Society of Cardiovascular Anesthesiologists

Routine vs. selective invasive strategies in patients with acute coronary syndromes.

Mehta SR, Cannon CP, Fox KAA, et al. JAMA 2005; 293:2908-17.

Reviewer: K.W. Tim Park, MD
Santa Clara Valley Medical Center
San Jose, CA

Background: The role and timing of percutaneous coronary intervention (PCI) in patients with acute coronary syndrome is still a matter of controversy and investigation (Circulation 2002; 106:1893-1900). A decision clinicians face commonly is whether patients with unstable angina (UA) or non-ST-elevation myocardial infarction (NSTEMI) should be routinely referred to invasive procedures or treated aggressively with pharmacological agents followed by selective referral of those with refractory or inducible angina to invasive intervention. This study compares by a meta-analysis the benefits and risks of routine vs. selective invasive strategies in such patients.

Methods: The MEDLINE and Cochrane databases from 1970 to June 2004 and abstracts from major cardiology meetings were screened to identify randomized controlled trials of routine vs. selective invasive strategies in patients with UA or NSTEMI. A routine strategy was defined as the referral of all patients with UA or NSTEMI for coronary angiography and revascularization in patients with suitable coronary anatomy. A selective strategy was defined as initial treatment of patients with pharmacological agents followed by cardiac catheterization and revascularization only in those patients with recurrent symptoms or objective evidence of inducible ischemia on noninvasive testing. Outcome variables compared were in-hospital and long-term mortality, nonfatal MI, the composite of death or nonfatal MI, Canadian Cardiovascular Society (CCS) class III or IV angina, and rehospitalization.

Results: Seven trials involving 9212 patients (4608 routine and 4604 selective) were included in the analysis. The mean age of the patients was 62.4, with 69.2% of them men. The incidence of diabetes was 18.9% and history of MI 32.5%. 59% of the patients presented with NSTEMI and 41% with UA. The mean follow-up was 17.3 months. During the initial hospitalization, 58.4% of the routine group and 24.0% of the selective group underwent revascularization. By the end of the follow-up, 63.6% of the routine group and 41.5% of the selective group underwent revascularization.

During the initial hospitalization, mortality, nonfatal MI, and the composite of death or nonfatal MI all tended to be higher in the routine invasive group (1.8% vs. 1.1%, OR 1.60, P = 0.007; 3.7% vs. 3.0%, OR 1.24, P = 0.07; 5.2% vs. 3.8%, OR 1.36, P = 0.002). On the other hand, following discharge, the incidence of adverse cardiac events was lower in the routine group. (mortality: 3.8% vs. 4.9 %, OR 0.76, P = 0.01; nonfatal MI: 3.8% vs. 6.6%, OR = 0.56, P < 0.001; death or nonfatal MI: 7.4% vs. 11.0 %, P < 0.001). For the whole study period from randomization to the end of the follow-up, mortality was not significantly different between the two groups (5.5% vs. 6.0%, OR 0.92, P = 0.33). However, nonfatal MI was reduced in the routine group (7.3% vs. 9.4%, OR 0.75, P < 0.001) as well as the composite of death or nonfatal MI. In addition, with the routine invasive strategy, there was a significant reduction in rehospitalization (32.5% vs. 41.3%, OR 0.66, P < 0.001) and the proportion of patients with CCS class III or IV angina (11.2% vs. 14.0%, OR 0.77, P < 0.001).

Heterogeneity among different trials was noted in the rate of in-hospital MI or the composite of in-hospital death or MI. The source of this heterogeneity was mostly in the TACTICS-TIMI 18 (Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Therapy - Thrombolysis in Myocardial Infarction 18) trial (N Engl J Med 2001; 344:1879-87). This trial was notable for the routine use of a glycoprotein IIb/IIIa antagonist tirofiban "upstream" of the coronary intervention (which may have reduced procedure-related complications) and significantly different cardiac marker threshold for an MI outcome after PCI (> 3 times the upper limit of normal) vs. for a non-procedure-related MI (> the upper limit of normal), (which may have contributed to an underdiagnosis of procedure-related MI.)

In stratified analyses, trials published after 1999 tended to favor a routine invasive strategy, whereas those prior to 1999 tended to be neutral. In the three most recent trials, the benefits of a routine invasive strategy in reducing death or nonfatal MI were limited to those patients who were troponin-positive, but not in those who were troponin-negative.

Comments: This meta-analysis demonstrates that in patients with UA or NSTEMI, a routine invasive strategy may be superior to a selective strategy in reducing long-term cardiac adverse events, but the benefits are not evident until after discharge from the hospital. The routine strategy may actually be associated with an increased early hazard. The study further suggests that better stratification of the patients will be needed to identify those who will truly benefit from a routine invasive strategy. Patients with positive cardiac markers may benefit from a routine invasive strategy, but not those with negative markers. By extension, other high-risk groups such as those with cardiogenic shock or significant heart failure may also benefit from a routine invasive strategy, although this needs to be demonstrated. Unfortunately, the current cardiology practice pattern is the opposite in that high-risk patients are treated more conservatively, while lower-risk patients are more likely to undergo an invasive strategy (JAMA 2004; 292:2096-2104).

Since the benefits of a routine invasive strategy mainly emerge over the long-term, the timing of intervention in the routine invasive strategy may need to be optimized further. Two randomized trials have been performed to address the potential benefits of early vs delayed intervention, with contrasting results. In the larger of these two trials involving 1,201 patients, early intervention led to an increase in MI. Of note, in this Dutch trial, a large proportion of patients were on clopidogrel and high-dose statins in both groups and the definition of MI was the same for spontaneous and procedure-related MIs. As mentioned above, "upstream" use of a glycoprotein IIb/IIIa antagonist for a period prior to PCI may reduce procedure-related complications.

Lastly, the role and importance of adjunctive pharmacological therapies with and without PCI should be further defined. Clopidogrel may confer benefits not only in patients treated prior to PCI (at least 6 hours), but also in those treated medically (N Engl J Med 2001; 345:494-502). Because, in the trials included in this meta-analysis, clopidogrel was used mainly in patients after coronary stents for 30 days, the imbalance in its use may have skewed the results of the trials. In addition to clopidogrel, statins, beta-adrenergic blockers, and angiotensin-converting enzyme inhibitors would be expected to improve outcome in patients with UA or NSTEMI, regardless of management approach.

Table of Contents:

Literature Reviews