Outcomes Comparable for PCI vs. CABG in Left Main CAD

Capodanno D. J Am Coll Cardiol. 2011;58:1426-1432.

Issue: January/February 2012

Patients with left main CAD treated with PCI had nonsignificantly different 1-year rates of MACCE, death and MI, but had a lower risk for stroke and a higher risk for target vessel revascularization compared with those treated with CABG, researchers of a meta-analysis reported.

Searching Medline and Cochrane databases from January 1980 to April 2011, researchers identified four randomized clinical trials that included a study population of left main CAD, randomization to PCI vs. CABG, and a report of safety and efficacy outcomes. Primary endpoints were 1-year incidence of MACCE, defined as death, MI, TVR or stroke.

Across the four studies, there were 1,611 patients, 809 of whom were assigned to PCI and 802 to CABG.

Overall, complete revascularization was achieved in 71% of patients treated with PCI vs. 76% of patients treated with CABG. In the PCI group, researchers found a nonsignificant trend toward a higher risk for MACCE events (OR=1.28; 95% CI, 0.95-1.72) and a higher 1-year rate of TVR (OR=2.25; 95% CI, 1.54-3.29).

Researchers also reported a less frequent occurrence of stroke (OR=0.15; 95% CI, 0.03-0.67) in patients treated with PCI, but found no significant differences in death (OR=0.74; 95% CI, 0.43-1.29) or MI (OR=0.98; 95% CI, 0.54-1.78).

Charanjit S. Rihal

This short-term analysis raises several interesting issues. First, are randomized trials in left main disease even ethical when CABG surgery has proven efficacy in this condition as compared to medical therapy? To answer this question, we must look at the original trials of CABG surgery vs. medical therapy. One trial published in The Lancet in 1994 by Yusuf et al enrolled 2,649 patients with CAD with only 6.6% (or about 175 patients) having left main disease. Moreover, since the time of the original trials, both medical and surgical therapies have improved significantly; and a new therapy, PCI, is now widely available for CAD. In the CABG vs. medical therapy trials, it took over 2 years for the long-term survival benefit of CABG to become apparent. Thus, the current meta-analysis with only a 1-year time horizon could be criticized for inadequate follow-up; and it will be interesting to see whether differences emerge with follow-up at 2 to 4 years.

A second important point relates to why a meta-analysis of only about 1,600 patients should be considered when we have large registries, such as the New York state registry with tens of thousands of patients. For example, in a 2005 study in The New England Journal of Medicine, Hannan et al showed that the long-term outcomes of CABG surgery are superior to PCI. Close inspection of these data, however, reveals that 85% of all three-vessel disease patients underwent CABG, from which it is safe to infer that the CV community, at least in New York state in the late 1990s, reached consensus that CABG is preferred for severe disease. Thus, the 15% of patients who did not undergo CABG did so for unclear reasons, such as being poor surgical candidates. This selection bias in registries (irrespective of size) cannot be overcome by multivariable adjustment since not all possible variables can be measured, and since many important variables are unequally or haphazardly measured. Thus, randomized trials in left main disease not only are ethical but also are imperative in improving our understanding of the relative merits of currently available therapeutic options.

How do we apply the data in clinical practice? I am a strong advocate of the ‘heart team’ concept for treatment of complex CVDs. Many, and probably most, patients are fully capable of understanding the relevant tradeoffs between PCI and CABG, both acutely and in the long term. The meta-analysis by Capodanno, along with the results of the larger SYNTAX trial, provides relevant data that can inform patient/physician discussions. Seeking the patient’s input and preferences are crucial in contemporary practice and likely to become more important as many newer therapies are developed and as patients become more and more knowledgeable. Decision aids to clarify the absolute risks involved (for example, a 1.6% absolute risk difference in stroke in favor of PCI vs. a 6% absolute risk difference in favor of CABG for redo procedures) can further help contextualize the relevant choices. From the patient standpoint, it is equally valid to point out that 99.9% of PCI patients and 98.3% of CABG patients are free of stroke at 1 year. Whether this is important can best be answered by the patient, and engaging patients in a discussion of tradeoffs and outcomes is crucial.

– Charanjit S. Rihal, MD
Cardiology Today Intervention Editorial Board member
Disclosure: Dr. Rihal reports no relevant financial disclosures.