CABG Appears Superior to PCI for Patients With Type 1 Diabetes, Multivessel Disease

The observational findings support existing recommendations favoring surgical revascularization in patients with diabetes.

CABG Appears Superior to PCI for Patients With Type 1 Diabetes, Multivessel Disease

BARCELONA, Spain—Over the long term, patients with type 1 diabetes and multivessel coronary disease fare better with CABG than with PCI, a population-based cohort study suggests.

Those findings, reported by Martin Holzmann, MD, PhD (Karolinska University Hospital, Stockholm, Sweden), at the European Society of Cardiology Congress 2017 here and published simultaneously online in the Journal of the American College of Cardiology, are consistent with both US and European guidelines, which recommend surgical revascularization over PCI in patients with diabetes.

“Our findings indicate that for patients with type 1 diabetes, CABG instead of PCI should be the preferred strategy for multivessel revascularization,” Holzmann said during his presentation.

Even so, he and his colleagues say in the paper that their “findings should be interpreted with some caution because of the observational nature of the study, and maybe more importantly, the large differences in risks in the first year of follow-up, indicating that there were large inherent differences in risk at baseline between the PCI and CABG groups.”

CABG has consistently been shown to improve outcomes over PCI in patients with diabetes—in the BARI and FREEDOM trials and in a subgroup analysis of the SYNTAX trial, for instance. But Holzmann noted that none of those studies has performed subgroup analyses based on the type of diabetes; this could be important when considering the findings because patients with type 1 diabetes are more likely to die after CABG than those with type 2 diabetes, whose prognosis is similar to that of nondiabetics after surgical revascularization. Also, because about 90% of diabetic patients have type 2 disease, the results of the prior trials might not be generalizable to those with type 1 diabetes, Holzmann said.

The investigators focused on patients with type 1 diabetes in the current study, which used data from the SWEDEHEART registry and other Swedish national registries. The analysis included 2,546 patients with multivessel disease who underwent revascularization (73% with PCI) between 1995 and 2013.

Although CABG was used in the majority of patients in the beginning of the study period, PCI steadily took over, and from 2010 to 2013, only 2% of patients underwent CABG. “The reason why guideline recommendations were not followed, and PCI was chosen over CABG to such a large extent in our observational study, remains elusive,” the authors say.

The investigators used inverse probability of treatment weighting based on propensity score to account for differences between the CABG and PCI groups. After adjustment, PCI carried higher risks of three of six outcomes, with no differences between groups for the others, through a mean follow-up of 10.6 years.

Long-term Outcomes: PCI vs CABG



(n = 1,863)


(n = 683)

Adjusted HR

(95% CI)




1.14 (0.99-1.32)




1.47 (1.23-1.78)

Heart Failure



1.10 (0.91-1.32)




1.00 (0.76-1.31)

Repeat Revascularization



5.64 (4.67-6.82)

Coronary Heart Disease Death



1.45 (1.21-1.74)


The main limitation of the study, the authors point out, is the large difference in outcomes between the CABG and PCI groups during the first year of follow-up, indicative of confounding by indication. “Probably there were patients in the PCI group who were denied CABG because of frailty or things that we were not able to actually adjust for in our analysis,” Holzmann said.

One of the moderators of the session where the results were presented, as well as an audience member, questioned the strength of Holzmann’s conclusion considering the limitations of the observational analysis.

Holzmann acknowledged that it is difficult to make causal inferences based on observational data, but noted that the data from this study are similar to those from the randomized FREEDOM study, which also showed an advantage for CABG over PCI.

“I believe the conclusion should be maybe that the results from the FREEDOM study may be generalizable to type 1 diabetes patients as well,” he said.

Findings ‘Should Influence Decision-Making’

In an accompanying editorial, Michael Domanski, MD, and Michael Farkouh, MD (University of Toronto, Canada), say the study “offers data specifically supporting CABG as the appropriate revascularization modality in patients with type 1 diabetes.”

In discussing why CABG has been shown to be better than percutaneous revascularization in diabetic patients, they note that the mechanisms of benefit differ between the two approaches. They explain that “CABG removes large segments of the artery that would have added to the total risk of necrosis by an occlusion,” whereas PCI does not reduce the probability of occlusion in nonstented regions of the vessel.

“Better stents alone cannot change the superiority of CABG compared with PCI for patients with aggressive CAD (diabetes or high SYNTAX score), because PCI addresses only a small portion of the coronary anatomy,” Domanski and Farkouh write. “This does not diminish the importance of continuing advances in stent technology, but rather, it puts into appropriate perspective what can be expected from these advances.”

The editorialists highlight the importance of the study because it provides information in an “essentially data-free zone,” but they also cite limitations, including the observational design and—as pointed out by the investigators—the likelihood that the PCI group included some patients who were not surgical candidates.

Nonetheless, Domanski and Farkouh say, “The findings of this important study help to better inform practice, and should influence decision-making for revascularization in patients with type 1 diabetes.”

ESC 2017


Todd Neale is the Associate News Editor for TCTMD and a Senior Medical Journalist. He got his start in journalism at …

Read Full Bio
  • Holzmann reports holding a research position funded by the Swedish Heart-Lung Foundation and has receiving consultancy honoraria from Actelion and Pfizer.
  • Domanski and Farkouh report no relevant conflicts of interest.