CKD Patients With Left Main Disease See Similar Outcomes at 3 Years After PCI vs CABG: EXCEL

Chronic kidney disease (CKD) in patients being treated for left main disease is linked to poorer outcomes, with similar results offered by both CABG surgery and PCI over the long term, according to a prespecified subgroup analysis from the EXCEL trial. But researchers also found that PCI offers up-front advantages for patients with CKD: fewer instances of acute renal failure, less bleeding, fewer blood transfusions, and a reduced risk of death, MI, and stroke in the first 30 days.

The 3-year findings echo those of the main EXCEL trial, which were presented at TCT 2016, the same meeting where the NOBLE trial, in contrast, showed an edge for CABG in patients with left main disease.

Gennaro Giustino, MD (Icahn School of Medicine at Mount Sinai and Cardiovascular Research Foundation, New York, NY), lead author of new analysis, told TCTMD it’s well known that patients with CKD are generally at higher risk of experiencing adverse events after revascularization, whether PCI or CABG.

"They have a lot of comorbidities that place them at risk for many types of adverse events, including bleeding, vascular complications, stroke, and mortality," he explained. "At the same time, chronic kidney disease in itself is a strong risk factor for worsening renal function [requiring dialysis], and obviously that is going to be a very important determinant for future prognosis."

Clinicians must carefully weigh the choice of revascularization strategy in these patients, what with the risks that contrast media during PCI and cardiopulmonary bypass during CABG can pose to their kidneys, Giustino advised. But randomized trial data to guide these decisions are lacking, he said. While the FREEDOM trial, for instance, shed light on how best to treat patients with diabetes, no such specific studies exist in the arena of CKD. In addition, patients with CKD are usually excluded from and underrepresented in randomized controlled trials.

Thus, the new EXCEL results, published in the August 14, 2018, issue of the Journal of the American College of Cardiology, add much to the literature, Giustino said.

Another complicating factor, one pointed out in an accompanying editorial by Patrick T. O'Gara, MD (Brigham and Women's Hospital, Boston, MA), is that the relative risks and benefits of PCI versus CABG shift over time.

As O’Gara asks: “Do the comparative differences in outcomes between PCI and CABG in the context of CKD enable the multidisciplinary heart team to make better decisions regarding the choice of revascularization strategy in individual patients with left main coronary artery disease? How can such nuanced considerations be communicated effectively to enable shared decision-making?”

Death, Stroke, and MI

In all, 361 of 1,869 randomized patients (19.3%) with available data on baseline renal function had CKD, defined as an estimated glomerular filtration rate of < 60 mL/min/1.73 m2. Acute renal failure, defined as an increase in serum creatinine ≥ 5 mg/dL or a new requirement for dialysis, was more frequent in the first 30 days for patients with versus without CKD (5.0% vs 0.8%; P < 0.0001) and nearly quintupled the likelihood of experiencing the primary endpoint of death, stroke, or MI by 3 years (HR 4.59; 95% CI 2.73-7.73). At 3-year follow-up, the CKD group as a whole had a higher rate of these major adverse events compared with the no-CKD group (20.8% vs 13.5%; P = 0.0005).

Looking at results based on revascularization type, acute renal failure was less common with PCI than with CABG. This was true whether patients had CKD (2.3% vs 7.7%) or better baseline renal function (0.3% vs 1.3%; P for interaction = 0.71). Patients in the PCI group also had less TIMI major or minor bleeding and less need for blood transfusion than those in the CABG group.

At 30-day follow-up, the combined rate of death, stroke, and MI was lower with PCI than with CABG in patients with CKD (6.2% vs. 9.3%) and those without CKD (4.5% vs 7.4%; P for interaction = 0.80). By 3 years, however, PCI and CABG had resulted in similar adverse event rates both for patients with (23.4% vs 18.1%) and without CKD (13.4% vs 13.5%; P for interaction = 0.38). Of note, the mortality rate was higher with PCI; all deaths occurred beyond 30 days and the difference was driven by noncardiac death related to sepsis.

No Clear Winner

Acute renal failure as was defined in the study represents a very serious adverse event, Giustino pointed out. "These are massive renal injuries compared to mild rises in creatinine that you commonly see after PCI or CABG in general," he stressed.

"The mechanism by which PCI is associated with lower risk of these important endpoints is likely due to the fact that putting a patient through surgery—including cardiopulmonary bypass, potential bleeding complications during surgery, need for transfusion, or increased oxidative stress, all these constellations of adverse events—can predispose patients to develop major adverse renal events," Giustino explained. On the flip side, PCI-related factors such as contrast-induced nephropathy, hemodynamic instability, or an atheroembolism are less common and would likely carry less risk for the kidneys than cardiac surgery, he added.

O’Gara, in his editorial, points out that while PCI initially held the lead, CABG had lower rates of ischemia-driven revascularization and noncardiac death over time. Only longer-term follow-up will provide clarity about the trajectory of clinical outcomes, he says, noting that EXCEL is slated to go through 5 years.

The factors informing today’s decision-making process are, as O’Gara notes, “nuanced.”

“These are complicated considerations that would be difficult for most patients to grasp, especially in urgent clinical situations. Attention to patient preferences and values is critical, but clinical experience would suggest that the majority would opt to avoid the early hazards associated with CABG and take their chances with the late sequelae related to PCI,” he writes. “It is not clear whether there are enough patients or events in this CKD subgroup to allow construction of multivariable risk models using patient-level data that incorporate quality of life and other functional outcomes to predict benefit and harm so that individual decision-making could be made more personal and more precise.”

While the current results do indeed add to the evidence base, O’Gara agrees, “in the absence of a clear winner, they also accentuate the need for longer-term follow-up in trials of coronary revascularization and argue strongly for the promotion of efforts to utilize validated risk models to help guide difficult decisions.”

For now, Giustino said, decisions should come down to patient preference and should involve discussions between the heart team, the patient, and the patient's family about the relative benefits and risks of the two revascularization strategies.

Note: Several co-authors of the paper are faculty members or employees of the Cardiovascular Research Foundation, the publisher of TCTMD.