised JVP | 20 |
| Killip class | III — pulmonary oedema | 39 |
| Killip class | IV — cardiogenic shock | 59 |
| Cardiac arrest at admission | Present | 39 |
| ST-segment deviation | Present | 28 |
| Elevated cardiac markers | Present | 14 |
The four published GRACE risk bands — and they are four different models
| Population and endpoint | Low | Intermediate | High |
|---|---|---|---|
| NSTE-ACS, in-hospital death (this calculator) | 1 – 108 (<1%) | 109 – 140 (1–3%) | 141 – 372 (>3%) |
| NSTE-ACS, death 6 months after discharge | 1 – 88 (<3%) | 89 – 118 (3–8%) | 119 – 263 (>8%) |
| STE-ACS, in-hospital death | 49 – 125 (<2%) | 126 – 154 (2–5%) | 155 – 319 (>5%) |
| STE-ACS, death 6 months after discharge | 27 – 99 (<4.4%) | 100 – 127 (4.5–11%) | 128 – 263 (>11%) |
GRACE against the other two ACS scores on this site
| GRACE | TIMI (NSTEMI) | HEART | |
|---|---|---|---|
| Answers | How likely is this patient to die in hospital? | How likely is a 14-day event in a patient already diagnosed? | Can this chest-pain patient go home? |
| Assumes | ACS is established | ACS is established | Nothing — it is a triage score |
| Variables | 8, four of them continuous | 7, all binary | 5, each scored 0–2 |
| Dominated by | Age, which can supply 100 of 372 points | Nothing — all items score 1 | History and troponin |
| Needs | Creatinine, Killip class, an ECG and a troponin | An ECG, a troponin and a history | An ECG, a troponin and a history |
| On this site | This page | TIMI risk score (NSTEMI) | HEART score |
Why there is no mortality percentage on this page, and no 6-month figure at all
GRACE is the risk score that the European guidelines actually name. It takes eight things known within minutes of admission — age, heart rate, systolic pressure, creatinine, Killip class, whether the patient arrested before arrival, whether the ECG shows ST-segment deviation, and whether the troponin is raised — and adds them to a total between 1 and 372. The registry’s own risk table then sorts non-ST-elevation ACS into three in-hospital mortality bands at 108 and 140 points, and the 2023 ESC acute coronary syndrome guideline lists “NSTE-ACS with GRACE score >140” among the high-risk criteria for which patients “should be considered for early invasive angiography (i.e. within 24 h)”. That single threshold is the reason most clinicians calculate it, which is why this page prints the signed distance from 140 beside the total rather than making you do the subtraction.
Two figures that other implementations print are missing here, and their absence is deliberate. The first is a mortality percentage for the individual score. Granger’s paper carries a nomogram that maps a point total onto a probability, and it could not be verified: the journal is not reachable from here, the sites that reproduce the card disagree with each other, and the GRACE registry’s own page publishes the three-band table and not a per-score lookup. A percentage that cannot be traced to the model it came from is worse than no percentage, so the band ranges are on the page, labelled with the population they were derived in, and nothing finer is claimed.
The second is a 6-month mortality, and that one cannot be derived from this total at all — which is the single most useful thing this page has to say. The in-hospital model is the eight-variable score implemented here, and its maximum is 372. The discharge-to-6-month model is a different model: it drops Killip class and cardiac arrest, adds a history of congestive heart failure, a prior myocardial infarction and the absence of in-hospital PCI, and has a maximum of 263. Printing a “6-month mortality” next to an in-hospital point total means reading one model’s total off another model’s table, and the answer is not approximately right — it is a category error. The four-row table above makes it visible: four maxima, four populations, four different scores that happen to share a name. If you need the 6-month estimate, the nine variables have to be collected and the 6-month model used.
A third version exists and is not implemented. GRACE 2.0 replaced the additive point table with a non-linear model and extended the horizon to one and three years, and it is the version the ESC guideline text refers to when it talks about GRACE-based estimates. Its coefficients are splines rather than a points chart, so there is no honest way to reproduce it as an additive table — which is exactly why a page like this has to say which version it is. This is the 2003 additive score. Two further cautions about the inputs: the creatinine ladder is in mg/dL, so a report in µmol/L needs the creatinine converter first, and Killip class is a bedside assessment that cannot be filled in from a laboratory report. This supports a clinician’s judgement rather than replacing it. It is arithmetic on the figures entered, and it knows nothing about the patient in front of you.
Frequently asked questions
What GRACE score is considered high risk?
For in-hospital death in non-ST-elevation ACS, the GRACE registry’s own risk table puts 1–108 points at under 1%, 109–140 at 1–3% and 141–372 at over 3%. Separately, the 2023 ESC guideline treats a score above 140 as a high-risk criterion for considering invasive angiography within 24 hours.
Why does this calculator not show a 6-month mortality?
Because the 6-month estimate comes from a different model. The in-hospital score has eight variables and a maximum of 372; the discharge-to-6-month score drops Killip class and cardiac arrest, adds congestive heart failure, prior myocardial infarction and no in-hospital PCI, and has a maximum of 263. Reading an in-hospital total off the 6-month table would be a category error, not an approximation.
Which variable contributes most to the GRACE score?
Age, by a wide margin. It contributes 0 points below 30 and 100 points at 90 or over — more than a quarter of the 372-point maximum, and usually the largest single term. This calculator prints the points coming from age alone for that reason.
Is this GRACE or GRACE 2.0?
This is the original 2003 additive eight-variable score for in-hospital death. GRACE 2.0 is a non-linear model reporting one- and three-year risk, and its coefficients are not published in a form an additive points table can reproduce, so it is described on this page rather than implemented.
What unit does the GRACE creatinine use?
mg/dL. The ladder runs from under 0.40 mg/dL (1 point) to 4.00 mg/dL and above (28 points), and note that the lowest row still scores 1 — creatinine never contributes zero. Convert a µmol/L result before entering it; 88.4 µmol/L is 1.0 mg/dL.
Related calculators
References
- Granger CB, Goldberg RJ, Dabbous O, et al. Predictors of hospital mortality in the Global Registry of Acute Coronary Events. Arch Intern Med. 2003;163(19):2345–53.
- 2023 ESC Guidelines for the management of acute coronary syndromes. Eur Heart J. 2023;44(38):3720–826 — “NSTE-ACS with GRACE score >140” among the high-risk criteria for early invasive angiography within 24 h. Verified against the published text.
- Center for Outcomes Research, University of Massachusetts Medical School. GRACE risk table — in-hospital and 6-month mortality bands for NSTE-ACS and STE-ACS. Verified against the registry’s own published table; the source of all four band sets on this page.
- GRACE risk score validation study, Egypt Heart J. doi:10.1186/s43044-021-00146-9 — independently quotes “low risk for GRACE score ≤108, intermediate risk … between 109 and 140, and high risk for GRACE score ≥141”. Cited by DOI; the author list was not displayed by the source consulted.
Medical Disclaimer: The tools and content provided here are for educational and reference purposes only. They are not intended to substitute for professional medical advice, diagnosis, or treatment. Clinical decisions should always be based on the comprehensive assessment of a qualified healthcare professional.