3.3. Initial Invasive versus Initial Conservative Strategies: Recommendations
(See Table 4, and Appendixes 5 and Table a6 for supplemental information.)
3.3.3.1. Timing of Invasive Therapy Among initially stabilized patients with UA/NSTEMI for whom an early invasive strategy of coronary angiography is chosen, optimal timing of angiography has not been well defined. Early or immediate catheterization with revascularization of unstable coronary lesions may prevent ischemic events that would otherwise occur during medical therapy. Conversely, pretreatment with intensive antithrombotic therapy may diminish thrombus burden and "passivate" unstable plaques, improving the safety of percutaneous revascularization and reducing the risk of periprocedural ischemic complications. Three trials have compared different strategies of "early" versus "delayed" intervention in patients with UA/NSTEMI and form the basis of the updated recommendations in this guideline.
The ISAR-COOL (Intracoronary Stenting with Antithrombotic Regimen Cooling-Off) trial carried out at 2 hospitals between 2000 and 2002 randomized 410 patients with unstable chest pain and either electrocardiographic ST-segment depression or elevated troponin levels to undergo coronary angiography within 6 hours of presentation (median 2.4 hours) or after 3 to 5 days (median 86 hours) of antithrombotic pretreatment. Patients with "large MI," defined by ST-segment elevation or creatine kinase–myocardial band isoenzyme activity >3 times normal, were excluded. Underlying medical therapy in both treatment arms included aspirin, clopidogrel, UFH, and tirofiban. By 30 days' follow-up, the primary endpoint of death or large MI (defined by new electrocardiographic Q waves, left bundle-branch block, or creatine kinase–myocardial band elevation >5 times normal) occurred in 11.6% of patients randomized to delayed catheterization versus 5.9% of those in the early angiography group (p=0.04). Differences between treatment groups were observed exclusively in the period before catheterization, with identical event rates in the 2 arms after angiography. Although providing evidence that a strategy of "cooling-off" for 3 to 5 days before angiography does not improve outcome in this setting, the findings of this trial were limited because of the small sample size and the prolonged delay before angiography in the medical pretreatment arm.
Information more relevant to contemporary practice patterns was provided in the 2009 publication of the large-scale multicenter TIMACS (Timing of Intervention in Acute Coronary Syndromes) trial, which compared early versus delayed angiography and intervention in patients with non–ST-segment elevation ACS. Patients were included if they presented within 24 hours of onset of unstable ischemic symptoms with advanced age (≥60 years), elevated cardiac biomarkers, or ischemic electrocardiographic changes, and were randomized to undergo angiography as rapidly as possible and within 24 hours of randomization (median 14 hours) versus after a minimum delay of 36 hours (median 50 hours). Anticoagulation included aspirin, clopidogrel in >80% of patients, heparin or fondaparinux, and GP IIb/IIIa inhibitors in 23% of patients. Although the trial was initially powered for enrollment of 4,000 patients to detect a 25% reduction in the primary endpoint of death, new MI, or stroke at 6 months, the steering committee chose to terminate enrollment at 3,031 patients because of recruitment challenges. Among the overall trial population, there was only a nonsignificant trend toward a reduced incidence of the primary clinical endpoint, from 11.3% in the delayed intervention group to 9.6% in the early intervention arm (HR for early intervention: 0.85; 95% CI: 0.68 to 1.06; p=0.15). However, a prospectively defined secondary endpoint of death, MI, or refractory ischemia was significantly reduced by early intervention from 12.9% to 9.5% (HR: 0.72; 95% CI: 0.58 to 0.89; p=0.003), mainly because of a difference in the incidence of refractory ischemia (3.3% versus 1.0% in the delayed versus early intervention arms, respectively; p<0.001). The occurrence of refractory ischemia was associated with a >4-fold increase in risk of subsequent MI. Moreover, significant heterogeneity was observed in the primary endpoint when stratified according to a prespecified estimation of baseline risk according to the GRACE (Global Registry of Acute Coronary Events) score. Patients in the highest tertile of the GRACE risk score (>140) experienced a sizeable and significant reduction in the incidence of the primary ischemic endpoint, from 21.0% to 13.9% (HR: 0.65; 95% CI: 0.48 to 0.89; p=0.006), whereas no difference in outcome (6.7% versus 7.6% in the delayed and early groups, respectively; HR: 1.12; 95% CI: 0.81 to 1.56; p=0.48) was observed among patients in the lower 2 risk tertiles (GRACE score ≤140).
Results of the TIMACS trial suggested superior outcome among patients managed by early rather than delayed intervention in the setting of UA/NSTEMI, although the reduction in the primary endpoint did not reach statistical significance for the overall trial population. Nevertheless, refractory ischemia was reduced by an early approach, as were the risks of death, MI, and stroke among patients at the highest tertile of ischemic risk as defined by the GRACE risk score.
To assess whether a more aggressive strategy of very early intervention, analogous to the standard of primary PCI for STEMI, would lead to improved outcomes in patients with non–ST-elevation ACS, the ABOARD (Angioplasty to Blunt the Rise of Troponin in Acute Coronary Syndromes) study investigators compared angiography and intervention performed immediately on presentation with intervention carried out on the next working day. A total of 352 patients with unstable ischemic symptoms, ECG changes, or troponin elevation were randomized at 13 hospitals to immediate (at a median 70 minutes after enrollment) versus delayed (at a median 21 hours) angiography and revascularization. Background antithrombotic therapy consisted of aspirin, clopidogrel with a loading dose of ≥300 mg, abciximab during PCI, and the anticoagulant of the investigator's choice. The primary trial endpoint was peak troponin I value during the hospitalization period. Immediate intervention conferred no advantage with regard to the primary endpoint (median troponin I value 2.1 versus 1.7 ng/mL in the immediate and delayed intervention groups, respectively), nor was there even a trend toward improved outcome in the prespecified clinical secondary endpoint of death, MI, or urgent revascularization by 1 month (13.7% versus 10.2% in the immediate and delayed intervention groups, respectively; p=0.31).
These 3 trials, taken together with earlier studies, do provide support for a strategy of early angiography and intervention to reduce ischemic complications in patients who have been selected for an initial invasive strategy, particularly among those at high risk (defined by a GRACE score >140), whereas a more delayed approach is reasonable in low- to intermediate-risk patients. The "early" time period in this context is considered to be within the first 24 hours after hospital presentation, although there is no evidence that incremental benefit is derived by angiography and intervention performed within the first few hours of hospital admission. The advantage of early intervention was achieved in the context of intensive background antithrombotic therapy Table 4.