Decline in Rates of Death and Heart Failure in Acute Coronary Syndromes, 1999-2006 FREE

Randomized trials provide robust evidence for the impact of pharmacological and interventional treatments in patients with ST-segment elevation and non–ST-segment elevation acute coronary syndromes (NSTE ACS), leading to changes in practice guidelines.^1- 4 However, the extent and time course of changes in clinical practice are uncertain, and it is unknown whether such changes are associated with improved outcome. Previous studies have documented substantial gaps between guideline recommendations and clinical practice.^5- 7 Thus, there is a clinical priority to determine the extent to which evidence is applied in practice, whether this is changing over time, and whether such changes are associated with improved outcomes.

Few studies in acute coronary disease offer a sufficiently long sampling interval with sufficiently robust sampling techniques to reveal changes in practice and outcome over time.^8- 10 The Global Registry of Acute Coronary Events (GRACE)⁷ is the only large-scale, multinational, observational study of the spectrum of patients hospitalized with an ACS, with continuous recruitment of patients for more than 6 years. Results from earlier studies suggest that hospital characteristics, access to resources, and geographic factors influence uptake of new therapies into practice.^7,11 The Can Rapid Risk Stratification of Unstable Angina Patients Suppress Adverse Outcomes With Early Implementation of the ACC/AHA Guidelines (CRUSADE)¹¹ quality improvement initiative involves centers throughout the United States and provides valuable insights into selected care processes and patient outcomes as measures of hospital quality.

A total of 62 935 patients have been enrolled in GRACE since it was launched in 1999, and there is a sufficiently large sample size (16 814 patients with ST-segment elevation myocardial infarction [STEMI]; 27 558 with NSTE ACS; and 2067 in-hospital deaths) and sufficient study duration (6.5 years of recruitment) to define changes in management and outcome. The hypothesis for this analysis is that changes in hospital management of patients with ACS are associated with improvements in clinical outcome, and that these changes are independent of the risk status of the study population on presentation to hospital.

Full details of the GRACE methods have been published.^12- 13 To be eligible, patients (≥18 years) had to be admitted for ACS as a presumptive diagnosis and have at least 1 of the following: electrocardiographic changes consistent with ACS, serial increases in serum biochemical markers of cardiac necrosis, documented coronary artery disease, or both. The qualifying patient with ACS must not have been precipitated by significant noncardiovascular comorbidity. Where required, study investigators received approval from their local hospital ethics or institutional review board for the conduct of this study and signed informed consent for follow-up contact was obtained from the patients at enrollment.

The aim was to enroll an unselected population of patients with an ACS, irrespective of geographic region. Sites were encouraged to recruit the first 10 to 20 consecutive eligible patients each month, and regular audits were performed. Data were collected by trained study coordinators using standardized case report forms. Demographic characteristics, medical history, presenting symptoms, duration of prehospital delay, biochemical and electrocardiographic findings, treatment practices, and outcomes were collected. Patients with a non-ACS discharge diagnosis and patients transferred into a GRACE site from a non-GRACE site were excluded from the analysis. All cases were assigned to 1 of the following categories: STEMI (including left bundle-branch block), non-STEMI, unstable angina, and other cardiac or noncardiac diagnoses.¹³ Hospital-specific feedback was provided to each center on a quarterly basis.

The primary outcomes for this study were in-hospital death, recurrent myocardial infarction, heart failure, stroke, and cardiogenic shock. At approximately 6 months after hospital discharge, patients were followed up for death, stroke or myocardial infarction. Of the patients who were alive at discharge from hospital, 2435 (16%) of 15 569 of those with STEMI and 3883 (15%) of 26 736 with NSTE ACS did not complete follow up. Standardized definitions of all patient-related variables, diagnoses, and in-hospital outcomes were used¹³ (details available at http://www.outcomes.org/grace). We applied guideline-based criteria for identifying patients eligible for pharmacological treatments or mechanical interventions; specifics are available from the authors on request.

The analysis focuses on the populations of patients diagnosed with STEMI or NSTE ACS (non-STEMI and unstable angina). Data used were collected from July 1, 1999, through December 31, 2006. Data are summarized as frequencies and percentages for categorical variables. Continuous variables are presented as medians and 25th and 75th percentiles. The period of patient enrollment into GRACE was divided into 13 periods of 6 months each. Six-month follow-up data for each patient were reported under the same period as their enrollment. The double-sided Cochran-Armitage test for trend or logistic regression was used to evaluate time trends at a significance level of α = .05. The analysis was performed using SAS software version 9.1 (SAS Institute Inc, Cary, NC).

