Outcomes After Complete vs. Incomplete Revascularization of Patients With Multivessel CAD

Editor's Note: Commentary based on Garcia S, Sandoval Y, Roukoz H, et al. Outcomes after complete versus incomplete revascularization of patients with multivessel coronary artery disease: a meta-analysis of 89,883 patients enrolled in randomized clinical trials and observational studies. J Am Coll Cardiol 2013;62:1421-31.


Patients with multivessel coronary artery disease (MVCAD) represent a high-risk group of patients with stable ischemic heart disease. Among patients with multivessel disease who undergo revascularization with either coronary bypass grafting (CABG) or percutaneous coronary intervention (PCI), there have been numerous studies examining outcomes related to complete revascularization (CR) vs. incomplete revascularization (ICR). However, many of these studies have used a variety of inclusion/exclusion criteria, and there is to date only one, single center, 219-patient randomized controlled trial (RCT) directly assessing a strategy of CR compared with ICR for MVCAD.1 In an attempt to amalgamate the existing data on CR vs. ICR as this relates to outcomes of patients undergoing coronary revascularization procedures, Garcia et al. performed a systematic review and meta-analysis of the existing literature on CR vs. ICR.2


The authors used the PubMed database to identify observational cohort studies and RCTs published from 1970-2012 that enrolled patients with MVCAD referred for revascularization with PCI or CABG. Included studies both specifically compared outcomes of CR vs. ICR and reported long-term mortality rates. Notably, studies assessing non-culprit vessel PCI in ST elevation myocardial infarction, studies examining outcomes in chronic total occlusions not specifically reporting on completeness of revascularization, studies examining CABG reoperation, and small single-center studies were excluded from the analysis. In addition to long-term mortality, other assessed outcomes included the occurrence of myocardial infarction (MI) and repeat revascularization procedures. Standard data collection/abstraction methodology was employed, study heterogeneity was assessed with the I2 statistic, and random effects methodology was utilized in order to generate summary estimates of effect.


A total of 35 studies were included in the analysis, enrolling 89,883 patients (50.5% with CR and 49.5% with ICR) with a mean follow-up of 4.6 years. The majority of patients in the included studies were revascularized with PCI (71%); 29% of included patients were revascularized with CABG. Notably, ICR was more common after PCI than after CABG (56% vs. 25%; p<0.001).

In the primary analysis, CR was associated with a 29% reduction in long-term mortality compared with ICR (RR: 0.71; 95%CI: 0.65-0.77; p<0.001). Although there was substantial heterogeneity among studies (I2=71%), the mortality reduction with CR compared with ICR was consistent among analyses examining revascularization by either PCI or CABG, and was also consistent across varying study designs and definitions of CR.

In addition to a reduction in mortality, CR was also associated with lower rates of MI (18 studies, RR: 0.78, 95% CI: 0.68-0.90; p=0.001) and repeat coronary revascularization procedures (20 studies: RR: 0.74, 95% CI: 0.65-0.83; p<0.001). There was less heterogeneity among studies with respect to the assessment of MI (I2=19%), but greater heterogeneity among studies with respect to the assessment of repeat revascularization procedures (I2=65%). Of note, the associations of CR with a reduction in MI and repeat revascularization procedures were only manifest among PCI-treated patients, but not patients treated with CABG, perhaps due to a smaller number of patients treated with CABG in the included analyses.


The authors observed that CR occurred in approximately half of the patients enrolled in these included studies and was less frequently observed among patients revascularized with PCI compared with CABG. However, given the findings of reduced mortality with CR compared with ICR across all included studies, the authors concluded that—irrespective of the revascularization strategy—CR might be the preferred strategy for patients with MVCAD.


The goal of coronary revascularization procedures is the restoration of normal blood flow and oxygen delivery to ischemic myocardium that has a compromised arterial supply due to obstructive epicardial coronary stenoses. Among patients with obstructive coronary artery disease, patients with MVCAD typically have the greatest amounts of ischemic myocardium, and additionally have the highest rates of subsequent adverse outcomes. Because of the adverse prognosis of MVCAD, appropriately performed revascularization procedures in this patient population potentially afford the treating clinician the smallest number of patients needed to treat in order to effect significant clinical benefit.

While it seems intuitive to revascularize all ischemic territories for patients with MVCAD, and an approach of CR was associated with improvements in survival outcomes in early CABG studies,3,4 a strategy of CR can also be associated with a theoretically greater incidence of adverse events (e.g. longer procedure times, more periprocedural events). Additionally, some of the prior data regarding CR vs. ICR has been contradictory. For example, in one of the more contemporary experiences for patients with complex MVCAD—the SYNTAX (Synergy Between PCI With Taxus and Cardiac Surgery) trial—CR was associated with lower rates of major adverse cardiac events and repeat revascularization procedures, but not death or MI. While SYNTAX enrolled a far more complex MVCAD population than many prior trials and as a result had a low overall rate of CR in both CABG and PCI patients enrolled in the trial (43% and 37% respectively),5,6 these findings have led some to question the importance of CR against a backdrop of modern medical therapy. Some of this belief stems from the failure of a strategy of revascularization (albeit not CR) to demonstrate benefit over optimal medical therapies in two recent randomized trials of stable ischemic heart disease.7,8 However, both of these trials enrolled "randomizable" lower-risk patients without a definite need for revascularization based upon coronary angiography, and not specifically with severe MVCAD. Additionally, both trials employed a strategy-based design allowing crossover but analyzed by intent-to-treat, which clouds the ability to make a pure comparison of outcomes related specifically to revascularization vs. no revascularization (an ultimate proxy for ICR).

