Author + information
- Received April 7, 2015
- Accepted May 19, 2015
- Published online August 4, 2015.
- Robert A. Henderson, DM∗∗ (, )
- Christopher Jarvis, MSc†,
- Tim Clayton, MSc†,
- Stuart J. Pocock, PhD† and
- Keith A.A. Fox, MB, ChB‡
- ∗Trent Cardiac Centre, Nottingham University Hospitals, Nottingham, United Kingdom
- †London School of Hygiene and Tropical Medicine, London, United Kingdom
- ‡University of Edinburgh, Edinburgh, United Kingdom
- ↵∗Reprint requests and correspondence:
Dr. Robert A. Henderson, Trent Cardiac Centre, Nottingham University Hospitals, City Hospital Campus, Hucknall Road, Nottingham NG5 1PB, United Kingdom.
Background The RITA-3 (Third Randomised Intervention Treatment of Angina) trial compared outcomes of a routine early invasive strategy (coronary arteriography and myocardial revascularization, as clinically indicated) to those of a selective invasive strategy (coronary arteriography for recurrent ischemia only) in patients with non–ST-segment elevation acute coronary syndrome (NSTEACS). At a median of 5 years’ follow-up, the routine invasive strategy was associated with a 24% reduction in the odds of all-cause mortality.
Objectives This study reports 10-year follow-up outcomes of the randomized cohort to determine the impact of a routine invasive strategy on longer-term mortality.
Methods We randomized 1,810 patients with NSTEACS to receive routine invasive or selective invasive strategies. All randomized patients had annual follow-up visits up to 5 years, and mortality was documented thereafter using data from the Office of National Statistics.
Results Over 10 years, there were no differences in mortality between the 2 groups (all-cause deaths in 225 [25.1%] vs. 232 patients [25.4%]: p = 0.94; and cardiovascular deaths in 135 [15.1%] vs. 147 patients [16.1%]: p = 0.65 in the routine invasive and selective invasive groups, respectively). Multivariate analysis identified several independent predictors of 10-year mortality: age, previous myocardial infarction, heart failure, smoking status, diabetes, heart rate, and ST-segment depression. A modified post-discharge Global Registry of Acute Coronary Events (GRACE) score was used to calculate an individual risk score for each patient and to form low-risk, medium-risk, and high-risk groups. Risk of death within 10 years varied markedly from 14.4 % in the low-risk group to 56.2% in the high-risk group. This mortality trend did not depend on the assigned treatment strategy.
Conclusions The advantage of reduced mortality of routine early invasive strategy seen at 5 years was attenuated during later follow-up, with no evidence of a difference in outcome at 10 years. Further trials of contemporary intervention strategies in patients with NSTEACS are warranted. (Third Randomised Intervention Treatment of Angina trial [RITA-3]; ISRCTN07752711)
Over recent years, 2 alternative management strategies have evolved for the management of patients with non–ST-segment elevation acute coronary syndrome (NSTEACS). The routine invasive strategy involves early coronary arteriography within 72 h of an index episode of myocardial ischemia, and myocardial revascularization as clinically indicated. The routine invasive strategy defines the coronary anatomy in all patients and facilitates selection of the most appropriate treatment strategy. In contrast, the selective invasive strategy involves continuation of medical therapy, with coronary arteriography and myocardial revascularization reserved for patients with persistent or recurrent myocardial ischemia. The selective invasive strategy aims at identifying patients most likely to benefit from an invasive procedure, thereby potentially minimizing the risks and costs of routine invasive management.
Several randomized trials have compared routine invasive and selective invasive strategies in patients with NSTEACS (1–9). Pooled data from these trials suggest that a routine early invasive strategy reduces the risk of myocardial infarction, severe angina, and rehospitalization (10–12). Moreover, in an individual patient data meta-analysis of 3 trials reporting long-term follow-up (FRISC-II [FRagmin and Fast Revascularisation during InStability in Coronary artery disease]; ICTUS [Invasive versus Conservative Treatment in Unstable coronary Syndromes], and RITA-3 [Third Randomised Intervention Treatment of Angina]), a routine invasive strategy was associated with a lower risk of death or myocardial infarction over 5 years, particularly in patients at increased baseline ischemic risk (13). These results have been incorporated into national and international clinical guidelines (14–16).
