The early diagnosis of acute coronary syndrome (ACS) remains challenging, and a considerable proportion of patients are diagnosed with “possible” ACS on admission. The Global Registry of Acute Coronary Events (GRACE/GRACE 2 ) and Canadian Registry of Acute Coronary Events (CANRACE) enrolled 16,618 Canadian patients with suspected ACS in 1999 to 2008. We compared the demographic and clinical characteristics, use of cardiac procedures, prognostic accuracy of the GRACE risk score, and in-hospital outcomes between patients given an admission diagnosis of “definite” versus “possible” ACS by the treating physician. Overall, 11,152 and 5,466 patients were given an initial diagnosis of “definite” ACS and “possible” ACS, respectively. Patients with a “possible” ACS had higher GRACE risk score (median 130 vs 125) and less frequently received aspirin, clopidogrel, heparin, or β blockers within the first 24 hours of presentation and assessment of left ventricular function, stress testing, cardiac catheterization, and percutaneous coronary intervention (all p <0.05). Patients with “possible” ACS had greater rates of in-hospital myocardial infarction (9.0% vs 2.0%, p <0.05) and heart failure (12% vs 8.9%, p <0.05). The GRACE risk score demonstrated excellent discrimination for in-hospital mortality in both groups and for the entire study population. In conclusion, compared to patients with “definite” ACS on presentation, those with “possible” ACS had higher baseline GRACE risk scores but less frequently received evidence-based medical therapies within 24 hours of admission or underwent cardiac procedures during hospitalization. The GRACE risk score provided accurate risk assessment, regardless of the initial diagnostic impression.
Acute coronary syndrome (ACS) encompasses a broad spectrum of clinical conditions and is associated with substantial morbidity and mortality. Despite a detailed history, electrocardiographic findings, and sensitive biomarkers, the early diagnosis of ACS remains challenging, and a considerable proportion of patients are diagnosed with “possible” ACS on admission. The ACS practice guidelines recommend accurate risk stratification using validated risk scores (e.g., Global Registry of Acute Coronary Events [GRACE], Thrombolysis In Myocardial Infarction [TIMI], and the Platelet IIb/IIIa in Unstable Angina: Receptor Suppression Using Integrilin Therapy [PURSUIT] ) to target patients who will derive the greatest absolute benefit from early and intensive therapies. Patients with suspected ACS could also require serial biomarker and electrocardiographic monitoring, stress testing, or coronary angiography to confirm or refute the diagnosis of ACS. It is plausible that the treating physician’s initial diagnostic impressions of “possible” versus “definite” ACS influence the timely delivery of evidence-based therapies, which could subsequently affect outcomes. However, the management patterns and outcomes in relation to physicians’ initial diagnostic impressions have not been well characterized. Accordingly, the present study aimed to (1) examine the clinical characteristics, management patterns, predictive accuracy of the GRACE risk score, and outcomes in relation to “definite” versus “possible” ACS diagnosis at hospital admission, and (2) determine the predictors (if any) for evolution toward a final discharge diagnosis of ACS in patients with an initial “possible” ACS.
Methods
The GRACE was a prospective, multicenter, observational study of patients admitted to hospitals for suspected ACS. An expanded version of the GRACE project, GRACE 2 , enrolled patients with ACS from 2003 to 2007 and was continued in Canada as the Canadian Registry of Acute Coronary Events (CANRACE) in 2008. The design of these studies has been previously described. The inclusion criteria for all 3 registries were (1) age ≥18 years, (2) symptoms compatible with ACS within 24 hours of hospital presentation, and (3) at least one of abnormal cardiac biomarkers, electrocardiographic changes, and/or documented history of coronary artery disease. These registries excluded patients with ACS accompanied by a serious concurrent illness, such as trauma or gastrointestinal bleeding. All patients enrolled in Canada were included in the present study, which focused on the clinical characteristics and treatment of patients initially diagnosed with “possible” ACS compared to patients initially diagnosed with “definite” ACS on admission by the treating physician. “Definite” ACS was defined as having a recorded diagnosis of myocardial infarction (MI), whether ST-segment elevation MI or non–ST-segment elevation MI, or “unstable angina.” “Possible” ACS was defined as a recorded working diagnosis of “rule out MI,” “chest pain,” or “other cardiac.” A final ACS diagnosis was defined as having a recorded discharge diagnosis of ST-segment elevation MI, non–ST-segment elevation MI, or unstable angina. A final non-ACS discharge diagnosis was defined as “other cardiac” or “other.” Patients with incomplete data on the initial diagnosis were excluded from the present analysis (n = 531). All data were recorded on standardized report forms by the local study coordinator or responsible physician and submitted to the centralized databases. Ethics approval was obtained at each hospital, and all patients provided informed consent.
