What should the cardiologist tell the patient facing revascularization for three-vessel disease?

Over time PCI has been used progressively more frequently compared with CABG, and despite the lack of randomized clinical trials that have been sufficiently powered, proponents of PCI have claimed equivalent survival with a strategy of initial PCI compared with a strategy of initial CABG. One example is the ARTS study that set out to compare CABG and PCI with stenting in patients with multivessel disease. A total of 1 205 patients were randomly assigned to CABG or stent implantation 1. In their 5-year follow-up data, the authors conclude that “there was no difference in mortality between stenting and surgery for multivessel disease” 2. However, the ARTS trial was not designed or powered to assess the relative benefit of the two treatments on death as a sole end-point 1, 2. Data from a 2003 meta-analysis of randomized controlled studies of CABG versus PCI actually showed a statistically significant survival advantage for CABG at 5 and 8 years in the subgroup of patients with multivessel disease 3. The long-term survival advantage for CABG is consistent with findings from early and late observational studies 4–6. One recent study deserves special attention. The New York cardiac registers identified nearly 60 000 patients with multivessel disease who underwent CABG or PCI from 1997 to 2000. The rates of adverse outcome were adjusted by means of proportional-hazards methods to account for differences in patients′ severity of illness before revascularization. The authors concluded that CABG was associated with higher adjusted rates of survival than PCI plus stenting within 3 years of follow-up 7.

In this issue of the Scandinavian Cardiovascular Journal, Per Mølstad presents register data of 10 815 patients (PCI 6 366, CABG 4 449) from the Feiring Heart Clinic during 1999–2005 8. Indeed, observational studies comparing different treatment strategies are susceptible to selection bias but this problem has been partly opposed by a propensity analysis. More than 98% of the patients were followed for a median of 3.3 years, and covariate adjusted survival was significantly improved by CABG compared to PCI in patients with 3-vessel disease with (hazard ratio 0.40) and without diabetes (0.61) 8. Opponents to these observations will emphasize that no statistical method can eliminate the weakness’ inherent in a non-randomized design, and surely this is true. On the other hand the patients included in a register most likely reflect the real world. Other opponents will be of the opinion that the low use of drug-eluting stents in PCI patients makes the Norwegian results obsolete. Indeed, drug-eluting stents have reduced the frequency of restenosis, but whether this fact alone is sufficient to switch more and more patients with 3-vessel disease from CABG to PCI remains controversial. At present there are no prognostic data on comparisons of CABG and PCI with drug-eluting stents, but trials are ongoing. However, in these trials equivalent survival rates from PCI with drug-eluting stents compared with CABG can only be obtained if the reduction in the frequency of restenosis is reflected in a lower mortality. So far, the evidence from randomized trials does not indicate that drug-eluting stents will influence mortality or myocardial-infarction rates favourably 9. Thus, although many cardiologists are of the opinion that PCI with stenting should be the preferred therapy in patients with 3-vessel disease and PCI amenable stenoses, we must admit that the scientific base supporting this view is still not there.

We as cardiologists have to remember that most acute myocardial infarctions do not arise from significant coronary stenoses. More than 2/3 of myocardial infarctions results from the progression of a vulnerable plaque which may not necessarily be visible on a coronary angiogram prior to plaque rupture and acute thrombosis 10. Therefore, our main problem in this setting is the lack of technical methods to identify the vulnerable plaque. Even if we were able to pin point the “next” culprit lesion, we would still not know how to treat it. Maybe CABG is superior to PCI simply because it bypasses several vulnerable plaques that potentially can develop into culprit lesions over time. Contrary, PCI including stenting only treat the focal area of most significant occlusion at the time of revascularization 11. Typically these lesions are stable and relative rarely turns into “culprit” lesions 10.

Whether we like it or not, the bulk of evidence from clinical randomized trials, observational studies and meta-analyses indicates that patients with 3-vessel disease in need off coronary revascularization have a better survival from a strategy of initial CABG compared with initial PCI. With the growing evidence of a higher frequency of late stent thromboses, it is not likely that an increased use of drug-eluting stents will change this pattern in a PCI favourable direction. Stent thrombosis is a serious adverse event commonly associated with acute myocardial infarction or sudden death. The Food and Drug Administration (FDA) have proposed that the absolute risk of thrombosis with drug-eluting stens is  < 2% throughout the first 3 years after implantation, when these are used for the approved indications 12. Such “on-label” indications for drug-eluting stents include only the treatment of discrete, previously untreated lesions in native coronary vessels, like those studied in the pivotal clinical trials. Nevertheless, more than 60% of drug-eluting stent use is off-label (complex conditions such as multiple vessel disease or acute myocardial infarction and complex lesions including chronic total occlusions, bifurcating lesions and saphenous-vein bypass grafts) 13. Unfortunately, it appears that compared with on-label indications, off-label use of drug-eluting stents is associated with increased risks of both early and late thrombosis, as well as death or myocardial infarction 12.

The interventional cardiologist has the opportunity to talk to the patient immediately after the coronary anatomy has been defined. With the increasing trend of doing PCIs in patients with 3-vessel disease, few of the involved patients might get the opportunity to talk to a cardiac surgeon. Consequently, the interventional cardiologist may be the only physician to counsel the patient and provide information about advantages and disadvantages associated with PCI procedures versus CABG. I totally agree with Per Mølstad 8 that the patient with 3-vessel disease should be informed that the summary of present scientific evidence suggest that CABG has a long-term survival benefit compared with PCI. In my opinion the patient should also be told that a strategy of initial CABG results in a higher likelihood of being free from angina and a reduced need for repeat revascularization. Additionally, the interventional cardiologist has an obligation to inform the patient that revascularization with PCI and stents – and in particular drug-eluting stents – will result in a 12 months period of clopidogrel treatment in addition to aspirin. The importance of a strict adherence to dual antiplatelet therapy should be emphasized, and the increased risk of bleeding complications must be addressed. Finally, the patient should be informed that in the past few years CABG without cardiopulmonary bypass and cardiac arrest (“off pump”) has been much more common.