External validation of the GRACE risk score and the risk-treatment paradox in patients with acute coronary syndrome

Abstract

Objectives: To validate the Global Registry of Acute Coronary Events (GRACE) risk score and examine the extent and impact of the risk-treatment paradox in contemporary patients with acute coronary syndrome (ACS).

Methods: Data from 5015 patients with ACS enrolled in the FORCE-ACS registry between January 2015 and December 2019 were used for model validation. The performance of the GRACE risk score for predicting in-hospital and 1-year mortality was evaluated based on indices of model discrimination and calibration. Differences in the delivery of guideline-recommended care among patients who survived hospitalisation (n=4911) per GRACE risk stratum were assessed and the association with postdischarge mortality was examined.

Results: Discriminative power of the GRACE risk score was good for predicting in-hospital (c-statistic: 0.86; 95% CI: 0.83 to 0.90) and 1-year mortality (c-statistic: 0.82; 95% CI: 0.79 to 0.84). However, the GRACE risk score overestimated the absolute in-hospital and 1-year mortality risk (Hosmer-Lemeshow goodness-of-fit test p<0.01). Intermediate-risk and high-risk patients were 12% and 29% less likely to receive optimal guideline-recommended care compared with low-risk patients, respectively. Optimal guideline-recommended care was associated with lower mortality in intermediate- and high-risk patients.

Conclusions: The GRACE risk score identified patients at higher risk for in-hospital and 1-year mortality, but overestimated absolute risk levels in contemporary patients. Optimal guideline-recommended care was associated with lower mortality in intermediate-risk and high-risk patients, but was less likely to be delivered with increasing mortality risk.

Trial registration: ClinicalTrials.gov NCT03823547.

Keywords: acute coronary syndrome; myocardial infarction; pharmacology, clinical.

Conflict of interest statement

Competing interests: Dr Wouter J Kikkert has received a research grant from AstraZeneca. Dr Georgios J Vlachojannis has research grants from MicroPort and Ferrer and personal fees from Terumo and AstraZeneca. Dr Yolande Appelman has received a research grant from the Dutch Heart Foundation. Professor Dr José PS Henriques has received research grants from Abbott Vascular, AstraZeneca, B. Braun, Getinge, Ferrer, Infraredx and ZonMw. Professor Dr Jurriën M ten Berg has received research grants from AstraZeneca and ZonMw and personal fees from AstraZeneca, Accu-Metrics, Bayer, Boehringer Ingelheim, Bristol-Myers Squibb, Daiichi Sankyo, Eli Lilly, Ferrer, Idorsia, Pfizer and The Medicines Company. All other authors have no relationships with industry to disclose.

© Author(s) (or their employer(s)) 2022. Re-use permitted under CC BY-NC. No commercial re-use. See rights and permissions. Published by BMJ.

Figures

Figure 1

Receiver operating characteristic curve of the Global Registry of Acute Coronary Events risk score for in-hospital mortality (blue) and 1-year mortality (green) in the validation cohort. AUC, area under the curve.

Figure 2

Calibration plot of the Global Registry of Acute Coronary Events risk score for in-hospital mortality (blue) and 1-year mortality (green) in the validation cohort. Patients were divided into deciles based on the predicted risk of mortality, each data point represents one decile. The dashed line shows absolute agreement between the observed and predicted rates.

Figure 3

All-cause mortality according to treatment status in the low-risk, intermediate- and high-risk categories.