GRACE Score Calculator
An interactive implementation of the GRACE risk score (Granger et al., Global Registry of Acute Coronary Events): eight admission variables estimating in-hospital and 6-month mortality in acute coronary syndrome.
Enter the eight admission variables, age, heart rate, systolic blood pressure, creatinine, Killip class, cardiac arrest at admission, ST deviation and cardiac enzyme elevation, to compute the GRACE score and read off the corresponding in-hospital and 6-month mortality bands. Use it at presentation to guide the timing of invasive management.
What is the GRACE score?
The GRACE score is the best-validated mortality prediction tool in acute coronary syndrome, and it underpins the timing of invasive management in every contemporary guideline. Derived from more than 11,000 patients in the Global Registry of Acute Coronary Events and published by Granger and colleagues in JAMA in 2003, it was later expanded with a 6-month mortality model. The registry spanned the full ACS spectrum, from unstable angina through non-ST-elevation MI to ST-elevation MI, which is why the score applies across presentations rather than to one subtype.
Its clinical job is early risk stratification. When a patient arrives with suspected ACS, the GRACE score computed from admission variables separates those who can safely wait for angiography from those who need it urgently. European Society of Cardiology guidance recommends an invasive strategy within 24 hours for non-ST-elevation ACS patients with a GRACE score above 140, alongside other high-risk features, while lower-risk patients can be managed on a less urgent timetable. The score therefore directly shapes logistics: catheter laboratory scheduling, transfer decisions and the intensity of monitoring.
The eight variables and their points
Age contributes the most points of any variable, rising from 0 below age 30 to 100 at age 90 or above, reflecting the steep age gradient of ACS mortality. Heart rate and systolic blood pressure capture haemodynamic stress in opposite directions: faster rates add points while higher pressures subtract them, so hypotension and tachycardia, the physiology of shock, score heavily. Serum creatinine adds points across its range because renal dysfunction is a powerful independent predictor of death after ACS. Killip class contributes 0, 20, 39 or 59 points for classes I through IV, making the bedside examination one of the highest-weighted inputs.
The three binary variables complete the picture. Cardiac arrest at admission adds 39 points, ST-segment deviation on the presenting ECG adds 28, and elevated cardiac enzymes or markers add 14. Note the hierarchy this implies: a patient can accumulate a high score through haemodynamic collapse alone, through advanced age and renal impairment, or through the classic ischaemic markers, and the score treats these as partly interchangeable paths to the same mortality risk. The total has no fixed maximum in practice but typically ranges from under 60 to above 300.
How to interpret the score
For in-hospital mortality the validated bands are: 108 or below, under 1 percent; 109 to 140, 1 to 3 percent; 141 to 173, 3 to 8 percent; and above 173, over 8 percent. For 6-month mortality the bands are: 88 or below, under 3 percent; 89 to 118, 3 to 8 percent; 119 to 150, 8 to 19 percent; and above 150, over 19 percent. The calculator reports both, because the admission decision and the post-discharge plan answer different questions.
The therapeutically important threshold is 140. In non-ST-elevation ACS, crossing it moves the patient into the group for whom early invasive management within 24 hours is recommended, assuming no contraindications. Below 109, the in-hospital risk is under 1 percent and a selective or delayed invasive approach is reasonable. Between these values, the score informs but does not dictate: frailty, bleeding risk, renal function, coronary anatomy and patient preference all modulate the timing decision. The score should be computed once, at presentation, from the first available values, since it was derived and validated that way.
Creatinine units and common pitfalls
The most frequent error in GRACE calculation is the creatinine unit. The point table uses mg/dL: 0 to 0.39 scores 1 point, 0.4 to 0.79 scores 4, 0.8 to 1.19 scores 7, 1.2 to 1.59 scores 10, 1.6 to 1.99 scores 13, 2.0 to 3.99 scores 21, and 4.0 or above scores 28. Laboratories reporting in micromol per litre need conversion by dividing by 88.4; entering a micromol value such as 110 directly would massively overstate renal dysfunction. The calculator expects mg/dL and rejects implausible values.
Other pitfalls include using blood pressure and heart rate values from after treatment rather than on admission, misclassifying Killip class, and forgetting that the score predicts mortality, not the composite ischaemic endpoints that the TIMI score addresses. GRACE also predates high-sensitivity troponin and contemporary rapid rule-out pathways; it remains valid for risk stratification but should sit alongside, not ahead of, the diagnostic workup for myocardial infarction itself.