To account for possible changes in patients' baseline risk characteristics, the probability of in-hospital death was calculated for each patient using the GRACE in-hospital risk score.¹⁴ The average risk score for each time period was then calculated. Although patients with STEMI showed no change in predicted risk of mortality over time, there was a slight increase in risk for those with NSTE ACS. To account for this, adjusted rates were computed and normalized to those of the first time period. To calculate the adjusted rate, a difference between the observed and predicted death rates was computed for each time period.

This study is based on data from 44 372 patients enrolled at 113 hospitals in 14 countries between July 1, 1999, and December 31, 2005, and were followed up for approximately 6 months after discharge. Of these patients, 27 558 were diagnosed with NSTE ACS and 16 814 with STEMI. Patients with STEMI were more often male and were slightly younger than those with NSTE ACS (Table 1). Patients with NSTE ACS had a more complex medical history than patients with STEMI but were less likely to be current smokers and more likely to be past smokers.

Temporal Trends in Pharmacological Management (Class 1A Indications). Temporal trends in medication use and interventions in eligible STEMI patients are shown in Figure 1 and Table 2. Over the study period, the use of aspirin exceeded 94% in eligible patients (Figure 1) and the use of any oral antiplatelet drug (aspirin or thienopyridine) did not change (97% in 1999 to 99% in 2005; 4% increase; 95% confidence interval [CI], 0.5-3.2 for percentage difference in rates). Treatment with β-blockers increased by 11 percentage points (95% CI, 7.6-14), statins by 48 percentage points (95% CI, 45-52), and angiotensin-converting enzyme (ACE) inhibitors or angiotensin II receptor blockers (ARBs) by 22 percentage points (95% CI, 18-25) (Figure 1 and Table 2).

The use of low-molecular-weight heparin increased by 20 percentage points (95% CI, 16-24), whereas the use of unfractionated heparin declined by 19 percentage points (95% CI, −24 to −15) over the same period (Figure 1). Some patients received both low-molecular-weight heparin and unfractionated heparin during hospitalization.¹⁵ Thienopyridine usage increased by 49 percentage points (95% CI, 45-53) overall, and by 56 percentage points (95% CI, 50-61) in the absence of percutaneous coronary intervention (PCI). Glycoprotein IIb/IIIa (GpIIb/IIIA) usage, overall, increased by 24 percentage points (95% CI, 20-27) over the course of the study, but among patients whose care was managed noninvasively, less than 1 in 10, received this therapy (2.3%-9.0%; 6.7 percentage points; 95% CI, 3.3-10). The use of calcium channel blockers, with no guideline-supported indication in STEMI, declined by 9.9 percentage points (95% CI, −13 to −6.9) between 1999 and 2005 (Figure 1B; Table 2).

Use of pharmacological reperfusion declined by 22 percentage points (95% CI, −27 to −17) (Figure 1; Table 2), whereas the rate of primary PCI increased by 37 percentage points (95% CI, 33-41) over the same interval (Figure 1; Table 2). Approximately one third of patients received primary or other PCI in 1999 and this increased to 64% of patients in 2005 (32.5 percentage points; 95% CI, 27-35). The proportion of patients with STEMI who did not receive pharmacological reperfusion therapy or primary rescue or facilitated PCI declined by 5.5 percentage points (95% CI, −9.8 to −1.2; Figure 1; Table 2).

The rates of death in hospital and cardiogenic shock declined in parallel, whereas the rate of stroke did not change significantly (Figure 2). Hospital deaths decreased by 3.9 percentage points (95% confidence interval [CI] −5.3 to −1.9) and cardiogenic shock by 2.4 percentage points (95% CI, −4.3 to −0.5; Table 3). The rate of in-hospital congestive heart failure or pulmonary edema declined by 9.0 percentage points (95% CI, −12 to −6; Table 3). Myocardial infarction diagnosed more than 24 hours after presentation to hospital or recurrent myocardial infarction were collected systematically in the revised case report form July 2002. The rate of these events declined by 1.6 percentage points (95% CI, −3.6 to 0.5). A linear regression of mean risk scores at hospital presentation, over time, was nonsignificant (data available from authors on request).