The RITA-3 trial assigned 1,810 patients with NSTEACS to receive either routine invasive treatment or selective invasive treatment strategies, and the short- and medium-term results of the trial have been reported previously (1,17,18). During the first year of follow-up, the routine invasive strategy approximately halved the risk of refractory angina, relative to that of the selective invasive strategy (1). At 5 years, the routine invasive strategy was associated with an estimated 26% reduction in the odds of cardiovascular death and myocardial infarction and a 24% reduction in the odds of all-cause death (17). Health economic analysis suggested that a routine invasive strategy might be cost effective, particularly in patients at intermediate or high baseline risk (19). In this report we extended follow-up of RITA-3 patients to 10 years and assessed the impact of baseline risk on long-term mortality.
The methods of the RITA-3 trial have been reported previously (1). Briefly, from November 1997 to October 2001, we enrolled 1,810 patients with NSTEACS in an open randomized trial of routine invasive versus selective invasive treatment strategies. Patients were randomized within 48 h of an index episode of myocardial ischemia if they had suspected cardiac chest pain at rest and documented evidence of coronary artery disease with at least 1 of the following: evidence of ischemia on an electrocardiogram; pathological Q waves, suggesting previous myocardial infarction; or coronary artery disease on a previous arteriogram. In all cases, the participating cardiologist had to be uncertain about the optimal management strategy, and continued medical therapy had to be an acceptable treatment option.
Patients were excluded if coronary arteriography was planned within 72 h of the index episode of myocardial ischemia or if the ischemia was thought to be due to an arrhythmia, anemia, or noncoronary disease. RITA-3 recruited patients before the introduction of serum troponin as a routine biomarker of myocardial necrosis, and patients were excluded if serum creatine kinase concentration was elevated to twice the upper limit of normal before randomization.
The trial protocol recommended that all patients be treated during the index hospital admission with aspirin and enoxaparin, 1 mg/kg subcutaneously twice daily for 2 to 8 days. Adenosine diphosphate (ADP) receptor antagonists and glycoprotein IIb/IIIa receptor inhibitors were prescribed as clinically appropriate. Secondary prevention treatment with aspirin, beta-blockers, angiotensin-converting enzyme inhibitors, and statins was encouraged in all patients and was monitored during the first 5 years of the trial by a member of the RITA-3 Executive Committee.
Patients assigned to receive routine invasive strategy underwent coronary arteriography as soon as possible and ideally within 72 h of randomization. The requirement for percutaneous coronary intervention (PCI) or coronary artery bypass graft (CABG) surgery was guided by the arteriographic findings, with no protocol restriction on the use of intracoronary stents, other interventional devices, or pharmacological treatments. All other aspects of patient management were initiated at the discretion of the supervising clinician.
Patients assigned to receive selective invasive strategy were managed with antianginal medication, with the objective of controlling angina symptoms. Coronary arteriography was indicated only for failure of the selective invasive strategy, defined by recurrence of ischemic pain at rest or on minimum exertion, with transient or persistent electrocardiographic evidence of ischemia despite full antianginal medication (usually including beta-blockers, nitrates, and antithrombotic treatment in clinically appropriate doses). After discharge from hospital, coronary arteriography could be performed for exertional angina despite appropriate antianginal medication or for evidence of ischemia on functional testing.