The demographic and clinical characteristics and data contributing to the GRACE risk score (e.g., age, heart rate, systolic blood pressure, Killip class, ST deviation, cardiac arrest, serum creatinine, and cardiac biomarker status on admission) were compared between the 2 groups. The in-hospital medical, diagnostic, and invasive management strategies were also compared. The in-hospital outcomes included in-hospital death, myocardial (re-)infarction (MI) defined as MI >24 hours after presentation or re-infarction, heart failure or pulmonary edema, cardiogenic shock, sustained ventricular tachycardia, and recurrent ischemia. We also evaluated the predictive performance of the GRACE risk score in these 2 patient groups.
Continuous variables are presented as the median and interquartile range and categorical variables as percentages. Group comparisons were made using chi-square tests for categorical variables and the Mann-Whitney U test for continuous variables. Multivariable logistic regression analysis was performed to determine the odds ratios and 95% confidence intervals of the independent predictors toward a final discharge diagnosis (ACS vs non-ACS) in patients with “possible” ACS. The logistic regression model included the individual components of the GRACE risk score, male gender, history of diabetes, history of MI, history of angina, and T-wave inversion, because these characteristics might be suggestive of ACS. Receiver operating characteristic curves and c-statistics were calculated to assess the utility of the GRACE risk score for predicting in-hospital mortality for both groups and the entire study population. A 2-tailed p value <0.05 was used for statistical significance. Statistical analyses were conducted using SPSS, version 15.0 (SPSS, Chicago, Illinois).
Results
A total of 16,618 patients were included in the present study: 11,152 patients with “definite” ACS and 5,466 patients with “possible” ACS on admission. Their baseline characteristics are listed in Table 1 . The patients with “definite” ACS were more likely to be smokers, have a history of a positive stress test, have abnormal electrocardiographic findings, and have positive biomarkers at 24 hours. Patients with “possible” ACS were more likely to be men, be older, and have a greater prevalence of cardiovascular risk factors. These patients also had a longer duration of symptoms and greater Killip class, creatinine, and GRACE risk scores.
