Global Registry Of Acute Coronary Events Risk Model Research Articles

Machine learning for predicting intrahospital mortality in ST-elevation myocardial infarction patients with type 2 diabetes mellitus

In an era of increasing need for precision medicine, machine learning has shown promise in making accurate acute myocardial infarction outcome predictions. The accurate assessment of high-risk patients is a crucial component of clinical practice. Type 2 diabetes mellitus (T2DM) complicates ST-segment elevation myocardial infarction (STEMI), and currently, there is no practical method for predicting or monitoring patient prognosis. The objective of the study was to compare the ability of machine learning models to predict in-hospital mortality among STEMI patients with T2DM. We compared six machine learning models, including random forest (RF), CatBoost classifier (CatBoost), naive Bayes (NB), extreme gradient boosting (XGBoost), gradient boosting classifier (GBC), and logistic regression (LR), with the Global Registry of Acute Coronary Events (GRACE) risk score. From January 2016 to January 2020, we enrolled patients aged > 18 years with STEMI and T2DM at the Affiliated Hospital of Zunyi Medical University. Overall, 438 patients were enrolled in the study [median age, 62 years; male, 312 (71%); death, 42 (9.5%]). All patients underwent emergency percutaneous coronary intervention (PCI), and 306 patients with STEMI who underwent PCI were enrolled as the training cohort. Six machine learning algorithms were used to establish the best-fit risk model. An additional 132 patients were recruited as a test cohort to validate the model. The ability of the GRACE score and six algorithm models to predict in-hospital mortality was evaluated. Seven models, including the GRACE risk model, showed an area under the curve (AUC) between 0.73 and 0.91. Among all models, with an accuracy of 0.93, AUC of 0.92, precision of 0.79, and F1 value of 0.57, the CatBoost model demonstrated the best predictive performance. A machine learning algorithm, such as the CatBoost model, may prove clinically beneficial and assist clinicians in tailoring precise management of STEMI patients and predicting in-hospital mortality complicated by T2DM.

Plasma desmosine for prediction of outcomes after acute myocardial infarction.

Elastin degradation is implicated in the pathology of vulnerable plaque. Recent studies show promising results for plasma desmosine (pDES), an elastin-specific degradation product, as a marker of cardiovascular disease (CVD) outcomes. The aim of this study was to investigate the potential role of pDES as a marker of clinical outcome in patients with acute myocardial infarction (AMI). In this case-control study, we studied 236 AMI patients: 79 patients who had death and/or myocardial infarction (MI) at 2 years, and 157 patients who did not have an event at 2 years. pDES was measured using a validated liquid chromatography-tandem mass spectrometry method. Association of pDES with adverse outcomes, and the incremental value of pDES to global registry of acute coronary events (GRACE) score for risk stratification was assessed. pDES levels were elevated in patients with the composite outcome of death/MI at 2 years (p = 0.002). Logistic regression analyses showed pDES to be associated with death/MI at 2 years [Odds ratio (OR) 5.99 (95% CI 1.81-19.86) p = 0.003]. pDES remained a significant predictor of death/MI at 2 years even after adjustment for age, sex, history of CVD, revascularisation, blood pressure, medications on discharge, Troponin I, and NT-proBNP levels.[OR 5.60 (95% CI 1.04-30.04) p = 0.044]. In another multivariable model including adjustment for eGFR, pDES was significantly associated with the composite outcome at 6 months, but not at 2 years follow up. DES was also able to reclassify risk stratification for death/MI at 6 months, when added to the GRACE risk model [Net Reclassification Index (NRI) 41.2 (95% CI 12.0-70.4) p = 0.006]. pDES concentrations predict clinical outcomes in patients with AMI, demonstrating its potential role as a prognostic marker in AMI.

Ethnicity-dependent performance of the Global Registry of Acute Coronary Events risk score for prediction of non-ST-segment elevation myocardial infarction in-hospital mortality: nationwide cohort study.