The rate of myocardial infarction occurring between hospital discharge and 6-month follow-up decreased by 2.8 percentage points (95% CI, −6.4 to 0.9) and stroke by 0.8 percentage points (95% CI, −1.7 to 0.1; Figure 2; Table 3).

Temporal Trends in Pharmacological Therapies (Class 1A Recommendations). Temporal trends in medication use in NSTE ACS patients without contraindications to these drugs are shown in Figure 3 and Table 2. The use of aspirin was high and increased modestly during the study by 2.6 percentage points (95% CI, 1.1-4.1). Similarly, the use of β-blockers increased by 9.9 percentage points (95% CI, 7.5-12). Use of ACE inhibitors (or ARBs) increased by 23 percentage points (95% CI, 20-26), statins by 42 percentage points (95% CI, 40-45), and thienopyridines by 51 percentage points (95% CI, 48-54). In contrast, the use of unfractionated heparin fell by 19 percentage points (95% CI, −22 to −16), whereas that of low-molecular-weight heparin increased by 23 percentage points (95% CI, 20-26; some patients received both low-molecular-weight heparin and unfractionated heparin during hospitalization¹⁵). Treatment with calcium-channel blockers (no class 1 recommendation) declined by 12 percentage points (95% CI, −15 to −9.4).

There was a modest decrease in the use of coronary artery bypass graft (CABG) surgery of 1.7 percentage points (95% CI, −3.2 to −0.3; Figure 3). The frequency of angiography increased markedly by 21 percentage points (95% CI, 18 to 24). The use of PCI increased substantially by 18 percentage points (95% CI, 15-20). Glycoprotein IIb/IIIa inhibitors exhibited an increase, with increasing usage between 1999 (9.0%, 194/2157) and 2003 (23%, 435/1859), then about 20% through 2005 (Figure 3) with an overall increase of 11 percentage points (95% CI, 8.3-13; Table 2). Glycoprotein IIb/IIIa usage in patients whose care was managed noninvasively was low, increasing from 2.7% (49/1792) in 1999 to 9.3% in 2002 and decreasing to 6.1% (123/2013) in 2005.

Risk-adjusted hospital deaths declined by 0.7 percentage points (95% CI, −1.7 to 0.3) in NSTE ACS patients (Figure 4; Table 3). The rate of stroke did not change significantly over time (Figure 4; Table 3). The rate of cardiogenic shock decreased slightly, but significantly, while the rate of congestive heart failure and pulmonary edema decreased by 6.5 percentage points (95% CI, −8.4 to −4.7). Myocardial infarction diagnosed more than 24 hours after presentation to hospital or recurrent myocardial infarction (recorded systematically in case report forms from July 2002) declined by 1.3 percentage points (95% CI, −2.7 to 0.1).

The rate of death between hospital discharge and 6-month follow-up decreased by 1.6 percentage points (95% CI, −3.0 to −0.1). Stroke decreased by 0.7 percentage points (95% CI, −1.4 to 0.1; Figure 4; Table 3).

The mean in-hospital risk score¹⁴ for patients with NSTE ACS showed a significant (P<.01) increasing trend over the study period, from 120 (95% CI, 118-121) in 1999 to 125 (95% CI, 123-127) in 2007. Hence, if all else remained unchanged, the expected risk of death would have risen from 2.5% to 3.2%. Thus, although the unadjusted death rate remained unchanged, when adjusted for risk, deaths decreased by 0.7 percentage points (95% CI, −1.7 to 0.3) from 1999, a relative change of 24% (Figure 4; Table 3).

These data, from the largest multinational observational cohort study of patients with an ACS, demonstrate evidence of a change in practice for both pharmacologic and interventional treatments in patients with either STEMI or NSTE ACS. These changes in practice are accompanied by significant decreases in the rates of in-hospital death, cardiogenic shock, recurrent myocardial infarction, and heart failure in patients presenting with STEMI and significant decreases in hospital death, heart failure, cardiogenic shock, and new myocardial infarction among patients with NSTE ACS. The risk status of patients at presentation with STEMI did not change over the course of the present study, so it is highly plausible that the changes in clinical outcomes are the direct consequence of changes in practice. The hospital death rate for patients with NSTE ACS showed a significant decrease after adjustment for baseline risk. It is disappointing that the proportion of patients who received no reperfusion therapy has not significantly improved over time.

