Drug-eluting stents vs coronary-artery bypass grafting in multivessel coronary disease

Submitted by:
Justin Fox, MD
Fellow in Cardiology, Feinberg School of Medicine, Northwestern University

Reviewed by:
Dan Fintel, MD, FCCP
Steering Committee Member, Cardiovascular Medicine and Surgery NetWork
Professor of Medicine, Feinberg School of Medicine, Northwestern University

Hannan EL, Wu C, Walford G, et al. Drug-eluting stents vs coronary-artery bypass grafting in multivessel coronary disease. N Engl J Med 2008; 358:331-341

Given the rapid pace of advancement in the fields of interventional cardiology and cardiac surgery, there are scant data comparing current catheter-based and surgical revascularization techniques. In a large observational study entitled, “Drug-eluting stents vs coronary-artery bypass grafting in multivessel coronary artery disease,” Hannan and colleagues1 attempt to differentiate between the outcomes of these two procedures using data from the New York State Cardiac Surgery Reporting System and Percutaneous Coronary Intervention Reporting System. New York state residents who received a drug-eluting stent (DES) or coronary artery bypass graft (CABG) between October 1, 2003, and December 31, 2004, were included in the analysis and divided into four anatomic groups: two-vessel coronary artery disease (2v CAD) and three-vessel coronary artery disease (3v CAD), with and without a proximal left anterior descending (LAD) stenosis. Patients with previous revascularization, left main artery disease, or acute infarction within 24 h were excluded. The outcomes assessed were death, death or myocardial infarction (MI), and revascularization during an 18-month follow-up. Three prespecified subgroups also were analyzed: patients with diabetes, >80 years, or with left ventricular ejection fraction (LVEF) <40%. A propensity model was used to assess for treatment bias.

The authors identified 7,437 patients who underwent CABG and 9,963 patients who underwent multivessel percutaneous coronary intervention (PCI) with DES. The two groups were significantly different in terms of several baseline characteristics, with patients who underwent CABG tending to be older, male, and white, with lower ejection fractions and more extensive CAD. The group of subjects who underwent CABG was also found to have a higher prevalence of cerebrovascular disease, peripheral vascular disease, and COPD. After adjustment for baseline characteristics, the hazard ratio for death after CABG, compared with PCI, was 0.80 (0.65-0.97, p=0.03) in patients with 3v CAD and 0.71 (0.57-0.89, p=0.003) in patients with 2v CAD. There was a similar effect for the combined endpoint of death and MI in these groups and across anatomic subgroups of 2v CAD and 3v CAD, with and without a proximal LAD stenosis. The rate of repeat revascularization was higher in the group who received a DES (29.6% vs 5.2%), with approximately two-thirds of revascularizations after PCI occurring within the first 3 months. In the subgroup analysis, there appeared to be a benefit of CABG in preventing death or MI in those with diabetes, >80 years old, and with LVEF <40%. However, a mortality difference could not be shown in patients with diabetes, and the outcome of death reached only borderline significance in the groups comprising elderly subjects and those with left ventricular dysfunction.

This study uses a robust data set to compare CABG and DES. Once adjustment is made for baseline risk factors, there appears to be a clear benefit favoring CABG in terms of hard clinical endpoints. The results of this study are similar to a prior study by the same group using the same databases, which purportedly showed superiority of CABG to bare metal stenting in patients with multivessel disease.2 In the absence of a randomized controlled clinical trial or metaanalysis of smaller trials, an observational study may provide the next highest level of evidence to determine which treatment is superior.  However, there are inherent biases to this type of analysis. Although the authors adjusted for what were major differences in baseline characteristics between the groups, it is reasonable, when two groups are very different, to suspect that many unmeasured or unmeasurable confounding variables exist, causing a treatment-selection bias to persist in the risk-adjusted analysis. In general, the validity of an observational trial approach in drawing conclusions about treatment effects has been called into question, particularly in the wake of debacles, such as the data for hormone replacement therapy. Hormone replacement therapy appeared to prevent cardiovascular disease in a large observational study3 but showed no such effect in a large, multicenter, randomized trial.4 

There also are several issues related to the use of the New York state databases that may have introduced bias into this analysis. Patients who underwent CABG appeared to have higher baseline rates of cardiovascular disease, peripheral vascular disease, and COPD. However, this is not consistent with other observational cohorts of patients who underwent CABG or PCI, such as in the BARI (Bypass Angioplasty Revascularization Investigation) registry.5 This difference may exist because the preoperative evaluation for CABG often includes carotid duplex and spirometry, whereas no such testing is routinely performed before catheterization. This risk-ascertainment bias would make the patients who underwent CABG appear to be more sick in the risk adjustment analysis, thereby inappropriately driving the results toward the superiority of CABG. Additionally, the high rate of early revascularization after DES is not consistent with other data sets, such as ARTS (Arterial Revascularization Therapy Study) II.6 It is unclear why this difference exists. However, it is unclear whether the New York state data are counting planned, staged interventions as subsequent revascularizations. The existence of public reporting of procedural outcomes in New York may also affect practice patterns in the state, thereby limiting our ability to generalize these data to clinical practice elsewhere. Notably, this has been recently shown in an analysis of the SHOCK registry,7 where patients with cardiogenic shock after an MI in New York are much less likely than patients elsewhere to receive a revascularization procedure. Those who do receive CABG in this setting wait an average of 101 h from shock onset until surgery. This pattern of clinical practice violates published guidelines8 and may reflect practitioner concerns about performing procedures on high-risk patients in a state that publishes interventionalist and surgeon report cards.

Ultimately, while the analysis by Hannan and colleaues1 is one of the few published reports comparing DES and CABG, the limitations of an observational study design, coupled with the multiple troublesome biases that appear to be influencing these data, make it very difficult to draw any meaningful conclusions from this study.

References

1. Hannan EL, Wu C, Walford G, et al.  Drug-eluting stents vs coronary-artery bypass grafting in multivessel coronary disease.N Engl J Med2008; 358:331-341
2. Hannan EL, Racz MJ, Walford G, et al. Long-term outcomes of coronary-artery bypass grafting versus stent implantation.N Engl J Med2005; 352:2174-2183
3. Grodstein F, Manson JE, Colditz GA, et al. A prospective, observational study of postmenopausal hormone therapy and primary prevention of cardiovascular disease. Ann Intern Med2000; 133:933-941
4. Rossouw JE, Anderson GL, Prentice RL, et al. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the Women’s Health Initiative randomized controlled trial. JAMA 2002; 288:321-333
5. Feit F, Brooks MM, Sopko G, et al. Long-term clinical outcome in the Bypass Angioplasty Revascularization Investigation Registry: comparison with the randomized trial. Circulation 2000; 101:2795-2802
6. Serruys PW, Daemen J, Morice MC, et al. Three-year follow-up of the ARTS-II: sirolimus-eluting stents for the treatment of patients with multivessel coronary artery disease. Eurointervention 2007; 3:450-459
7. Apolito RA, Greenberg MA, Menegus MA, et al. Impact of the New York State Cardiac Surgery and Percutaneous Coronary Intervention Reporting System on the management of patients with acute myocardial infarction complicated by cardiogenic shock. Am Heart J 2008; 5:267-273
8. Antman EM, Anbe DT, Armstrong PW, et al. ACC/AHA guidelines for the management of patients with ST-elevation myocardial infarction: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines (Writing Committee to revise the 1999 Guidelines for the Management of Patients With Acute Myocardial Infarction). Circulation 2004; 110:e82-e292