The Global Registry of Acute Coronary Events (GRACE) score was developed to evaluate risk in patients with the acute coronary syndrome with or without ST-segment elevation. Little is known about its performance at predicting in-hospital mortality for ethnic minority patients. We identified 326 160 admissions with non-ST-segment elevation myocardial infarction (NSTEMI) in the Myocardial Infarction National Audit Project (MINAP), 2010-17, including White (n = 299 184) and ethnic minorities (excluding White minorities) (n = 26 976). We calculated the GRACE score for in-hospital mortality and assessed ethnic group baseline characteristics by low, intermediate and high risk. The performance of the GRACE risk score was estimated by discrimination [area under the receiver operating characteristic curve (AUC)] and calibration (calibration plots). Ethnic minorities presented younger and had increased prevalence of cardiometabolic risk factors in all GRACE risk groups. The GRACE risk score for White [AUC 0.87, 95% confidence interval (CI) 0.86-0.87] and ethnic minority (AUC 0.87, 95% CI 0.86-0.88) patients had good discrimination. However, whilst the GRACE risk model was well calibrated in White patients (expected to observed (E : O) in-hospital death rate ratio 0.99; slope 1.00), it overestimated risk in ethnic minority patients (E : O ratio 1.29; slope: 0.94). The GRACE risk score provided good discrimination overall for in-hospital mortality, but was not well calibrated and overestimated risk for ethnic minorities with NSTEMI.

Modern NCDR and ACTION risk models outperform the GRACE model for prediction of in-hospital mortality in acute coronary syndrome in a German cohort

Background and purposeRisk prediction with the Global Registry of Acute Coronary Events (GRACE) risk model is guideline-recommended in acute coronary syndrome (ACS) patients. However, the performance of more contemporary scores derived from ACTION (Acute Coronary Treatment and Intervention Outcomes Network) and National Cardiovascular Data (NCDR) registries remains incompletely understood. We aimed to compare these models in German ACS patients. Methods and resultsA total of 1567 patients with (Non-)ST-segment elevation myocardial infarction (NSTEMI: 1002 patients, STEMI: 565 patients) undergoing invasive management at University Hospital Düsseldorf (Germany) from 2014 to 2018 were included. Overall in-hospital mortality was 7.5% (NSTEMI 3.7%, STEMI 14.5%). Parameters for calculation of GRACE 1.0, GRACE 2.0, ACTION and NCDR risk models and in-hospital mortality were assessed and risk model performance was compared. The GRACE 1.0 risk model for prediction of in-hospital mortality discriminated risk superior (c-index 0.84) to its successor GRACE 2.0 (c-index 0.79, pGRACE1.0vsGRACE2.0 = 0.0008). The NCDR model performed best in discrimination of risk in ACS overall (c-index 0.89; pACTIONvsNCDR < 0.0001; pGRACEvsNCDR < 0.0001) and showed superior performance compared to GRACE in NSTEMI and STEMI subgroups (pGRACEvsNCDR both < 0.02). ACTION and GRACE risk models performed comparable to each other (both c-index 0.84, pGRACEvsACTION = 0.68), with advantages for ACTION in NSTEMI patients (c-index 0.87 vs. 0.84 (GRACE); pGRACEvsACTION = 0.02). ACTION and GRACE 2.0 showed the most accurate calibration of all models. ConclusionsIn a contemporary German patient population with ACS, modern NCDR and ACTION risk models showed superior performance in prediction of in-hospital mortality compared to the gold-standard GRACE model.

Nocturnal respiratory rate as a predictor of mortality in patients with acute coronary syndrome

BackgroundPatients with acute coronary syndrome (ACS) are at risk especially in the period shortly after the event. Alterations in respiratory control have been associated with adverse prognosis. The aim of...