Patients had yearly follow-up examinations at the hospital to 5 years, to document symptoms, cardiac events, and vital status. Vital status at 5 years was known for 1,802 of 1,810 patients (99.6%), and we have previously reported mortality and rates of myocardial infarction and revascularization procedures at that time point (17). The trial was not designed or funded to collect information on nonfatal outcomes beyond 5 years, but all surviving patients were prospectively registered with the Office of National Statistics (ONS) in England and with the General Register Office (GRO) in Scotland to ascertain deaths from national mortality data, with 10-year all-cause mortality as the pre-specified secondary outcome. A change in ONS policy prevented collection of mortality information beyond March 2011. Deaths before 5 years were evaluated by an event validation committee and classified as cardiovascular or noncardiovascular. Deaths between 5 and 10 years were classified as cardiovascular or noncardiovascular by an investigator blinded to treatment assignment, on the basis of the cause of death recorded on the death certificate.
Multicenter national and local ethics committee approvals were obtained. All patients provided written informed consent to participate in the trial before randomization.
All analyses were done by intention to treat. Kaplan-Meier estimates of 10-year all-cause and cardiovascular mortality were calculated together with risk differences and confidence intervals (CIs), and treatment effect was assessed with a log-rank test. Treatment by time interactions were used to assess whether any impact of treatment differed over follow-up time.
To identify baseline characteristics associated with all-cause mortality, multivariate logistic regression was used; Cox regression was not appropriate because of clear nonproportionality between the treatment groups. In order to allow for the 220 patients (12%) with between 9.4 and 10 years’ follow-up due to the changes in ONS policy, an adjustment was made by including a covariate of log of total time in the study for each patient, although this had little impact on the estimates of effect of the model. Treatment, age, and sex were included in the model regardless of their statistical significance. Other candidate baseline variables were added in a stepwise manner, using a p value of <0.01 as the criterion for inclusion. A Wald test was used to assess overall effect for a 3-category variable (e.g., smoking). Time-updated analyses incorporated myocardial infarction and revascularization procedures occurring up to 5 years after randomization into the multivariate model.
In addition, an analysis was undertaken to assess the impact of the assigned treatment strategy according to a patient’s underlying risk, using a risk score developed independently from RITA-3. We used the post-discharge Global Registry of Acute Coronary Events (GRACE) model, an external, validated model that was developed to predict 6-month mortality in patients with acute coronary syndromes who survived to hospital discharge (20). This model was selected because patients enrolled in RITA-3 were randomized in hospital up to 48 h following the index episode of ischemia and after the very early high-risk period for further cardiovascular events. A risk score was calculated for each patient by using a modified version of the post-discharge GRACE model nomogram. RITA-3 was designed before the impact of renal function on long-term outcome of patients with acute coronary syndromes was fully recognized, and baseline renal function was not recorded routinely in the trial. Renal function, therefore, was excluded from the risk model, and this approach was validated previously in a large cohort with acute coronary syndrome cases in the United Kingdom (21). In addition, we excluded in-hospital PCI from the risk model, as the procedures occurred after randomization, were strongly influenced by assigned treatment strategy, and could potentially confound comparison of risk scores between treatment strategy groups. Individual patient risk scores were then used to stratify patients into low-, medium-, and high-risk groups such that each group contained approximately one-third of the deaths. Kaplan-Meier estimates of 10-year all-cause mortality were compared between the strategies in each risk group. Interactions between risk groups and treatment were used to formally assess whether the impact of treatment differed in the 3 groups on the relative or absolute scale. A logistic regression analysis of individual patient risk scores was used to calculate predicted 10-year mortality, which was then plotted against risk score.
Baseline characteristics were well matched between the randomized groups; the mean age at randomization was 62 years old, and 38% of patients were female. Electrocardiographic evidence of myocardial ischemia was present at baseline in 92%, and 41% had ST-segment deviation of at least 0.1 mV. An elevated level of a serum cardiac biomarker (creatine kinase or troponin) was detected in 25%, a history of previous myocardial infarction was recorded in 28%, and 13% had diabetes at baseline (1,17).
