Coronary artery bypass surgery improves outcomes in patients with diabetes and left ventricular dysfunction

Selected in Journal of the American College of Cardiology by G. Di Gioia



Jeevan Nagendran, Sabin J. Bozso, Colleen M. Norris, Finlay A. McAlister, Jehangir J. Appoo, Michael C. Moon, Darren H. Freed and Jayan Nagendran




February 2018


Read the abstract

My Comment

Why this study – the rationale/objective?

While various studies addressed the issue of the best revascularization strategy in patients with diabetes mellitus and multivessel coronary artery disease, little is known about the impact of left ventricular dysfunction on the outcome of diabetic patients treated either with coronary artery bypass grafting (CABG) or percutaneous coronary intervention (PCI).

How was it executed – the methodology?

This was a propensity matched study based on the prospective APPROACH database (Alberta Provincial Project for Outcome Assessment in Coronary Heart Disease). From 2004 to 2016, 869 diabetic patients with coronary multivessel disease and left ventricular dysfunction treated with PCI were matched, in a 1 to 1 fashion, with a corresponding number of patients treated with CABG. The primary outcome of this study was the combined endpoint of major adverse cardiac and cerebrovascular events (MACCE), defined as the composite of death, stroke, MI, and repeat revascularization.

What is the main result?

At 5 years follow up, MACCE and mortality were significantly more frequent in the PCI group rather than in the CABG group, both in patients with EF between 35 and 49% (respectively 51% vs. 28%; p<0.01 and 26% vs. 16%; p<0.01) and in patients with EF < 35% (respectively 61% vs. 29%; p<0.01 and 35% vs. 19%; p<0.01). Repeat revascularization rates were higher in the PCI groups in both patients with EF < 35% and >35%EF. Myocardial infarction also occurred more frequently in the PCI group, though only in patients with EF<35%. No difference in stroke was found between the 2 groups in either of the 2 EF layers.

Critical reading and the relevance for clinical practice

In this propensity matched study including a large number of patients, CABG is seemingly consecrated as the revascularization therapy of choice in patients with diabetes mellitus and multi-vessel coronary artery disease. Actually, in patients with left ventricular dysfunction the difference in terms of survival and survival free from adverse events in favour of CABG is apparently even larger than the one reported in previous studies investigating mainly patients with preserved ejection fraction. On the other hand, this study has several limitations:

  • First, it is not randomized. Propensity score matching, although methodologically rigorous, will never be able to ensure the absence of selection bias, for the factors guiding decision-making in this very high-risk subset of patients are often intangible and thus unmeasurable.
  • Second, there is a lack of procedural information (i.e. on-pump time, number of grafts and anastomosis, number and type of stents, total stent length...), that would be useful to have a comprehensive picture of the situation.
  • Third, a considerable amount of patients were not enrolled in this study because treated conservatively.

Notwithstanding, this study has the merit of being a very well conducted non-randomized trial, with reliable outcome data based on a regional prospective registry, whose results will likely stimulate researchers to conduct a highly needed randomized trial in this patient population.

No comments yet!