ASCERT: Long-term Survival Advantage with CABG in Multivessel Disease

CHICAGO, IL—For older patients with multivessel disease, percutaneous coronary intervention (PCI) and coronary artery bypass graft (CABG) surgery yield equivalent mortality at 1 year, but CABG patients fare better in the longer-term, according to results presented March 27, 2012, at the annual American College of Cardiology (ACC)/i2 Scientific Session.

Preliminary results from the ASCERT (ACCF-Society of Thoracic Surgeons Database Collaboration on the Comparative Effectiveness of Revascularization Strategies) study were presented in February at the annual meeting of the Society of Thoracic Surgeons (STS) in Fort Lauderdale, FL, and the current findings were simultaneously published in the New England Journal of Medicine.

William S. Weintraub, MD, of the Christiana Care Health System (Newark, DE), and colleagues analyzed outcomes for patients with 2- or 3-vessel disease who underwent CABG (n = 86,244) or PCI (n = 103,549) from 2004 through 2007.

The CABG population was taken from the STS National Database and the PCI population from the ACC National Cardiovascular Data Registry. To obtain long-term follow-up, patient records were linked to administrative data from the Centers for Medicare and Medicaid Services. Propensity scores and inverse probability weighting were used and numerous clinical subgroups were prespecified.

CABG Bests PCI in the Long-Term

Follow-up ranged up to 5 years with a median of 2.67 years. At 1 year, there was no difference in adjusted mortality between the 2 groups. However, at 4 years, survival in the CABG group was higher vs. PCI (table 1).

Table 1. Short- and Long-Term Mortality, CABG vs. PCI

 

The 4-year risk ratios showed a benefit of CABG across subgroups defined according to sex, age, presence or absence of diabetes, body-mass index, presence or absence of chronic lung disease, ejection fraction, and glomerular filtration rate in both a high- and low-risk group. CABG was also associated with a benefit across subgroups defined according to quintile of propensity score for CABG. Thus, survival was with CABG even among patients whose propensity scores were most consistent with selection for PCI.

Linking Clinical, Administrative Data a Good Idea?

Dr. Weintraub and colleagues write that ASCERT “shows the potential benefits of linking large clinical and administrative databases to assess the comparative effectiveness of therapies in large patient populations. Perhaps the most compelling advantage of this approach is the ability to evaluate outcomes in broadly representative patient populations rather than the selected population of a randomized, controlled trial.” Data collected in this fashion provides continuity for future studies, they add, and linking clinical and administrative data “capitalizes on the advantages of each.”

While the advantages are aplenty, the researchers note several limitations. First, the elapsed time from the onset of an event was tabulated differently for some variables, which led to small differences in definitions between the databases. Next, the angiographic data are not as detailed as they are in randomized trials, limiting the ability to establish balance with respect to angiographic variables. Also, even after adjustment, the potential for unmeasured confounders remains. Lastly, because the data consist entirely of Medicare patients, the information might not be generalizeable to younger patients.

In an accompanying editorial, Laura Mauri, MD, MSc, of Brigham and Women’s Hospital (Boston, MA), questioned the value of observational registries in comparing treatment strategies if they require randomized trials to explain their results.

“Patients who consent to participate in the controlled framework of a randomized study are systematically different from those who do not, and unselected registries are the only way to examine the generalizability of results from randomized trials,” she said. “Observational studies provide detail on how and in whom treatments are being performed and how patient selection varies between treatments, but there is no substitute for randomized trials to eliminate selection bias between treatments.”

Going forward, she concluded, priority must be given to randomized trials for the “most salient questions regarding treatment strategy,” but efforts should be made to “simplify their design and conduct to be more inclusive and efficient.”

Analysis ‘Sets the Bar’

Panelist Alice K. Jacobs, MD, of Boston University School of Medicine (Boston, MA), said that this study “has certainly set the bar in conducting these analyses” since many efforts were made to eliminate residual confounders. “But I think it’s fair to say that in this data set, it’s probably impossible to [entirely eliminate confounders] because…all those inherent things that we integrate [into clinical practice] are not collected in this database.”

Dr. Weintraub said that he was surprised at the consistent results across all subgroups in favoring CABG at 4 years, and that he hopes these findings will lead to “more shared decision making between patients and their doctors, and also between surgeons and interventional cardiologists” in addition to greater dialogue.

“This is not the end of the story,” he continued. “This is the beginning of [studies] like this. There’s already been discussions both in the STS and the ACCF about what we can do to improve data collection so registries will be more robust and allow us to answer questions in greater detail with greater accuracy, and that is no small task.”

Still, Bernard J. Gersh, MB, ChB, DPhil, of the Mayo Clinic (Rochester, MN), was not completely convinced. “The problem is you can do multivariable propensity analyses to kingdom come, but it’s never going to eliminate the selection bias,” he said.

The problem with randomized trials, Dr. Weintraub said, is that they “can undo the criticality of decision making” in clinical practice. “We have to look at this all together or we’re never going to overcome all the problems.”

Study Details

Compared with patients undergoing CABG, those undergoing PCI were, on average, older and more often women. CABG patients more often had heart failure, diabetes, hypertension, chronic lung disease, cerebrovascular disease, a history of smoking, or peripheral arterial disease. PCI patients more often had prior MI or unstable angina requiring urgent procedures, and had higher ejection fraction than CABG patients. Patients in the PCI group more often had 2-vessel disease, while patients in the PCI group more often had 3-vessel disease.

Among patients who underwent PCI, 78% received DES, 16% received BMS, and 6% underwent the procedure without stent placement.

 

 

---

 

Sources:1. Weintraub WS, Grau-Sepulveda MV, Weiss JM, et al. Comparative effectiveness of revascularization strategies. N Engl J Med. 2012;Epub ahead of print.

 

2. Mauri L. Why we still need randomized trials to compare effectiveness. N Engl J Med. 2012;Epub ahead of print.

 

Related Stories:

• ASCERT: Long-term Survival Appears to Favor CABG Over PCI for Multivessel Disease
• Updated PCI, CABG Guidelines Highlight Interventional-Surgical Collaboration
• Study Suggests PCI Use Exceeds ACC-AHA Guideline Recommendations