Patient variables may tip scale in multivessel CABG-PCI debate

Multivessel CABG may offer a survival benefit over multivessel PCI for the “average” patient, according to a study published online April 23 in the Annals of Internal Medicine. But for patients with no diabetes, heart failure, tobacco use and peripheral artery disease, multivessel PCI may provide better life expectancy.

Randomized clinical trials comparing multivessel CABG and multivessel PCI typically enroll a selected patient population that may not reflect real-world practice. The sample sizes also may not allow for subgroup analyses to assess treatment effects based on the clinical characteristics of patients. For instance, in 10 randomized trials, only 3 percent of patients had a history of heart failure, wrote Mark A. Hlatky, MD, of Stanford University School of Medicine in Stanford, Calif., and colleagues.

To analyze patient variables and their association with multivessel CABG and multivessel PCI, Hlatky et al designed an observational comparative study based on claims data from Medicare beneficiaries with multivessel coronary disease and an index coronary revascularization procedure between 1992 and 2008. They used a propensity score analysis to control for potential treatment selection bias.

They identified 251,553 eligible patients (194,223 with multivessel CABG and 57,330 with multivessel PCI) and matched 10,080 patient pairs for the analysis. The matched populations had similar baseline patient characteristics. They found that race, diabetes, tobacco use, peripheral artery disease and heart failure modified all-cause mortality over a five year follow-up for CABG compared with PCI.

CABG had a mean increase in life expectancy of 0.053 life-years compared with PCI, but the estimated life-years gained varied widely across the patient population. Patients with diabetes, heart failure, tobacco use and peripheral artery disease had a survival advantage from CABG while patients without those characteristics were predicted to have better survival after PCI.

Citing their findings and results from clinical trials, Hlatky et al wrote, “This body of evidence suggests that the use of CABG rather than PCI is likely to reduce mortality for the average patient with multivessel coronary disease.”

They noted that it was important to not only consider the treatment effects for the average patient, as is often provided through results from randomized trials, but also effectiveness for specific patients. “This variation in the comparative effectiveness of CABG and PCI underscores the importance of individualizing treatment.”

Their analysis included 18,000 patients with peripheral artery disease and 13 percent of patients in their study had heart failure, they observed. These sample sizes provided sufficient power for subgroup analyses, but they added that the data also lacked important clinical details and information on medications and behaviors during follow-up.

“This study provides strong evidence that clinical characteristics modify the comparative effectiveness of CABG and PCI on mortality, especially for diabetes,” the researchers wrote. “Our study suggests that additional factors, particularly smoking, peripheral artery disease and heart failure, also modify the CABG-PCI treatment effect.”

They encouraged physicians to consider the clinical details of each patient with multivessel coronary disease when making treatment recommendations.