For CKD Patients, CABG Favored Long-term Over PCI

In patients with chronic kidney disease (CKD), CABG is associated with improved early and late clinical outcomes compared with PCI, according to a registry study published in the January 2015 issue of Circulation: Cardiovascular Interventions. Mortality at 30 days was similar between the treatment groups, but survival at 3 years favored surgery.Take Home: For CKD Patients, CABG Favored Long-term Over PCI

“These data would suggest that patients with CKD would be better served with CABG over PCI,” the authors say, although they acknowledge that there is a paucity of data and to date no randomized controlled trial has compared the strategies in this high-risk group.

Vladimír Džavík, MD, of Toronto General Hospital (Toronto, Canada), and colleagues identified 4,006 patients (average age 75 years) with CKD and multivessel disease who underwent revascularization from October 2008 to September 2011 and were enrolled in the Cardiac Care Network of Ontario Cardiac Registry. Most patients had CABG (n = 3,010), while the rest received PCI with DES (n = 996).

Patients from each treatment arm were also sorted into 893 propensity-matched pairs. Baseline clinical and procedural characteristics were similar between the groups with the exception of CABG patients having lower weight and being more likely to be in Canadian Cardiovascular Society angina class IV. Diabetes was present in about 44% of each group.

Higher Long-term Survival With CABG

Mortality and admission for stroke were similar between the PCI and CABG groups at 30 days, but MACCE and its components—including repeat revascularization and MI—were higher with PCI. Late follow-up at 3 years showed higher mortality, repeat revascularization, MI, and MACCE with PCI vs CABG (P < .01 for all; table 1).

Table 1. Propensity-Matched Analysis of Clinical Outcomes

Kaplan-Meier survival estimates favored CABG over PCI at 1, 2, and 3 years (P < .001 for all). Additionally, CABG resulted in greater freedom from MACCE at each of those time points, driven by a greater occurrence of MI and repeat revascularization in the PCI cohort (P < .001 for all).

In the unmatched analysis, PCI with DES was associated with increased late mortality (HR 1.58; 95% CI 1.32-1.90) and MACCE (HR 2.62; 95% CI 2.28-3.01). Other independent predictors of these outcomes included CKD stage, age, diabetes, LV dysfunction, and urgent revascularization priority. DES use resulted in similar late MACCE outcomes across a multitude of subgroups compared with CABG (P < .001 for all).

Of the more than 80% of patients who had stage 3 CKD, mortality rates at 30 days and at 1 year were similar between PCI and CABG. However, late mortality and MACCE were higher in the PCI group (P < .05 for both). In those with stage 4 CKD, all adverse outcomes occurred more frequently in the PCI arm with the exception of 30-day death.

Several Advantages to CABG

“These important findings underscore the results of 2 recent large RCTs comparing the effectiveness of PCI with DES versus CABG in patients with complex multivessel disease and diabetes mellitus (SYNTAX and FREEDOM),” Dr. Džavík and colleagues write. “In both these trials, CABG demonstrated superiority over PCI with respect to MACCE, MI, and repeat revascularization, particularly in the subset of patients with intermediate to high SYNTAX scores (with mortality reduction at 5 years in FREEDOM).”

They offer possible reasons why surgery seems to be favored in this challenging patient group:

  • Complete revascularization may provide a survival advantage, and it is more likely to be achieved with CABG than with PCI
  • In CKD patients, procedural success is less common with PCI, contributing to the risk of incomplete revascularization
  • Surgery may help protect against future coronary events by bypassing vulnerable plaques

In an accompanying editorial, Usman Baber, MD, MS, and Roxana Mehran, MD, both of the Icahn School of Medicine at Mount Sinai (New York, NY), confirm the consistency of these data with past studies in patients with diffuse or complex CAD as well as in those receiving dialysis. However, this study is at odds with “a recent network meta-analysis suggesting that newer-generation DESs yield mortality reductions comparable with CABG when either is compared with medical therapy, whereas first-generation DESs confer less benefit.”

The editorialists note an outstanding baseline difference even after propensity matching—adherence to dual antiplatelet therapy—that “as a potential mediator of DES-attributable mortality in the setting of CKD, may be particularly relevant given the increased bleeding risk observed in such patients.”

Drs. Baber and Mehran also express surprise at the lack of difference in stroke risk between the treatment groups. “Although this rate is comparable with the CABG arms in both FREEDOM and SYNTAX trials, it is much higher than the [approximate] 1.5% stroke rate observed in the PCI arms of these randomized studies,” they write, adding that this may be due to chance or a selection bias favoring PCI.

Patient Choice Always a Factor

In a telephone interview with TCTMD, Jeffrey A. Breall, MD, of Indiana University (Indianapolis, IN), said he too was “a little bit surprised” about the stroke rates. “To my knowledge, [this is] the first time that stroke ever [did not favor PCI],” he said.

A strength of the study is that it represents real-world practice, according to Richard Solomon, MD, of the University of Vermont (Burlington, VT), who noted that 3 times as many patients underwent surgery as received PCI.

He told TCTMD in a telephone interview that patient factors often influence treatment choice since “patients tend to look at the near future rather than long-term” and PCI is less invasive. “For people with lifespans that are already shortened by comorbidities, PCI may be a reasonable choice,” he said.

The study authors observe that the best way to approach decision making is “in the framework of a Heart Team to define the optimal revascularization strategy for each individual patient.”

Dr. Breall said a randomized trial in the CKD population—much like the FREEDOM trial in diabetics—would be “tremendously valuable” since the majority of PCI and CABG studies previously excluded patients with CKD. Given that “we see patients like this all the time,” he noted, enrollment would be easy.

Dr. Solomon agreed but said he doubted that the outcomes would depart from what is seen in the current study. Another question is how the choice of revascularization affects the need for dialysis or worsening kidney function, he suggested.


1. Chan W, Ivanov J, Ko D, et al. Clinical outcomes of treatment by percutaneous coronary intervention versus coronary artery bypass graft surgery in patients with chronic kidney disease undergoing index revascularization in Ontario. Circ Cardiovasc Interv. 2014;Epub ahead of print.

2. Baber U, Mehran R. Coronary artery revascularization in chronic kidney disease: time for a randomized trial [editorial]. Circ Cardiovasc Interv. 2014;Epub ahead of print.

Related Stories:

  • Dr. Džavík reports receiving support from the Brompton Funds Professorship in Interventional Cardiology at the Peter Munk Cardiac Centre and research and educational grants and speaker’s honoraria from Abbott Vascular and educational grants from St. Jude Medical.
  • Dr. Mehran reports receiving institutional research grant support from, serving as a consultant to, and holding equity in multiple pharmaceutical and device companies.
  • Dr. Breall reports serving on the advisory board for Siemens Interventional Cardiology and as a consultant to Fujifilm.
  • Drs. Baber and Solomon report no relevant conflicts of interest.