The use of evidence-based therapies and PCI interventions increased in the STEMI population. This included the use of ACE inhibitors (or ARBs) and Gp IIb/IIIa inhibitors in patients undergoing PCI. This increase was matched by a statistically significant decrease in the rates of death, cardiogenic shock, and heart failure or pulmonary edema.

In patients with NSTE ACS, increases were observed in in-hospital treatment with low-molecular-weight heparin, thienopyridines, ACE inhibitors (or ARBs), and statins. Similarly, the rate of cardiac intervention (cardiac catheterization and PCI) increased. In contrast, the use of Gp IIb/IIIa inhibitors increased between 1999 and 2003 but changed little thereafter; a similar pattern was observed for Gp IIb/IIIa inhibitors without PCI. This change in practice may reflect a response to the findings from the Global Use of Strategies to Open Occluded Coronary Arteries-IV Acute Coronary Syndrome (GUSTO-IV ACS,¹⁶ published in June 2001) and guidelines for managing patients with NSTE ACS (published in 2000^17- 18 and 2002^4,19). In the update, the use of Gp IIb/IIIa inhibitors was limited in patients who were to undergo PCI.

In NSTE ACS the rates of new heart failure or pulmonary edema, in hospital, decreased significantly. The risk score for patients with NSTE ACS increased modestly over the course of the study. After adjustment for risk status, the rate of hospital death declined by approximately a quarter. More marked changes may be evident with longer follow-up because the beneficial impact of interventional therapy may be seen with later follow-up.^20- 23 The results of a meta-analysis of trials of interventional therapy suggest an additional hazard of death or myocardial infarction during the hospital phase that is subsequently overcome by postdischarge benefit.^23- 24

Since GRACE was launched in 1999, NSTE ACS guidelines from both the American College of Cardiology/American Heart Association (ACC/AHA)^1- 2 and the European Society of Cardiology (ESC)^3- 4 have been updated to include data from recent clinical trials. In an earlier study from the GRACE registry, data from nearly 13 000 patients with an ACS were analyzed to evaluate the uptake of selected cardiac medications and invasive therapies between 1999 and 2001 and to assess the impact of geographical and hospital characteristics on this pattern.⁷ During this period, new guidelines were released by the ESC and the ACC/AHA, and the results of 2 major clinical trials (Fragmin during Instability in Coronary Artery Disease [FRISC] II,^25- 26 GUSTO IV-ACS²⁷) were published. The results of our earlier study suggest that hospital status, access to resources, and geographical factors, rather than the publication of guidelines, influence the uptake of new therapies into practice.⁷ We also reported increases in the use of evidence-based cardiac medications between July 1999 and December 2001 but no significant decline in death rates was observed over that interval.⁷ Data from the present study (1999 to 2006) demonstrate substantial temporal changes in the use of pharmacological and interventional therapies in ACS and clear evidence for a reduction in new heart failure and in-hospital deaths especially in patients with STEMI. Our multinational data, combined with those from CRUSADE,²⁸ suggest the potential for further reductions in patient outcomes with greater uptake of evidence-based therapies and interventions.

GRACE is the largest multinational observational cohort study to include the spectrum of patients with ACS. GRACE is designed to be representative of regional communities and uses standardized criteria for defining ACS and hospital outcomes, rigorous quality control, and audit measures. The participating clusters reflect regional practices and outcomes but do not necessarily reflect practice for specific countries. Participating hospitals are sent feedback on a 6-month basis, so we cannot determine whether improvements in adherence to evidence-based medication are taking place nationwide or are limited to participating sites. Increasing use of troponin measurement throughout the study may have led to underestimation of the detection of small reinfarctions if troponin was already elevated at presentation and if the patient did not evolve new electrocardiographic changes of myocardial infarction.

These contemporary multinational observational data show substantial changes in the management of patients with STEMI or NSTE ACS studied between 1999 and 2006. The main changes demonstrate a marked increase in interventional therapy in both ST elevation and NSTE ACS and changes in pharmacological therapy, including increases in use of β-blockers, statins, ACE inhibitors (or ARBs), and thienopyridines in patients with an ACS, and Gp IIb/IIIa inhibitors in patients with STEMI. The changes are consistent with trial evidence and national and international guidelines. The risk profile for patients with STEMI has not changed significantly over this interval, whereas the risk for patients with NSTE ACS has increased slightly. This study population is the first demonstration of significant reductions observed in hospital rates of new heart failure in ACS patients, over time, and of reductions in mortality.