Although multivessel coronary artery disease (CAD) is commonly encountered in patients with acute coronary syndrome (ACS), no randomized controlled trials have evaluated the effi cacy of singlevessel percutaneous coronary intervention (PCI) versus multivessel PCI.
Many patients (>40%) with acute coronary syndrome (ACS) and no previous history of coronary artery bypass graft (CABG) surgery have angiographically significant multivessel stenosis.1 Although multivessel coronary artery disease (CAD) is prevalent in these patients, no definitive treatment guidelines exist, and deciding whether to intervene only on the culprit lesion or on multiple epicardial vessels is left up to the treating clinician.
Most of the data regarding percutaneous coronary intervention (PCI) for multivessel CAD come from single-center studies that are not randomized or controlled. These studies have indicated that when single-vessel PCI is performed solely on the culprit lesion, procedural times are usually decreased. The risk of non-culprit-vessel closure, which is a rare procedural complication that can lead to other adverse events, such as additional loss of viable myocardium, is also eliminated. On the other hand, multivessel PCI may offer several advantages, including complete or near-complete revascularization, reduced need for further interventions or medications, and a potentially lower rate of new ischemic events in the follow-up period. The risk of future arrhythmias might also be reduced because of more complete revascularization. We sought to add to the limited data on single-vessel PCI versus multivessel PCI in the setting of non—ST-segment elevation (NSTE) ACS.
We retrospectively reviewed the records of 105,866 ACS patients undergoing PCI for multivessel CAD from 402 centers reported to the American College of Cardiology National Cardiovascular Database Registry between 2000 and 2004. We sought to identify the predictors of performance of single-vessel intervention versus multivessel PCI and determine their respective in-hospital outcomes. To adjust for numerous imbalances in baseline characteristics between the 2 groups, a propensity score for performance of single-vessel intervention versus multivessel PCI was constructed using demographic, clinical, and angiographic characteristics of the patients.
In the absence of randomized trials, the largest population of patients that could be evaluated to determine the frequency of single-vessel versus multivessel PCI for NSTE-ACS derives from the American College of Cardiology National Cardiovascular Database Registry. Of 1,639,174 admissions for PCI reported from a total of 402 centers between 2000 and 2004, 105,866 patients underwent either single-vessel or multivessel intervention; multivessel intervention was defined as PCI of lesions in more than 1 epicardial territory during the initial hospitalization in 1 setting without additional PCI procedures during that hospitalization.1 Patients excluded from analysis were those without ACS (n = 321,059), those with a history of CABG surgery (n = 74,300), those with single-vessel CAD (n = 155,907), those requiring staged PCI during the initial hospitalization (n = 5298), and those with missing angiographic information (n = 33). The peri-PCI complications reported were myocardial infarction (MI), cardiogenic shock, tamponade, bleeding (drop in hemoglobin >3 g/dL or hematoma >10 cm), dissection, pseudoaneurysm, arteriovenous fistula or occlusion of the entry vessel by thrombus requiring surgical repair, congestive heart failure, and renal failure (increase in serum creatinine >2.0 mg/dL, >50% increase from baseline, or need for dialysis). We created a propensity score for undergoing single-vessel versus multivessel PCI by including all the clinical and angiographic variables available in the registry.
Of the 105,866 patients with NSTE-ACS and multivessel CAD, 68% underwent single-vessel PCI and the remaining 32% underwent multivessel PCI. Although the 2 groups were statistically different in many baseline characteristics, the absolute differences in clinical and angiographic parameters were very small and, probably, of no clinical relevance. Patients with single-vessel PCI more frequently had non—ST-segment elevation MI (NSTEMI), 1 or more chronically occluded arteries, and a slightly lower ejection fraction. Those who underwent multivessel PCI more often had significant left main coronary artery stenosis or proximal left anterior descending artery stenosis (Table).
Important independent predictors of single-vessel PCI included presentation with NSTEMI versus unstable angina (odds ratio [OR], 1.29; 95% confidence interval [CI], 1.24-1.34), increasing age (OR, 1.09; 95% CI, 1.08-1.11), presence of 1 or more non-culprit total occlusions (OR, 1.25; 95% CI, 1.16-1.36), renal dysfunction, previous PCI, lower ejection fraction, and previous MI. The C statistic for the model was 0.70, indicating moderate discriminating ability.
Procedural success, defined as TIMI-3 flow and less than 20% stenosis at the end of PCI, was achieved in 91% of single-vessel PCIs and 88% of multivessel PCIs (P <.001). Multivessel PCI patients more frequently experienced periprocedural MI than the single-vessel group (Figure). In contrast, most other complications were more common in the single-vessel group, including heart failure (P = .006), emergency repeat PCI (P = .003), and unplanned CABG surgery (P <.001). In-hospital mortality was similar for both groups (unadjusted OR for single-vessel vs multivessel intervention, 1.08 [0.96-1.21; P = .18]; adjusted OR, 1.11 [0.97-1.27; P = .13]). Finally, no correlation between the propensity to perform single-vessel or multivessel PCI and in-hospital mortality was found using the Mantel-Haenszel test.
Other studies evaluating multivessel CAD intervention
The TACTICS-TIMI (Treat Angina with Aggrastat and Determine Cost of Therapy with an Invasive or Conservative Strategy—Thrombolysis in Myocardial Infarction)-18 trial included 2220 patients without persistent ST-segment elevation (NSTE-ACS) who were randomly assigned to early invasive intervention or to a conservative strategy. A post-hoc, subgroup analysis of this trial identified 290 patients with multivessel CAD, of which 66 underwent multivessel PCI and 224 underwent single-vessel PCI. By 6 months, the incidence of death (3%), MI (6.1%), and the composite of death, MI, or rehospitalization for ACS (21.2%) were insignificantly lower in the multivessel group than in the single-vessel group.2
In a retrospective analysis from the Mayo Clinic, patients with ST-segment elevation and NSTE-ACS were followed for 3 years. Of 1384 patients with multivessel CAD, 239 had undergone multivessel PCI and 1145 had undergone single-vessel PCI. There was no significant difference in survival at 1 and 3 years between the groups, and multivessel PCI was not associated with higher risk of major adverse ischemic events (hazard ratio [HR], 0.83; 0.62-1.10).3
In a 2007 report from the Cleveland Clinic, the authors sought to examine the safety and efficacy of multivessel stenting (culprit and non-culprit lesions) compared with stenting of only the culprit lesion in patients with multivessel disease presenting with NSTE-ACS.4 Of the 1240 patients evaluated between 1995 and 2005, 479 underwent multivessel intervention and 761 underwent single-vessel stenting. There were 442 events during a median follow-up of 2.3 years. Multivessel intervention was associated with a lower composite of death, MI, or revascularization after adjusting for baseline clinical and angiographic characteristics (HR, 0.80; 95% CI, 0.64-0.99; P = .04) and propensity score for the performance of multivessel PCI (HR, 0.67; 95% CI, 0.51-0.88; P = .004).
Based on our review of the American College of Cardiology National Cardiovascular Database Registry and others’ observations, it appears that multivessel PCI in the setting of NSTE-ACS is safe and can offer advantages compared with single-vessel PCI, including a lower composite of death, MI, or revascularization,4 as well as a lower incidence of heart failure, emergency repeat PCI, and unplanned CABG surgery.1 Adequate long-term follow-up of large cohorts is still lacking. The European and American Societies of Cardiology recommend performing multivessel PCI using an individualized approach5 or when the likelihood of success is high and the territory subtended by non-culprit lesions is moderate or large.6
The authors have no relationship with any commercial entity that might represent a conflict of interest with the content of this article.