Less Stroke With PCI at 5 Years, but CABG Risks Front-loaded in First 30 Days

People who get CABG surgery for coronary artery disease have a higher risk of stroke at 5 years than those who undergo PCI, an increased risk that was largely confined to patients with multivessel disease and diabetes, according to the results of a new meta-analysis.

Beyond the 30-day postprocedural period, however, those differences disappear, reported Stuart Head, MD, PhD (Erasmus Medical Center, Rotterdam, the Netherlands), and colleagues July 16, 2018, in the Journal of the American College of Cardiology. In an analysis restricted to between 31 days and 5 years, the risk of stroke was similar between the two revascularization strategies: 2.2% with PCI and 2.1% with CABG surgery (HR 1.05; 95% CI 0.80-1.38).

Michael Farkouh, MD (University of Toronto, Canada), one of the study investigators, said the meta-analysis can help inform physicians and patients about the trade-offs between the two coronary revascularization strategies. “In the end with CABG and PCI, the rates of stroke after the first 30 days are comparable,” said Farkouh. “The problem is getting somebody through those first 30 days.”

The bottom line, he told TCTMD, “is that the risk of stroke needs to be incorporated into the risk-benefit equation for every patient.”

Patient-Level Data From 11 Randomized Trials

The new meta-analysis included pooled, patient-level data from 11 randomized trials—ERACI II, ARTS, MASS II, SoS, SYNTAX, PRECOMBAT, FREEDOM, VA-CARDS, BEST, NOBLE, and EXCEL—comparing CABG versus PCI. In total, 5,753 patients were randomized to PCI and 5,765 were treated with surgery. During a mean follow-up of 3.8 years, 293 strokes occurred.

At 30 days, the stroke rate was 0.4% among those treated with PCI and 1.1% in those who underwent CABG surgery (HR 0.33; 95% CI 0.20-0.53). At 5 years, the risk of stroke remained significantly higher among those who underwent surgery, with stroke occurring in 2.6% of those treated with PCI and 3.2% of those treated surgically (HR 0.77; 95% CI 0.61-0.97).

The problem is getting somebody through those first 30 days. Michael Farkouh

The risk of stroke at 5 years was lower in diabetic patients treated with PCI than with CABG surgery (2.6% versus 4.9%, respectively; HR 0.52; 95% CI 0.37-0.75), but there was no difference in stroke risk at 5 years in those without diabetes. Similarly, the risk of stroke at 5 years was lower in patients with multivessel disease treated with PCI compared with CABG, but there was no difference in stroke between the two approaches in those without multivessel disease.

Farkouh, the lead investigator of FREEDOM, a 5-year randomized, controlled clinical trial showing that CABG surgery reduced the risk of death and MI compared with PCI among patients with multivessel disease and diabetes, said that stroke after CABG is inevitably higher in FREEDOM-like patients given the nature of their disease.

“Diabetics have more atherothrombotic disease, are more likely to have atrial fibrillation, and are more likely to have an acute coronary syndrome,” said Farkouh. “They are higher-risk people. As a result, they have a perioperative stroke rate that is higher. Multivessel disease with diabetes is a marker of bad disease in the rest of the body, including the brain.”

While CABG surgery is recommended in FREEDOM-like patients, some individuals refuse the procedure because they’re concerned about stroke. “It’s not minimal to a patient,” said Farkouh. “They want to live free of stroke. The worst thing that can happen to an elderly person is not necessarily dying, but having a devastating stroke. That’s why this needs to be put into an individual formula to understand the patient’s stroke risk.”

In the meta-analysis, stroke in the first 30 days was associated with particularly poor long-term clinical outcomes. Patients who had an early stroke following PCI and CABG surgery had a significantly higher risk of death at 5 years. In fact, 41.5% of surgical patients and 45.7% of PCI-treated patients who had an early stroke died by 5 years. Comparatively, only 8.9% and 11.1% of CABG and PCI patients who did not have a stroke were dead by 5 years. 

Focusing on Perioperative Approaches

Subodh Verma, MD, PhD (University of Toronto), a cardiothoracic surgeon who was not involved in the present analysis, praised the investigators, noting that the patient-level analysis strengthens their conclusions. Like Farkouh, he highlighted the front-loaded risk of stroke that was primarily observed in individuals with diabetes and multivessel disease.

“Surgeons must focus their attention to perioperative approaches—safer aortic cannulation techniques, more intensive postoperative monitoring for atrial fibrillation beyond hospital discharge, etc—to mitigate these risks,” he told TCTMD. And again like Farkouh, Verma stressed the importance of assessing the risks and benefits of each revascularization approach in the individual patient. “The very patients who have a mortality benefit with CABG over PCI, for example, those with diabetes and multivessel disease, are the ones at greater risk of stroke,” he noted.

Amar Krishnaswamy, MD, and Samir Kapadia, MD (Cleveland Clinic, OH), who wrote an editorial accompanying the study, state that the findings of higher stroke risk with CABG confirm data observed in real-world registries, including the American College of Cardiology’s National Cardiovascular Data Registry. Focusing on the stroke risk with PCI, they find the results reassuring given that there appears to be no late “catch-up phenomenon for stroke among PCI patients due to the generally higher need for repeat revascularization.”

The editorialists recommend physicians engage in thoughtful conversations about the relative risks and benefits of CABG surgery and PCI. Close postprocedural follow-up with a focus on lipid management and oral anticoagulation for atrial fibrillation is advised. They also suggest longer use of dual antiplatelet therapy in diabetic patients after coronary revascularization as well as surgical techniques to minimize stroke risk, such as minimizing proximal aortic anastomoses and performing off-pump CABG, among other recommendations.

To TCTMD, Farkouh noted there are international differences seen in the rate of stroke in the first 30 days after surgery. Variations in care or treatment might contribute to these differences.

“There are clear regional differences and in most clinical trials we see that the excess of stroke occurs mainly outside of North America,” said Farkouh. “We can’t explain it on the basis of the operator because they’re all good operators. We think it might be due to practices like how we monitor for atrial fibrillation, how we restart anticoagulation, and how we clamp the aorta. We’re looking into those mechanisms now.”