At 1 year, 86%, 61%, 29%, and 74% (and at 5 years, 80%, 58%, 37%, and 82%) of the patients were taking aspirin, beta-blocker, angiotensin-converting enzyme inhibitor, and statin therapy, respectively. There were no differences between the 2 treatment strategy groups in the use of any of these medications. Thienopyridines were used routinely in both groups after implantation of a coronary artery stent.
There was good separation of treatment strategies between the randomized groups, and during the index hospital admission, coronary arteriography was performed in 857 patients (96%) assigned to routine invasive strategy and 142 patients (16%) assigned to selective invasive strategy. In the routine invasive strategy group, the assigned coronary arteriogram resulted in revascularization in 397 patients (55%) (consisting of PCI in 35% [stents inserted in 88% of procedures] and CABG in 21%). In the selective invasive strategy group, myocardial revascularization during the index hospital admission was performed in 94 patients (10%) (consisting of PCI in 7% and CABG in 4%). During subsequent follow-up, the rate of revascularization procedures was slightly lower in the routine invasive group than in the selective invasive group, and at 5 years, at least 1 revascularization procedure had been carried out in 546 (61%) and 347 (38%) of the routine invasive and selective invasive strategy groups, respectively (17).
All-cause and cardiovascular mortality
In total, 457 deaths (25% of the randomized population) occurred by 10 years (501 occurred over a median of 10.6 years), including 242 deaths (53%) before 5 years and 215 deaths (47%) between 5 and 10 years. Deaths were classified as cardiovascular in 282 cases by 10 years (305 in total), accounting for 62% of all deaths (Table 1).
Overall, there were no differences in all-cause mortality between the treatment groups (225 deaths [25.1%] in the routine invasive group vs. 232 deaths [25.4%] in the selective invasive group, for a risk difference of −0.2%; 95% CI: −4.2% to 3.8%; p = 0.94). Cardiovascular mortality was also similar between the 2 groups (135 deaths [15.1%] vs. 147 deaths [16.1%], respectively; p = 0.65). The mortality curves diverged over the first 5 years but then progressively converged over the following 5 to 10 years (Figures 1 and 2⇓). Rates of all-cause mortality at 2-year intervals in the routine invasive and selective invasive arms are shown in Online Table 1. The hazard ratio (HR) of all-cause mortality in the routine invasive group versus the selective invasive group was lower during years 0 to 5 (HR: 0.76; 95% CI: 0.59 to 0.98) but higher during years 5 to 10 (HR: 1.28; 95% CI: 0.98 to 1.68; p = 0.006 for treatment-time interaction). Similar results were found for cardiovascular mortality (HR: 0.70; 95% CI: 0.51 to 0.97; and HR: 1.29; 95% CI: 0.70 to 1.83, for years 0 to 5 and 5 to 10, respectively; treatment-time interaction p = 0.013).
Baseline characteristics independently associated with all-cause 10-year mortality are shown in Table 2, and a predictive model on the basis of these variables showed good discrimination (c-statistic: 0.809). Variables associated with all-cause mortality at 10 years were also associated with all-cause mortality at 5 years (Online Table 2). Time-updated multivariate analyses suggest that the HR of all-cause mortality at 10 years was increased by new myocardial infarction within 5 years of randomization but was not associated with revascularization procedures (Online Tables 3 and 4).
A risk score was calculated with the modified post-discharge GRACE score for all 1,810 patients. Patients were stratified into risk groups, with approximately equal numbers of deaths in each group, resulting in low-risk, intermediate-risk, and high-risk categories that included 63.0%, 22.8%, and 14.2% of the patients, respectively. The 10-year mortality observed across these risk groups ranged widely from 13.4% in the low-risk group to 58.0% in the high-risk group (Table 3). The median risk score was 82 (interquartile range [IQR]: 64 to 97), and the distribution of scores compared with predicted cumulative 10-year mortality for all patients is shown in Figure 3. There was close agreement between predicted and observed 10-year mortality rates in the 3 risk groups (Figure 3).