Prognostic significance of low QRS voltage on the admission electrocardiogram in acute coronary syndromes

PurposeTo examine the prognostic significance of low QRS voltage in a large contemporary cohort of patients with a broad spectrum of acute coronary syndromes (ACS). Methods12409 patients with STEMI or NSTE-ACS from the Global Registry of Acute Coronary Events (GRACE) and Canadian ACS I registries were stratified based on the presence of low QRS voltage (<0.5mV in all limb leads and <1.0mV in all precordial leads) on the admission ECG. We performed multivariable logistic regression to assess the independent association between low voltage and in-hospital and 6-month mortality, and tested for its interaction with ST-segment deviation for these outcomes. ResultsPatients with low voltage (3.2%) had higher GRACE risk scores, rates of prior myocardial infarction, and pathological Q waves, with less prevalent ST-segment deviation and ST-segment depression. They had worse left ventricular function and higher unadjusted rates of in-hospital and 6-month mortality. After adjustment for established prognosticators in the GRACE risk models in multivariable analysis, low voltage was independently associated with higher in-hospital mortality (adjusted OR 1.77, 95% CI 1.13–2.78, P=0.013) and mortality/re-infarction (adjusted OR 1.42, 95% CI 1.05–1.93, P=0.023), but not 6-month mortality (adjusted OR 1.25, 95% CI 0.85–1.84, P=0.27). There was no significant interaction between low voltage and ST-segment deviation for any endpoint (interaction P>0.10 for all endpoints). ConclusionsLow QRS voltage was associated with previous myocardial infarction and adverse hemodynamic variables at presentation. After adjusting for other prognosticators, low voltage independently predicted higher in-hospital mortality. This increased risk was not modulated by concomitant ST-segment deviation.

IN-HOSPITAL MANAGEMENT AND OUTCOMES OF ACUTE CORONARY SYNDROMES IN RELATION TO PRIOR HISTORY OF HEART FAILURE

Acute heart failure (HF) complicating acute coronary syndromes (ACS) portends an ominous outcome. Previous history of HF is a predictor of acute decompensated HF in ACS, but the incremental prognostic significance of previous HF has not been well studied. Accordingly, we evaluated the baseline characteristics, management patterns, and clinical outcomes in relation to prior history of HF across a broad spectrum of ACS patients. Our study population consisted of ACS patients from 53 Canadian sites in the Global Registry of Acute Coronary Events (GRACE), the subsequently expanded GRACE2, and Canadian Registry of Acute Coronary Events (CANRACE) between 1999 and 2008. Patients were stratified into two groups based on history of prior HF. Multivariable logistic regression analysis was performed to assess the impact of prior history of heart failure on in-hospital mortality, adjusting for the components of the GRACE risk model. Of the 13937 eligible patients (mean age 66±13 years, 33% female, and 28.3% with ST-elevation MI), 1498 (10.7%) patients had a history of HF. Those with prior HF tended to be older, female, with higher prevalence of diabetes and previous myocardial infarction; they also had lower systolic blood pressure, higher heart rate, worse Killip class, higher creatinine, and higher GRACE risk score on presentation (all p<0.001). Previous HF was also associated with significantly worse left ventricular systolic function (p<0.001). The table summarizes the in-hospital management and outcomes. After adjusting for components of the GRACE risk model including Killip class on presentation, prior HF remained an independent predictor of in-hospital mortality (OR 1.48, 95% CI 1.08-2.03, p=0.02). Prior HF was associated with other high risk features in ACS including HF on presentation, and other adverse outcomes in hospital. After adjusting for other prognosticators, prior HF remained an independent predictor of in-hospital mortality. However, patients with prior HF were less likely to receive evidence-based medical and invasive therapies, suggesting an opportunity for more intensive treatment to improve outcomes.

Clinical characteristics and outcomes of acute coronary syndrome patients with left anterior hemiblock

ObjectiveWe aimed to study the relationships between left anterior hemiblock (LAHB) and the patient characteristics, management, and clinical outcomes in the setting of acute coronary syndromes (ACS).MethodsAdmission ECGs of patients...