By performing a systematic meta-analysis of CR vs. ICR studies, Garcia et al quantitatively amalgamated numerous analyses published over four decades and demonstrated a consistently beneficial effect of CR compared with ICR for patients with MVCAD. The consistency of results within the included studies is generally reassuring to those proponents of CR, but key limitations inherent to the study design are relevant. First, the included studies were virtually all non-randomized comparisons. The results of RCTs and observational analyses of non-randomized treatment decisions can be strikingly different, particularly if covariates related to the outcome of interest also influence the treatment decision/process.9 Thus, is possible that systematic selection bias within the included studies influenced the results. For example, patients in whom CR was unable to be achieved likely had a greater burden of residual CAD (or may have been more complex in some other way), and as a result may have subsequently suffered more adverse clinical events. Second, significant study heterogeneity was observed (and in fact quantified using the I2 statistic). Despite attempts to perform sensitivity analyses in order to address the study heterogeneity, the overall variability across studies is not uncommonly observed when combining observational analyses, and weakens the overall findings. Other limitations regarding the lack of standard definitions of CR, ICR, and other key variables are also raised and addressed by the authors in the manuscript itself.

Ultimately, despite the findings in support of a strategy of CR for patients with MVCAD, the authors rightly state that an appropriately powered RCT is needed to definitively answer this question. The ongoing, multicenter ISCHEMIA trial—randomizing patients with moderate-severe ischemia to an invasively guided vs. medical therapy-based management strategy—should provide further insights regarding the impact of ICR and residual disease upon clinical outcomes. Nonetheless, it is important to point out that particularly for high-risk patients, studies such as this meta-analysis support the current standard of care (diagnostic angiography followed by CR for MVCAD patients), and that the conservative arm of the ISCHEMIA trial (medical therapy without anatomic risk stratification for patients with moderate-severe ischemia) remains "experimental" whether within or outside of the confines of that trial.

In the interim until that trial is complete, a strategy of CR for the high-risk subset of patients with MVCAD should remain the goal, provided that the patient is an appropriate candidate for revascularization and the revascularization can be safely and efficaciously undertaken.


  1. Ijsselmuiden AJ, Ezechiels J, Westendorp IC et al. Complete versus culprit vessel percutaneous coronary intervention in multivessel disease: a randomized comparison. Am Heart J 2004;148:467-74.
  2. Garcia S, Sandoval Y, Roukoz H et al. Outcomes after complete versus incomplete revascularization of patients with multivessel coronary artery disease: a meta-analysis of 89,883 patients enrolled in randomized clinical trials and observational studies. J Am Coll Cardiol 2013;62:1421-31.
  3. Jones EL, Craver JM, Guyton RA, Bone DK, Hatcher CR, Jr., Riechwald N. Importance of complete revascularization in performance of the coronary bypass operation. Am J Cardiol 1983;51:7-12.
  4. Bell MR, Gersh BJ, Schaff HV et al. Effect of completeness of revascularization on long-term outcome of patients with three-vessel disease undergoing coronary artery bypass surgery. A report from the Coronary Artery Surgery Study (CASS) Registry. Circulation 1992;86:446-57.
  5. Serruys PW, Morice MC, Kappetein AP et al. Percutaneous coronary intervention versus coronary-artery bypass grafting for severe coronary artery disease. N Engl J Med 2009;360:961-72.
  6. Head SJ, Mack MJ, Holmes DR, Jr. et al. Incidence, predictors and outcomes of incomplete revascularization after percutaneous coronary intervention and coronary artery bypass grafting: a subgroup analysis of 3-year SYNTAX data. Eur J Cardiothorac Surg 2012;41:535-41.
  7. Boden WE, O'Rourke RA, Teo KK et al. Optimal medical therapy with or without PCI for stable coronary disease. N Engl J Med 2007;356:1503-16.
  8. Frye RL, August P, Brooks MM et al. A randomized trial of therapies for type 2 diabetes and coronary artery disease. N Engl J Med 2009;360:2503-15.
  9. Kirtane AJ, Gupta A, Iyengar S et al. Safety and efficacy of drug-eluting and bare metal stents: comprehensive meta-analysis of randomized trials and observational studies. Circulation 2009;119:3198-206.

Keywords: Coronary Artery Bypass, Coronary Artery Disease, Coronary Disease, Myocardial Ischemia, Percutaneous Coronary Intervention

< Back to Listings