Cumulative 10-year mortality by assigned treatment strategy for the 3 risk groups is shown in the Central Illustration. There were no differences in all-cause mortality between the routine invasive and selective invasive strategies in the low-risk and intermediate-risk categories (Table 3). In the high-risk subgroup, patients undergoing routine invasive strategy appeared to be at lower risk of mortality than patients in the selective invasive group (risk difference at 5 years was 11.4% [95% CI: −22.8% to −0.1%]) but this difference was attenuated later in follow-up because of a low mortality rate in the selective invasive group. At 10 years, the absolute risk difference in the high-risk group was −0.3% (95% CI: −12.4% to 11.7%), and overall, there was no evidence for an interaction between risk score and treatment effect on all-cause mortality at 10 years on the relative (p = 0.79) or absolute scale (p = 0.88).
RITA-3 compared outcomes of routine invasive to those of selective invasive treatment strategies in patients with NSTEACS, and in a previous report, the routine invasive strategy was associated with lower all-cause and cardiovascular mortality at 5 years’ follow-up (17). We extended follow-up of RITA-3 patients to 10 years and report a progressive diminution of any mortality differences beyond 5 years, suggesting that the differences in revascularization rates between the routine invasive and selective invasive strategies during the index admission (55% versus 10%, respectively) and sustained over 5 years (61% vs. 34%, respectively) are not associated with longer-term survival benefit. It is therefore unclear whether the mortality difference at 5 years in RITA-3 is due to a direct treatment effect that attenuates over time, perhaps due to treatment crossovers, or due to the play of chance.
The relatively low rates of revascularization in the invasive strategy group in RITA-3 reflect practice in the United Kingdom at the time of enrollment, but there was clear separation in revascularization rates between the 2 treatment strategies. Moreover, in other large trials reporting higher revascularization rates, a routine invasive strategy did not reduce all-cause or cardiovascular mortality at 5 years (13,22,23). In individual patient data meta-analyses of FRISC-II, ICTUS, and RITA-3, there were no differences in 5-year all-cause mortality, and a difference in cardiovascular mortality of borderline statistical significance was driven largely by RITA-3 (13). Hence, definitive evidence for the impact of a routine invasive strategy on longer-term mortality is lacking, but our report suggests that, for most patients with NSTEACS, a major survival benefit over 10 years is unlikely.
The multivariate analysis of RITA-3 patients reported in this paper identified several factors that were independently associated with 10-year mortality. For example, a 10-year increment in age was associated with a 4.4-fold increase in the odds of death, whereas markers of impaired cardiac function (e.g., history of previous myocardial infarction or heart failure), and the severity of underlying coronary artery disease (e.g., ST-segment depression) also had substantial effects on all-cause mortality.
The multivariate model developed from RITA-3 data showed good discrimination, but we preferred a modified post-discharge GRACE score to stratify individual patients by baseline risk because it was developed independently of RITA-3 to predict 6-month mortality in patients with acute coronary syndrome who survived to hospital discharge and was validated in an unselected cohort of patients from the large international GRACE registry (20). The post-discharge GRACE score has not been validated in external datasets with longer-term follow-up, but several of the variables included in the score were also independently associated with 10-year mortality in RITA-3 (e.g., age, history of heart failure, ST-segment depression). Moreover the modified post-discharge GRACE score was shown to reliably discriminate between those at higher or lower risk of 10-year mortality, suggesting that the GRACE score could be used to predict mortality in the longer term.
