Electronic GRACE risk score updated, validated out to 3 years

AMSTERDAM — An updated version of the electronic GRACE risk score has been validated for accuracy up to 3 years after a patient presents with ACS, according to a presentation at the European Society of Cardiology Congress 2013.

The GRACE risk score has been validated and is recommended by the ESC guidelines for the acute management of ACS; however, it was previously unknown whether the risk score is an accurate predictor of longer-term outcomes.

Researchers have developed an updated, user-friendly, fully electronic version of the GRACE risk score that can be downloaded for free on any smartphone, mobile or electronic device, according to presenter Keith A.A. Fox, MB, ChB, Duke of Edinburgh professor of cardiology at the University of Edinburgh, Scotland.

Fox presented the first study of longer-term outcomes with determinations of predictive accuracy up to 3 years after ACS presentation with external validation of the updated risk score in the French FAST-MI registry.

The updated version of the GRACE risk score was developed using data from 32,037 patients, 3,655 of whom experienced death or MI. The accuracy of the risk score was validated externally using patients with available variables from FAST-MI 2005, a French registry of patients with STEMI or non-STEMI (n=2,959; mean age, 66.9 years; 69% men; 53% STEMI).

The updated version of the GRACE risk score had a C-index of 0.83 for predicting overall death after 1 year and 0.82 after 3 years for patients in FAST-MI. It was associated with a C-index of 0.773 for predicting death or MI after 1 year and 0.773 after 3 years. The closer a C-index is to 1, the better the set of predictions; a C-index of 0.5 is equivalent to a coin flip. Overall survival after 3 years was 79% and infarct-free survival was 73%. The C statistics are similar to those in the original version of the GRACE risk score for death after 6 months (0.82) and death or MI after 6 months (0.7), according to the study abstract.

A clinician can use the GRACE risk score to determine which ACS patients qualify as high risk and are candidates for more intensive long-term treatment. The clinician receives a risk score after inputting eight factors:

Age
Heart rate
Systolic BP
Killip class
Creatinine
ST-deviation
Biomarkers of necrosis
Cardiac arrest

“The limitations of GRACE and the previous risk scores were that they were rather clumsy to use,” Fox told Cardiology Today before the presentation. “They often needed look-up tables or numerical calculations. What’s different about GRACE 2.0 is that it’s a smart, electronic version of the risk score. It’s actually more accurate than the original because it uses nonlinear algorithms.”

The updated version allows clinicians to input history of renal dysfunction instead of creatinine and use of diuretics instead of Killip class if creatinine and Killip class are not available, as is the case with a lot of patients, Fox said. “The figures [using the substitutions] are almost as accurate as the full GRACE score.”

This tool “provides the clinician and, for the first time, the paramedic … with a tool that allows them to say, ‘Here is a high-risk non-STEMI patient that needs to go to a major intervention center for early intervention,’” Fox said. “What we’ve calculated is that based upon all the randomized trials, if patients were stratified correctly according to high and low risk, then a much larger proportion would get guideline-indicated interventions and improved outcomes.” – by Erik Swain

For more information:

Fox KAA. Registry Hot Line: Registries on risks, therapies and outcomes. Presented at: the European Society of Cardiology Congress; Aug. 31-Sept. 4, 2013; Amsterdam.

Disclosure: The GRACE program is supported by grants from the British Heart Foundation, The Chief Scientist of Scotland and AstraZeneca. The researchers report no relevant financial disclosures.

PAGE BREAK