| Variable | “Possible” ACS (n = 5,466) | “Definite” ACS (n = 11,152) | p Value |
|---|---|---|---|
| Men | 62% | 68% | <0.001 |
| Age (yrs) | <0.001 | ||
| Median | 69 | 66 | |
| Interquartile range | 58–78 | 56–76 | |
| History of angina pectoris | 45% | 44% | 0.46 |
| Transient ischemic attack/stroke | 10% | 8.5% | 0.001 |
| Diabetes mellitus | 30% | 26% | <0.001 |
| Current smoker | 22% | 27% | <0.001 |
| History of myocardial infarction | 35% | 32% | <0.001 |
| History of heart failure | 15% | 9.2% | <0.001 |
| Previous percutaneous coronary intervention | 19% | 18% | 0.09 |
| Previous coronary bypass | 13% | 12% | 0.52 |
| Previous positive stress test | 8.0% | 13% | <0.001 |
| Peripheral artery disease | 8.6% | 8.5% | 0.81 |
| Hypertension | 62% | 59% | <0.001 |
| Dyslipidemia | 52% | 54% | 0.03 |
| Atrial fibrillation | 13% | 8.2% | <0.001 |
| Symptom onset to presentation (min) | <0.001 | ||
| Median | 163 | 145 | |
| Interquartile range | 74–427 | 69–357 | |
| Killip class | <0.001 | ||
| I | 79% | 86% | |
| II | 13% | 10% | |
| III | 7.6% | 3.8% | |
| IV | 0.4% | 0.4% | |
| Systolic blood pressure (mm Hg) | 0.85 | ||
| Median | 143 | 143 | |
| Interquartile range | 124–162 | 124–161 | |
| Diastolic blood pressure (mm Hg) | 0.36 | ||
| Median | 80 | 80 | |
| Interquartile range | 69–91 | 69–91 | |
| Heart rate (beats/min) | <0.001 | ||
| Median | 80 | 76 | |
| Interquartile range | 68–96 | 65–91 | |
| Index electrocardiographic findings | |||
| ST deviation | 38% | 48% | <0.001 |
| ST elevation (≥1 mm) | 11% | 30% | <0.001 |
| ST depression (≥1 mm) | 30% | 28% | 0.03 |
| Significant Q waves | 5.6% | 11% | <0.001 |
| T-wave inversion | 16% | 20% | <0.001 |
| Initial creatinine (μmol/L) | <0.001 | ||
| Median | 94 | 92 | |
| Interquartile range | 78–118 | 78–111 | |
| Positive initial biomarkers | 47% | 46% | 0.22 |
| Positive biomarkers by 24 h | 68% | 71% | 0.001 |
| Initial troponin level (ng/ml) | <0.001 | ||
| Median | 0.10 | 0.10 | |
| Interquartile range | 0.03–0.39 | 0.04–0.70 | |
| 24-hour Troponin level (ng/ml) | <0.001 | ||
| Median | 0.55 | 2.20 | |
| Interquartile range | 0.10–3.89 | 0.20–15.7 | |
| Cardiac arrest | 1.0% | 1.6% | 0.002 |
| Global Registry of Acute Coronary Events risk score | <0.001 | ||
| Median | 130 | 125 | |
| Interquartile range | 104–162 | 102–153 |
Before presentation, patients with “possible” ACS had greater rates of chronic cardiac medication use ( Table 2 ). Within 24 hours of presentation ( Table 3 ), patients with “definite” ACS were more likely to receive aspirin, clopidogrel, anticoagulation, glycoprotein IIb/IIIa inhibitors, β blockers, and intravenous inotropes. Significantly more patients with “definite” ACS underwent an assessment of left ventricular function during their hospitalization, exercise stress testing, and cardiac catheterization with shorter times to cardiac catheterization and percutaneous coronary intervention.
| Drug | “Possible” ACS (n = 5,466) | “Definite” ACS (n = 11,152) | p Value |
|---|---|---|---|
| Aspirin | 45% | 42% | 0.01 |
| Warfarin | 9.8% | 5.9% | <0.001 |
| Clopidogrel | 12% | 11% | 0.06 |
| Ticlopidine/clopidogrel | 13% | 11% | 0.01 |
| Angiotensin-converting enzyme inhibitor | 33% | 30% | <0.001 |
| β Blocker | 36% | 34% | 0.01 |
| Calcium channel blocker | 20% | 21% | 0.16 |
| Diuretic | 27% | 22% | <0.001 |
| Nitrate | 22% | 21% | 0.18 |
| Statin | 41% | 39% | 0.02 |
