Format

Send to

Choose Destination
Int J Cardiol. 2017 Oct 1;244:24-29. doi: 10.1016/j.ijcard.2017.06.055. Epub 2017 Jun 15.

GRACE risk score: Sex-based validity of in-hospital mortality prediction in Canadian patients with acute coronary syndrome.

Author information

1
University of Toronto, Toronto, ON, Canada.
2
University of Toronto, Toronto, ON, Canada; Terrence Donnelly Heart Centre, Division of Cardiology, St Michael's Hospital, Canadian Heart Research Centre, Toronto, ON, Canada.
3
Department of Cardiology, Concord Hospital, University of Sydney, Sydney, Australia.
4
Leeds Institute of Cardiovascular and Metabolic Medicine, University of Leeds, Leeds, UK.
5
Flinders University, South Australia, Australia.
6
Mazankowski Alberta Heart Institute, University of Alberta, Edmonton, AB, Canada.
7
McGill University Health Centre, McGill University, Montreal, QC, Canada.
8
Cornwall Community Hospital, Cornwall, ON and University of Ottawa, Ottawa, ON, Canada.
9
The Moncton Hospital, Moncton, NB and Dalhousie University, Halifax, NS, Canada.
10
Cardiovascular Division, Women's College Hospital, Peter Munk Cardiac Centre of the University Health Network, University of Toronto, Toronto, ON, Canada.
11
Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, UK.
12
University of Toronto, Toronto, ON, Canada; Terrence Donnelly Heart Centre, Division of Cardiology, St Michael's Hospital, Canadian Heart Research Centre, Toronto, ON, Canada. Electronic address: yana@smh.ca.

Abstract

BACKGROUND:

Although there are sex differences in management and outcome of acute coronary syndromes (ACS), sex is not a component of Global Registry of Acute Coronary Events (GRACE) risk score (RS) for in-hospital mortality prediction. We sought to determine the prognostic utility of GRACE RS in men and women, and whether its predictive accuracy would be augmented through sex-based modification of its components.

METHODS:

Canadian men and women enrolled in GRACE and Canadian Registry of Acute Coronary Events were stratified as ST-segment elevation myocardial infarction (STEMI) or non-ST-segment elevation ACS (NSTE-ACS). GRACE RS was calculated as per original model. Discrimination and calibration were evaluated using the c-statistic and Hosmer-Lemeshow goodness-of-fit test, respectively. Multivariable logistic regression was undertaken to assess potential interactions of sex with GRACE RS components.

RESULTS:

For the overall cohort (n=14,422), unadjusted in-hospital mortality rate was higher in women than men (4.5% vs. 3.0%, p<0.001). Overall, GRACE RS c-statistic and goodness-of-fit test p-value were 0.85 (95% CI 0.83-0.87) and 0.11, respectively. While the RS had excellent discrimination for all subgroups (c-statistics >0.80), discrimination was lower for women compared to men with STEMI [0.80 (0.75-0.84) vs. 0.86 (0.82-0.89), respectively, p<0.05]. The goodness-of-fit test showed good calibration for women (p=0.86), but suboptimal for men (p=0.031). No significant interaction was evident between sex and RS components (all p>0.25).

CONCLUSIONS:

The GRACE RS is a valid predictor of in-hospital mortality for both men and women with ACS. The lack of interaction between sex and RS components suggests that sex-based modification is not required.

KEYWORDS:

Acute coronary syndrome; GRACE risk score; Risk stratification; Sex differences; Validation

PMID:
28645803
DOI:
10.1016/j.ijcard.2017.06.055
[Indexed for MEDLINE]
Free full text

Supplemental Content

Full text links

Icon for Elsevier Science Icon for White Rose Research Online
Loading ...
Support Center