Limitations
GRACE was derived from registry patients hospitalised with ACS, so it does not apply to undifferentiated chest pain in the emergency department before ACS is diagnosed. Its discrimination for mortality is good, with c-statistics around 0.8, but it is a population model and individual trajectories vary. The score does not incorporate frailty, bleeding risk, left ventricular function on imaging, or the results of angiography, all of which refine prognosis once available. Absolute event rates have improved since the derivation era with faster reperfusion and modern secondary prevention, so contemporary mortality may run slightly below the published bands. Finally, the score guides timing and intensity of care; it never replaces the clinical diagnosis of ACS or the indication for reperfusion in ST-elevation MI, which is an emergency whatever the number.
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):2345-2353.
- Eagle KA, Lim MJ, Dabbous OH, et al. A validated prediction model for all forms of acute coronary syndrome: estimating the risk of 6-month postdischarge death in an international registry. JAMA. 2004;291(22):2727-2733.
- Collet JP, Thiele H, Barbato E, et al. 2020 ESC Guidelines for the management of acute coronary syndromes in patients presenting without persistent ST-segment elevation. Eur Heart J. 2021;42(14):1289-1367.
Key takeaways
- The GRACE score sums eight admission variables, age, heart rate, systolic blood pressure, creatinine, Killip class, cardiac arrest, ST deviation and cardiac enzyme elevation, to estimate mortality in acute coronary syndrome.
- In-hospital mortality bands are 108 or below (under 1 percent), 109 to 140 (1 to 3 percent), 141 to 173 (3 to 8 percent) and above 173 (over 8 percent).
- Six-month mortality bands are 88 or below (under 3 percent), 89 to 118 (3 to 8 percent), 119 to 150 (8 to 19 percent) and above 150 (over 19 percent).
- A score above 140 in non-ST-elevation ACS identifies patients for an early invasive strategy within 24 hours.
- Calculate the score once at presentation from admission values; enter creatinine in mg/dL (divide micromol/L by 88.4).
- The score predicts mortality and guides timing of care; it does not diagnose ACS or replace the emergency reperfusion indication in ST-elevation MI.
Frequently asked questions
What is the GRACE score used for?
The GRACE score estimates the risk of in-hospital death and of death within 6 months for patients presenting with acute coronary syndrome, covering unstable angina, non-ST-elevation myocardial infarction and ST-elevation myocardial infarction. It sums points for eight admission variables: age, heart rate, systolic blood pressure, serum creatinine, Killip class, cardiac arrest at admission, ST-segment deviation and elevated cardiac enzymes. Guidelines use it to identify high-risk patients who benefit from early invasive management.
What GRACE score indicates high risk?
For in-hospital mortality, a GRACE score above 140 is generally considered high risk with 1 to 3 percent mortality, and above 173 very high risk with over 8 percent mortality. For 6-month mortality, scores above 118 are high risk. In practice, a score over 140 in non-ST-elevation ACS identifies patients for whom guidelines recommend an early invasive strategy within 24 hours.
When should the GRACE score be calculated?
At presentation, using the first available admission values: the initial heart rate and blood pressure, the admission creatinine, the Killip class on examination, whether cardiac arrest occurred at admission, and the presenting ECG and first troponin. It is an admission risk stratification tool, not a serial monitoring score.
Can GRACE be used in ST-elevation myocardial infarction?
Yes. The GRACE registry enrolled the full spectrum of acute coronary syndrome, including ST-elevation MI, and the score is validated across it. In ST-elevation MI, immediate reperfusion takes priority regardless of the score, but GRACE still informs prognosis and the intensity of subsequent care.
What is the difference between GRACE and TIMI risk scores?
Both stratify risk in acute coronary syndrome, but GRACE uses eight weighted continuous and categorical variables with finer point gradations and is better calibrated for mortality prediction, while the TIMI risk score for unstable angina or NSTEMI uses seven equally weighted binary variables and predicts a composite of death, MI or urgent revascularization at 30 days. Guidelines favour GRACE for mortality risk stratification.
Which creatinine units does the GRACE score use?
The point table uses mg/dL. If the laboratory reports micromol per litre, divide by 88.4 before entering the value. Entering a micromol/L value directly would greatly overstate renal dysfunction and inflate the score.