| Variable | “Possible” ACS (n = 5,466) | “Definite” ACS (n = 11,152) | p Value |
|---|---|---|---|
| Medication use within first 24 h | |||
| Aspirin | 89% | 92% | <0.001 |
| Clopidogrel | 61% | 63% | 0.02 |
| Warfarin | 8.6% | 4.0% | <0.001 |
| Unfractionated heparin | 29% | 38% | <0.001 |
| Enoxaparin | 54% | 54% | 0.03 |
| Any heparin | 82% | 88% | <0.001 |
| Glycoprotein IIb/IIIa inhibitor | 5.4% | 11% | <0.001 |
| Angiotensin receptor blocker | 13% | 9.7% | <0.001 |
| Angiotensin-converting enzyme inhibitor | 53% | 54% | 0.21 |
| β Blocker | 76% | 78% | 0.03 |
| Statin | 66% | 66% | 0.97 |
| Nitrates | 68% | 68% | 0.53 |
| Calcium channel blocker | 22% | 21% | 0.02 |
| Intravenous inotropes | 2.4% | 3.2% | 0.004 |
| Noninvasive testing | |||
| Left ventricular function assessment | 62% | 66% | <0.001 |
| Normal | 51% | 51% | 0.01 |
| Mildly diminished | 28% | 30% | |
| Moderately to severely diminished | 21% | 19% | |
| Exercise stress testing | 17% | 21% | <0.001 |
| Invasive management | |||
| Cardiac catheterization | 56% | 61% | <0.001 |
| Interval to cardiac catheterization (days) | <0.001 | ||
| Median | 3 | 2 | |
| Interquartile range | 2–5 | 1–4 | |
| Percutaneous coronary intervention | 29.1% | 34.4% | <0.001 |
| Coronary artery bypass grafting | 4.2% | 3.5% | 0.04 |
| Percutaneous coronary intervention or coronary artery bypass grafting | 33% | 35% | 0.10 |
Of the 5,466 patients with “possible” ACS, 76% received a final discharge diagnosis of ACS. In contrast, 91% of patients with “definite” ACS received a final diagnosis of ACS ( Table 4 ). The “possible” ACS group had a longer length of stay in the hospital. The independent predictors of a final ACS diagnosis in patients with “possible” ACS were older age, Killip class III to IV, initial positive biomarkers, ST-segment deviation, T-wave inversion, and greater systolic blood pressure ( Table 5 ). The independent predictors of a final non-ACS diagnosis were female gender, Killip class II, a history of angina, and a greater heart rate ( Table 6 ).
| Variable | “Possible” ACS (n = 5,466) | “Definite” ACS (n = 11,152) | p Value |
|---|---|---|---|
| Discharge diagnosis | <0.001 | ||
| ST-segment elevation myocardial infarction | 13% | 30% | |
| Non–ST-segment elevation myocardial infarction | 43% | 38% | |
| Unstable angina pectoris | 20% | 23% | |
| Other cardiac | 17% | 6.4% | |
| Noncardiac | 7.1% | 2.7% | |
| Length of stay (days) | 0.01 | ||
| Median | 5 | 5 | |
| Interquartile range | 3–8 | 3–8 |
| Independent Variables | Adjusted OR (95% CI) | p Value |
|---|---|---|
| Age, per decade older | 1.06 (1.00–1.062) | 0.04 |
| Female gender | 0.66 (0.57–0.77) | <0.001 |
| History of diabetes mellitus | 1.14 (0.97–1.33) | 0.12 |
| History of myocardial infarction | 1.09 (0.93–1.29) | 0.28 |
| Angina pectoris | 0.82 (0.70–0.96) | 0.01 |
| Heart rate, per 10 beats/min greater | 0.96 (0.94–0.99) | 0.004 |
| Systolic blood pressure, per 10-mm Hg greater | 1.07 (1.04–1.09) | <0.001 |
| Killip class I | Reference | |
| Killip class II | 0.72 (0.59–0.89) | 0.002 |
| Killip class III–IV | 1.53 (1.12–2.08) | 0.01 |
| Cardiac arrest on presentation | 0.93 (0.43–2.02) | 0.86 |
| Initial positive biomarker status | 1.58 (1.36–1.82) | <0.001 |
| Creatinine, per 10 μmol/L greater | 0.99 (0.98–1.00) | 0.01 |
| ST-segment deviation | 1.45 (1.25–1.69) | <0.001 |
| T-wave inversion | 1.30 (1.06–1.58) | 0.01 |
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