We used the modified post-discharge GRACE score to stratify patients into low-risk, medium-risk, and high-risk subgroups (Central Illustration). There was no evidence of benefit from routine invasive management in the low- and medium-risk groups, which comprised more than 80% of all patients. We observed a trend toward benefit in the high-risk group, with a median difference in survival between the routine invasive and selective invasive strategies over 10 years of 1.09 years. If real, the magnitude of this difference in survival is potentially clinically important, but these data should be interpreted cautiously as this post-hoc subgroup analysis is based on relatively small numbers of deaths and lacks statistical power, and there was no evidence at 10 years for an interaction between risk score and treatment. Further investigation of the effect of underlying risk on the impact of a routine invasive strategy on outcome is warranted, and an updated individual patient data meta-analysis of the FRISC-II, ICTUS, and RITA-3 trials is planned.
RITA-3 recruited patients considered suitable for either routine invasive or selective invasive strategies, and patients in whom early coronary arteriography was considered inappropriate (or mandatory) were excluded. The results of RITA-3 therefore cannot be generalized to the wider unselected population of patients with NSTEACS. In RITA-3, the observed 6-month mortality in the low-, intermediate-, and high-risk subgroups (defined by the modified version of the post-discharge GRACE score) was 1.8%, 5.6%, and 8.0%, respectively. These mortality rates suggest that RITA-3 patients were at low to intermediate levels of risk compared with the spectrum of risk seen in unselected patients with NSTEACS in the U.K. Myocardial Ischaemia National Audit Project registry, more than one-half of whom had a 6-month mortality risk exceeding 9% (24). In both the FRISC-II and ICTUS trials, mortality rates in the routine and selective invasive strategy groups at 6 months were <3% (4,25), and the results of these trials are therefore also not directly relevant to those patients with NSTEACS at highest risk. These data suggest that the randomized trials of routine invasive versus selective invasive strategies in patients with NSTEACS have consistently excluded patients at highest risk and that the optimal treatment strategy for these patients has not been defined.
The long-term results of RITA-3 have important implications for clinical practice. For most patients with NSTEACS who are eligible for routine invasive or selective invasive treatment strategies, neither strategy confers a prognostic advantage over 10 years. This lack of prognostic benefit must be balanced against the beneficial impact of a routine invasive strategy on other outcomes, (10–12,17,18,22,23), but, for many patients at lower levels of baseline risk, a conservative treatment strategy remains a reasonable treatment option.
We relied on national mortality data to determine vital status during follow-up from 5 to 10 years and classified deaths after 5 years as cardiovascular and noncardiovascular on the basis of the certified cause of death. In the United Kingdom, there is a legal requirement that all deaths be registered with ONS or GRO, but we cannot exclude the possibility that some deaths were not recorded (for example, among patients who emigrated) and that mortality might therefore be underestimated. Nevertheless, Kaplan-Meier plots show relatively constant mortality over 10 years, suggesting that substantial underestimation or biased reporting of mortality is unlikely.
The lack of benefit of reduced mortality from either treatment strategy at 10 years must be balanced against earlier reductions in the risk of recurrent ischemia and myocardial infarction with the routine invasive strategy (13,17). We did not collect information about these endpoints or about revascularization procedures beyond 5 years, and this limits scope for exploratory analyses to examine why any mortality advantage of a routine invasive strategy at 5 years attenuates during longer-term follow-up. Nevertheless, in time-updated multivariate analysis, myocardial infarction during the first 5 years of follow-up was associated with 10-year all-cause mortality, but revascularization procedures had no substantial impact on outcome.
RITA-3 recruited patients from 1997 to 2001 and prior to the availability of drug-eluting stents and routine use of ADP receptor antagonists in patients with NSTEACS (15). Evidence from mixed-treatment comparison meta-analysis suggests that second-generation everolimus-eluting stents may reduce mortality relative to that of bare-metal stents (26), but this benefit has not been demonstrated specifically in patients with NSTEACS managed by a routine invasive strategy.
In the RITA-3 trial of routine invasive versus selective invasive strategies in patients with NSTEACS, the advantage of reduced mortality of a routine early invasive strategy seen at 5 years attenuated during later follow-up. Risk of death within 10 years was influenced by several baseline variables and varied markedly from 14.4% in the low-risk group to 56.2% in the high-risk group, but this mortality trend did not depend on the assigned treatment strategy. Further trials of contemporary intervention strategies in patients with NSTEACS are warranted.