Anaemia to predict outcome in patients with acute coronary syndromes

Owing to the heterogeneous population of patients with acute coronary syndromes (ACS), risk stratification with tools such as the GRACE risk score is recommended to guide therapeutic management and improve outcome. To evaluate whether anaemia refines the value of the GRACE risk model to predict midterm outcome after an ACS. A prospective registry of 1064 ACS patients (63 ± 14 years; 73% men; 57% ST-segment elevation myocardial infarction [MI]) was studied. Anaemia was defined as haemoglobin less than 13 mg/dL in men or less than 12 mg/dL in women. The primary endpoint was 6-month death or rehospitalization for MI. The primary endpoint was reached in 132 patients, including 68 deaths. Anaemia was associated with adverse clinical outcomes (hazard ratio 3.008, 95% confidence interval 2.137-4.234; P<0.0001) in univariate analysis and remained independently associated with outcome after adjustment for the Global Registry of Acute Coronary Events (GRACE) risk score (hazard ratio 2.870, 95% confidence interval 1.815-4.538; P<0.0001). Anaemia provided additional prognostic information to the GRACE score as demonstrated by a systematic improvement in global model fit and discrimination (c-statistic increasing from 0.633 [0.571;0.696] to 0.697 [0.638;0.755]). Subsequently, adding anaemia to the GRACE score led to reclassification of 595 patients into different risk categories; 16.5% patients at low risk (≤ 5% risk of death or rehospitalization for MI) were upgraded to intermediate (>5-10%) or high risk (>10%); 79.5% patients at intermediate risk were reclassified as low (55%) or high risk (24%); and 45.5% patients at high risk were downgraded to intermediate risk. Overall, 174 patients were reclassified into a higher risk category (17.3%) and 421 into a lower risk category (41.9%). Anaemia provides independent additional prognostic information to the GRACE score. Combining anaemia with the GRACE score refines its predictive value, which often overestimates the risk.

Comparative prognostic value of T-wave inversion and ST-segment depression on the admission electrocardiogram in non–ST-segment elevation acute coronary syndromes

ST-segment depression (STD) is predictive of adverse outcomes in non-ST-segment elevation acute coronary syndromes (NSTE-ACS), but there are conflicting data on the incremental prognostic value of T-wave inversions (TWIs) on the admission electrocardiogram. Admission electrocardiograms of 7,343 patients with NSTE-ACS from the Global Registry of Acute Coronary Events (GRACE) and ACS I registry were independently analyzed at a core laboratory and stratified by TWI and STD status. We performed multivariable analyses to determine the independent prognostic significance of TWI and tested for interaction between TWI and STD for adverse outcomes. Patients with TWI and/or STD had a higher prevalence of cardiovascular risk factors, higher Killip class, and higher GRACE risk scores. Among the 2,708 patients with available angiographic data, rates of 3-vessel or left main disease were similar between patients with TWI and those without TWI/STD. After adjusting for other established prognosticators, TWI did not independently predict in-hospital (adjusted odds ratio 1.03, 95% CI 0.75-1.42, P = .85) or 6-month mortality (adjusted odds ratio 1.02, 95% CI 0.80-1.30, P = .88); STD remained a strong independent predictor. There was no interaction between TWI and STD for these outcomes. No contiguous lead groups or cumulative number of leads with TWI provided independent prognostic information. TWI is associated with other high-risk clinical features but is not an independent predictor of adverse short- and long-term mortality in NSTE-ACS. T-wave inversion does not provide additional prognostication beyond the GRACE risk model, and its concomitant presence does not alter the prognostic value of STD.

