Combined use of the GRACE ACS risk score and comorbidity indices to increase the effectiveness of hospital mortality risk assessment in patients with acute coronary syndrome

Abstract

Aim. To assess the possibilities of using comorbidity indices together with the GRACE (Global Registry of Acute Coronary Events) scale to assess the risk of hospital mortality in acute coronary syndrome (ACS). Materials and methods. The registry study included 2,305 patients with ACS. The frequency of coronary angiography was 54.0%, percutaneous coronary intervention (PCI) 26.9%. Hospital mortality with ACS was 4.8%, with myocardial infarction 9.4%. All patients underwent a comorbidity assessment according to the CIRS system (Cumulative Illness Rating Scale), according to the CCI (Charlson Comorbidity Index) and the CDS (Chronic Disease Score) scale, according to their own scale, which is based on the summation of 9 diseases (diabetes mellitus, atrial fibrillation, stroke, arterial hypertension, obesity, peripheral atherosclerosis, thrombocytopenia, anemia, chronic kidney disease). All patients underwent a mortality risk assessment using the GRACE ACS Risk scale. Results. It was found that the CDS and CIRS indices are not associated with the risk of hospital mortality. With CCI3, the frequency of death outcomes increased from 4.1 to 6.1% (2=4.12, p=0.042). With an increase in the severity of comorbidity from minimal (no more than 1 disease) to severe (4 or more diseases) according to its own scale, hospital mortality increased from 1.2 to 7.4% (2=23.8, p0.0001). In contrast to other scales of comorbidity, our own model more efficiently estimates the hospital prognosis both in the conservative treatment group (2=8.0, p=0.018) and in the PCI group (2=28.5, p=0.00001). It was in the PCI subgroup that the comorbidity factors included in their own model made it possible to increase the area under the ROC curve of the GRACE scale from 0.80 (0.740.87) to 0.90 (0.850.95). Conclusion. CCI and its own comorbidity model, but not CDS and CIRS, are associated with the risk of hospital mortality. The model for assessing comorbidity on a 9-point scale, but not CCI, CDS and CIRS, can significantly improve the predictive value of the GRACE scale.