Risk Prediction Tools Improve Shared Decision Making in Diabetes With CAD

doctor speaking to patient
Researchers created a personalized risk prediction tools that can estimate long-term MACE and angina outcomes after revascularization in patients with diabetes and multivessel disease.

Personalized risk prediction tools may be used to improve shared medical decision making and increase the use of coronary artery bypass grafting (CABG) over percutaneous coronary intervention (PCI) in patients with diabetes and multivessel coronary artery disease (CAD), according to study results published in the Journal of the American College of Cardiology.

Results from the FREEDOM trial (Future Revascularization Evaluation in Patients with Diabetes Mellitus: Optimal Management of Multivessel Disease; ClinicalTrials.gov identifier: NCT00086450) have shown CABG to be superior to PCI for major acute cardiovascular events (MACE) and angina reduction in patients with diabetes and CAD. However, PCI continues to be preferred as a revascularization strategy in many settings for reasons, including patients’ preference for a less invasive treatment and lack of shared medical decision making. The objective of this study was to establish a more evidence-based infrastructure to help physicians better weigh the risks and benefits of CABG vs PCI, and to develop prediction models for long-term MACE and angina in patients with diabetes and multivessel CAD.

In this study, researchers evaluated data from 1900 participants in the FREEDOM trial, which was an international, multicenter, randomized clinical trial that took place between 2005 and 2010 and compared CABG with PCI outcomes in patients with diabetes and multivessel CAD. The primary outcome of the FREEDOM trial was all-cause mortality, nonfatal myocardial infarction, and nonfatal stroke. Researchers used the patient-level data to develop models to predict 5-year MACE and 1-year angina following CABG and PCI using baseline characteristics and treatment interactions. Parsimonious models were created and internally validated using bootstrap resampling with 500 replications, and the MACE model was externally validated in a real-world registry.

Results revealed that 18.2% of patients (n=346) experienced MACE within 5 years after randomization (CABG vs PCI: 15.4% vs 20.9%) and that 16.3% of patients (n=310) reported angina at 1 year (CABG vs PCI: 14.1% vs 18.4%). The MACE model included 8 covariates and a treatment interaction with a history of smoking (P =.04). Patients with a history of smoking were shown to be at lower risk for MACE with CABG than with PCI. 

External validation in stable CAD and acute coronary syndrome demonstrated comparable performance (c=0.65 and c=0.68, respectively). The angina model included 6 variables and a treatment interaction with SYNTAX score (P =.02), and showed that patients with intermediate or high SYNTAX scores experienced less angina with CABG than with PCI.

PCI was not found to be superior to CABG, and 54.5% of patients were expected to have a lower risk of MACE with CABG. All patients with a history of smoking were expected to have a lower risk of MACE with CABG than with PCI. Lastly, all patients with a SYNTAX score greater than 22 were expected to experience angina relief with CABG vs PCI.

Related Articles

Limitations to this study were as follows. First, the models used may not work as well in more general populations, since this was a post-hoc analysis from a randomized clinical trial. Second, researchers modeled the MACE and angina prediction tools at 5 and 1 years, respectively, meaning it’s possible CABG may offer higher or lower benefit to symptom control vs PCI over a longer period. Third, the FREEDOM trial involved the use of first-generation stents, meaning newer stents may provide better outcomes. Lastly, clinicians may continue to face the challenge of being able to apply trial results to patients ineligible for randomization.

The study researchers concluded that “CABG was the preferred strategy in the majority of patients, especially among those with a history of smoking. Although PCI was never the preferred revascularization strategy, 45% of patients were expected to have similar risks of poor outcomes with CABG or PCI.” The new MACE and angina prediction tools may be used in clinical settings to improve the shared medical decision making process for choosing CABG vs PCI in patients with diabetes and multivessel CAD who need MACE and angina treatment.

Disclosure: Several study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


Qintar M, Humphries KH, Park JE, et al. Individualizing revascularization strategy for diabetic patients with multivessel coronary disease. J Am Coll Cardiol. 2019;74(16):2074-2084.