Prognostic significance of electrocardiographic-determined left ventricular hypertrophy and associated ST-segment depression in patients with non–ST-elevation acute coronary syndromes

Left ventricular hypertrophy (LVH) is frequently associated with ST depression (STD) on the electrocardiogram (ECG), a so-called strain pattern. Although STD is a well-established adverse prognosticator in non-ST-elevation acute coronary syndrome (NSTE-ACS), the relative prognostic importance of LVH and associated STD has not been elucidated. A total of 7,761 patients with NSTE-ACS in the Global Registry of Acute Coronary Events (GRACE) and ACS-I registries had admission ECGs analyzed at a core laboratory. Left ventricular hypertrophy (determined by Sokolow-Lyon and/or Casale criteria) was observed in 296 (3.8%) patients. We examined the independent association between LVH (determined by the admission ECG) and outcomes in relation to STD. Patients with LVH were older, had more comorbidities and STD, and presented with a higher Killip class. They were less likely to undergo cardiac catheterization (43.1% vs 51.2%, P = .006) and percutaneous coronary intervention (18.3% vs 24.6%, P = .014). Patients with LVH had higher unadjusted mortality at 6 months (10.5% vs 7.1%, P = .038), but similar rates of in-hospital mortality (4.1% vs 3.4%, P = .54) and reinfarction (7.1% vs 7.6%, P = .75). Patients with LVH were more likely to have heart failure in-hospital (21.8% vs 11.8%, P < .001). Among LVH patients, degree of quantitative STD did not predict higher short- or long-term mortality, but was associated with in-hospital heart failure. Multivariable analysis adjusting for other clinical prognosticators of the GRACE risk models revealed that LVH was not a significant independent predictor of in-hospital mortality (adjusted odds ratio = 0.75, 95% CI 0.40-1.41, P = .37) or 6-month mortality (adjusted odds ratio = 0.83, 95% CI 0.52-1.35, P = .44). In contrast, STD remained a strong independent predictor of adverse outcomes. There was no significant interaction between STD and LVH. Across the broad spectrum of NSTE-ACS, LVH is associated with adverse prognostic factors including STD. Electrocardiographic-determined LVH provides no significant additional prognostic utility beyond comprehensive risk assessment using the GRACE risk score. The adverse prognosis associated with LVH in NSTE-ACS may be attributable to other prognosticators such as STD.

Validity of a risk-prediction tool for hospital mortality: The Global Registry of Acute Coronary Events

The Global Registry of Acute Coronary Events (GRACE) risk model provides a simple method for determining the probability of hospital death in acute coronary syndrome (ACS). The aim of this study was to explore the impact of modeling techniques on the risk model when generating predictions. Patients with ACS (n = 48,023) with or without ST-segment elevation myocardial infarction (STEMI) were enrolled (123 hospitals, 14 countries) between April 1999 and June 2006. The original GRACE model did not include terms to account for possible differences in outcomes between patients with STEMI, non-STEMI, and unstable angina, nor did it account for changing risk across continuous measures. In this cohort, the influence on outcome of region of hospitalization and cardiac arrest at presentation changed over the 7-year study. Other interactions included previous percutaneous coronary intervention and age with type of ACS. However, these interactions were insufficient to affect the final risk score. The same variables as in the original score comprise the new score. Inclusion of nonlinearity and differential effects did little to change the model's discrimination but influenced predictions for patients at extremes of risk. Irrespective of the inclusion of nonlinear and interaction terms, the updated GRACE risk model provides an excellent means to discriminate risk of death in patients with ACS and can be used as a simple nomogram to estimate risk in patients seen in clinical practice.

Usefulness of Quantitative Versus Qualitative ST-Segment Depression for Risk Stratification of Non-ST Elevation Acute Coronary Syndromes in Contemporary Clinical Practice

This aim of this study was to assess the clinical utility of quantitative ST-segment depression (STD) for refining the risk stratification of non-ST elevation acute coronary syndromes in the prospective, multinational Global Registry of Acute Coronary Events (GRACE). Quantitative measurements of STD on admission electrocardiograms were evaluated independently by a core laboratory, and their predictive value for in-hospital and cumulative 6-month mortality was examined. Although more severe STD is a marker of increased short- and long-term mortality, it is also associated with higher risk clinical features and biomarkers. Thus, after adjustment for these clinically important predictors, quantitative STD does not provide incremental prognostic value beyond simple dichotomous evaluation for the presence of STD. Furthermore, adopting quantitative instead of the prognostically proven qualitative evaluation of STD does not improve risk discrimination afforded by the validated GRACE risk models. In conclusion, the findings do not support the quantification of STD in routine clinical practice beyond simple evaluation for the presence of STD as an integral part of comprehensive risk stratification using the GRACE risk score.