COMPETENCY IN MEDICAL KNOWLEDGE: Patients with NSTEACS who are managed with routine invasive strategies have total and cardiovascular mortality rates comparable to those of patients managed with selective invasive strategies over 10 years.
TRANSLATIONAL OUTLOOK: Future trials should prospectively seek to identify and validate specific clinical risk factors that predict benefit from an initially invasive treatment strategy for patients presenting with NSTEACS.
The authors thank all members of the RITA-3 trial committees, the investigators and coordinators of the RITA-3 trial, the medical and nursing staff in the participating centers responsible for recruitment and follow-up of patients, and most of all, the patients who participated in the trial.
RITA-3 was funded by a competitive grant from the British Heart Foundation, and the British Heart Foundation received a donation from Aventis Pharma. Additional governmental support was obtained to reimburse interventional centers for part of the costs of PCI procedures on trial patients. Mr. Clayton has received grant support from The Medicines Company. All other authors have reported that they have no relationships relevant to the contents of this paper to disclose.
- Abbreviations and Acronyms
- coronary artery bypass graft
- confidence interval
- Global Registry of Acute Coronary Events
- General Register Office
- hazard ratio
- non–ST-segment elevation acute coronary syndrome
- Office of National Statistics
- percutaneous coronary intervention
- Received April 7, 2015.
- Accepted May 19, 2015.
- American College of Cardiology Foundation
- Fox K.A.A.,
- Poole-Wilson P.A.,
- Henderson R.A.,
- et al.,
- for the Randomized Intervention Trial of unstable Angina (RITA) investigators
- Cannon C.P.,
- Weintraub W.S.,
- Demopoulos L.A.,
- et al.,
- for the TACTICS–Thrombolysis in Myocardial Infarction 18 Investigators
- Spacek R.,
- Widimsky P.,
- Straka Z.,
- et al.
- McCullough P.A.,
- O'Neill W.W.,
- Graham M.,
- et al.
- Michalis L.K.,
- Stroumbis C.S.,
- Pappas K.,
- et al.
- Boden W.E.,
- O'Rourke R.A.,
- Crawford M.H.,
- et al.,
- for the Veterans Affairs Non-Q-Wave Infarction Strategies in Hospital (VANQWISH) Trial Investigators
- TIMI-IIIB Investigators
- Bavry A.A.,
- Kumbhani D.J.,
- Rassi A.N.,
- et al.
- Fox K.A.A.,
- Clayton T.C.,
- Damman P.,
- et al.,
- for the FIR Collaboration
- Jneid H.,
- Anderson J.L.,
- Wright R.S.,
- et al.
- Hamm C.W.,
- Bassand J.P.,
- Agewall S.,
- et al.
- National Institute for Health and Clinical Excellence. Unstable angina and NSTEMI: the early management of unstable angina and non-ST-segment-elevation myocardial infarction. 2013. Available at: http://www.nice.org.uk/guidance/cg94/resources/guidance-unstable-angina-and-nstemi-pdf. Accessed May 28, 2015.
- Kim J.,
- Henderson R.A.,
- Pocock S.J.,
- et al.
- Henriksson M.,
- Epstein D.M.,
- Palmer S.J.,
- et al.
- Simms A.D.,
- Reynolds S.,
- Pieper K.,
- et al.
- Lagerqvist B.,
- Husted S.,
- Kontny F.,
- et al.,
- for Fast Revascularisation during InStability in Coronary artery disease (FRISC-II) Investigators
- Damman P.,
- Hirsch A.,
- Windhausen F.,
- et al.,
- for the ICTUS Investigators
- Gray H.H.,
- Henderson R.A.,
- de Belder M.A.,
- et al.
- Bangalore S.,
- Toklu B.,
- Amoroso N.,
- et al.