Download this article's Factoid in PDF (& PPT for Gold Subscribers)

For patients with less complex left main coronary artery disease, percutaneous coronary intervention (PCI) and coronary artery bypass grafting (CABG) yield comparable 1-year rates of death and MI, according to a meta-analysis of randomized trials published in the September 27, 2011, issue of the Journal of the American College of Cardiology. However, PCI is associated with more repeat revascularization, while CABG is linked to more strokes.

Investigators led by Davide Capodanno, MD, of Ferrarotto Hospital (Catania, Italy), searched the medical literature for prospective randomized controlled trials (or prespecified subanalyses thereof) comparing PCI with CABG for left main disease. Four trials, spanning the period from 2008 through 2011, qualified for analysis:

• LEMANS (n = 105)
• SYNTAX Left Main (n = 705)
• Boudriot et al (n = 201)
• PRECOMBAT (n = 600)

The meta-analysis included 1,611 patients, 809 randomized to PCI and 802 to surgery. First-generation DES were implanted in 96% of PCI patients, while left anterior internal mammary artery grafts were used in 95% of CABG patients.

Repeat Revascularization, Stroke the Main Differentiators

At 1 year, there was no difference between the PCI and CABG arms for the primary endpoint of MACCE (composite of death, MI, stroke, or TVR) or the individual endpoints of death and MI. However, stroke favored PCI, while TVR favored surgery (table 1).

Table 1. One-Year Outcomes

The results were consistent on meta-regression analysis across numerous subgroups analyzed according to:

• Number of patients in the trial
• Diabetes status
• Distal involvement
• Mean Syntax score
• Mean logistic EuroScore
• Complete revascularization 

Three-Vessel Disease a Tipping Point for CABG

The SYNTAX and PRECOMBAT trials reported MACCE stratified by the number of vessels involved. No differences were seen between PCI and CABG for isolated left main disease (OR 0.66; 95% CI 0.18-2.40; P = 0.53) or left main disease plus 1 vessel (OR 0.58; 95% CI 0.22-1.51; P = 0.26) or 2 vessels (OR 1.28; 95% CI 0.74-2.23; P = 0.38). However, 3-vessel involvement favored CABG over PCI (OR 1.80; 95% CI 1.06-3.07; P = 0.03).

Because the current meta-analysis included only randomized data, the risk of confounding is low, the authors say. Moreover, the underlying trials reflect contemporary practice.

CABG may yet show a long-term advantage over PCI, the investigators acknowledge, so “a fair assessment of the 2 revascularization techniques is likely to require longer follow-up than in the present study,” they write.

However, “[b]ased on the present study, revision of the guidelines regarding left main PCI is warranted, raising the level of evidence of current recommendations from B to A,” Dr. Capodanno and colleagues argue.

Strong Evidence for Equipoise

The meta-analysis may add some statistical weight to previous findings, but there is very little heterogeneity to reconcile among the included studies in the first place, said Jeffrey W. Moses, MD, of Weill Cornell Medical College (New York, NY). “With more than 1,600 randomized patients in this study and an enormous amount of data from registries, I think the current evidence for equipoise [between PCI and CABG] is pretty strong,” he told TCTM in a telephone interview.

“That’s why we’re doing [the randomized] EXCEL [trial],” he said. “We don’t have any qualms about performing PCI in low- and moderate-complexity left main patients [EXCEL is enrolling only those with a Syntax score equal to or less than 32] and comparing it to surgery. Frankly, I think EXCEL is just being done to seal the deal.”

Dr. Moses reported that in the EXCEL trial, PCI patients will receive Xience stents, and the previously documented reductions in restenosis and thrombosis achieved by the newer device compared with Taxus, which was used in the SYNTAX trial, are magnified in complex lesions. 

Making the Primary Endpoint More Patient Oriented 

Also, in EXCEL, TVR, the main driver of MACCE in trials of PCI vs. CABG in left main disease, has been removed from the primary endpoint. That is because patients do not give repeat revascularization the same weight as the other component endpoints, he said, adding that most are willing to trade about a 1-in-20 chance of needing another angioplasty in the future for avoiding the trauma, morbidity, and long recovery time of surgery. In fact, Dr. Moses said, the only reason TVR was included in the primary endpoint in previous trials was to provide sufficient power in terms of events.

The patient’s upfront risk of surgery is a factor clinicians regularly consider in deciding on a revascularization strategy, although it is rarely discussed in papers, Dr. Moses said. “That can shift the scale,” he explained.

Dr. Moses agreed with the authors that 1-year follow-up is inadequate, but he noted that longer-term registry data have not found a divergence in mortality rates between PCI and surgery except in patients with higher Syntax scores and triple-vessel disease. It is unclear whether clinical and anatomic factors such as completeness of revascularization, diffuse disease, and total occlusion may be driving the prognostic value of the Syntax score, he said, adding that such variables may be more important than the score itself.

“If you look at all the studies—between the registries and randomized trials there must be about 50 over the last 5 years—the consistency in their findings makes the conversation [with a surgeon and other members of the patient’s ‘heart team’] pretty easy,” Dr. Moses said. “For nondistal left main lesions without extensive coronary disease, we’re very comfortable offering PCI.”

With regard to the guidelines, he said he did not believe that PCI needs to be restricted to patients at high risk of surgery, as they currently suggest. Moreover, he added, based on the totality of evidence, the current IIb recommendation should be elevated to IIa.

Related Stories:

• SYNTAX Left Main Published: Equivalent Outcomes at 1 Year for PCI, CABG
• Long-term Safety of PCI vs. CABG for Left Main Disease Depends on Syntax Score
• DES, CABG Yield Similar Outcomes for Left Main Disease