Relationship of ST elevation in lead aVR with angiographic findings and outcome in non–ST elevation acute coronary syndromes

Limited data suggest that ST elevation (ST elevation) in aVR is associated with higher mortality and more extensive coronary artery disease in the setting of non-ST elevation acute coronary syndromes (ACS). In the prospective Global Registry of Acute Coronary Events (GRACE) electrocardiographic substudy, the admission electrocardiograms were analyzed by a blinded core laboratory. We performed multivariable analysis to determine (1) the independent prognostic significance of ST elevation in aVR and (2) its association with significant (> or = 50% stenosis) left main or 3-vessel disease (LM/3-vd). Among 5064 patients with non-ST elevation ACS, 4696 had no ST elevation in aVR, 292 (5.8%) had minor (0.5-1 mm) ST elevation in aVR, and 76 (1.5%) had major (>1 mm) ST elevation in aVR; their in-hospital mortality rates were 4.2%, 6.2%, and 7.9%, respectively (P for trend =.03). At 6 months follow-up, the cumulative mortality rates were 7.6%, 12.7%, and 18.3%, respectively (log-rank P for trend <.001). However, minor and major ST elevation in aVR were not independent predictors of in-hospital or 6-month death after adjusting for other validated prognosticators in the GRACE risk model. Of the 2416 patients without prior coronary bypass surgery who underwent cardiac catheterization, the prevalence of LM/3-vd was 26.1%, 36.2%, and 55.9% for the groups with no, minor, and major ST elevation in aVR, respectively (P for trend <.001). After adjusting for other clinical characteristics, major ST elevation in aVR remained an independent predictor of LM/3-vd (adjusted odds ratio, 2.68; 95% confidence interval, 1.29-5.58; P = .008). ST elevation in aVR is less prevalent than reported in previous smaller studies. Although it is associated with higher unadjusted in-hospital and 6-month mortality, it does not provide incremental prognostic value beyond comprehensive risk stratification using the validated GRACE risk model. However, ST elevation greater than 1 mm in aVR may be useful in the early identification of LM/3-vd in ACS patients with ST depression.

Global Registry of Acute Coronary Events (GRACE) hospital discharge risk score accurately predicts long-term mortality post acute coronary syndrome

The Global Registry of Acute Coronary Events (GRACE) hospital discharge risk score (GRACE score) developed from a multinational registry involving all subsets of acute coronary syndrome (ACS) predicted 6-month survival. There is currently no validated risk model to predict mortality beyond 6 months. Of the 1143 consecutive patients with ACS admitted to coronary care unit in 2000 to 2002 (mean age, 64.9 +/- 12.6 years), 39% had ST-elevation myocardial infarction, 39% had non-ST-elevation infarction, and 22% had unstable angina. The mortality was 7.5% during index admission, 12.1% at 6 months, 14.8% at 1 year, 18.7% at 2 years, 25.0% at 3 years, and 39.2% at 4 years. The GRACE hospital discharge risk score calculated for 1057 hospital survivors discriminated survival from death at 6 months (C index, 0.81), 1 year (C index, 0.82), 2 years (C index, 0.81), 3 years (C index, 0.81), and 4 years (C index, 0.80). The risk score worked for all 3 subsets of ACS at all time points, with C index >0.75 in all analyses. A separate multivariable mortality model for these 1057 patients over the 4-years follow-up period identified 10 independent predictors of mortality. Seven were in the GRACE risk model (age, history of ischemic heart disease, heart failure, increased heart rate on admission, serum creatinine level, evidence of myonecrosis, not receiving in-hospital percutaneous coronary intervention). The GRACE postdischarge risk score contains relevant prognostic factors and accurately discriminate survivors from nonsurvivors over the longer term (up to 4 years) in all subsets of ACS patients.