Drug Information Center

What are the findings of the BARI 2D Study?

The classification and treatment of ischemic heart disease (IHD) is guided by patient presentation and objective diagnostic tests.¹ Medical therapy with antiplatelet agents (aspirin), beta blockers and/or calcium channel blockers, angiotensin converting enzyme inhibitors, and lipid lowering agents (statins) are recommended for treatment of chronic, stable ischemic heart disease. Revascularization therapies, such as coronary artery bypass grafting (CABG) and percutaneous coronary intervention (PCI), are generally reserved for patients with disease of multiple vessels and/or left main coronary artery disease. Patients with type 2 diabetes are considered to be at an elevated risk for IHD compared to patients without diabetes. Class IIa (level evidence B) recommendations from the American Heart Association and the American College of Cardiology suggest that CABG for patients with diabetes who present with unstable angina and multivessel disease and PCI for patients with single vessel disease and inducible ischemia can be effective.² The guidelines also recognize the need for further studies of this issue with focused objectives. Furthermore, the benefits of revascularization in patients with diabetes who have mild symptoms of ischemia are still unclear.

The initial Bypass Angioplasty Revascularization Investigation (BARI) study demonstrated a higher survival rate in symptomatic patients with multivessel coronary artery disease randomized to CABG compared to PCI in the subgroup of patients with diabetes.³ However, only 19% of the patients in the BARI study had diabetes. The BARI 2 Diabetes (BARI 2D) trial, therefore, was designed to evaluate the effects of treatments for IHD only in patients with diabetes.

Design/Methods
BARI 2D compared the effects of treatments for IHD and diabetes on rates of death and cardiovascular events.⁴ Patients were required to have a diagnosis of type 2 diabetes (defined as the need for treatment or elevated blood glucose) and a diagnosis of coronary artery disease (defined as ≥ 50% stenosis of a major epicardial artery along with a positive stress test or ≥ 70% stenosis of a major epicardial artery and symptoms of angina). Patients with a need for immediate revascularization, PCI or CABG within the previous 12 months, left main coronary disease, creatinine level > 2 mg/dL, glycated hemoglobin >13%, class III or IV heart failure, or hepatic dysfunction were excluded from the trial.

Patients were stratified according to the type of revascularization procedure (CABG or PCI) as deemed appropriate by the physician.⁴ With the dual diagnosis of diabetes and IHD, patients were first randomized to prompt revascularization or medical therapy for management of ischemic heart disease and then, were further randomized to diabetes treatment with either insulin-providing treatment (such as sulfonylureas and/or insulin) or insulin-sensitizing agents (such as metformin and/or thiazolidinediones). As a result, patients were categorized into 8 treatment groups (Figure 1). Patients randomized to revascularization were required to have the procedure within 4 weeks of randomization. Patients initially randomized to medical therapy could receive delayed revascularization and those randomized to a particular diabetes treatment group could have other medications added to their regimen if deemed medically necessary. Medical management of heart disease and diabetes followed current guidelines and goals of therapy including a glycated hemoglobin <7%, blood pressure (BP) ≤130/80 mm Hg and a low density lipoprotein (LDL) cholesterol level <100 mg/dL. Death from any cause was the primary outcome and the composite of death, myocardial infarction (MI) or stroke (major cardiovascular events) was the secondary endpoint.

Results
Of the 2368 patients who were randomized, 70% were males, 66% were white, 32% had a history of MI, and 24% had prior revascularization.⁴ The mean duration of diabetes was 10.4 years and the mean baseline glycated hemoglobin was 7.7%. Baseline characteristics were similar across treatment groups with the exception of unstable angina, which was more common in the prompt revascularization group. At baseline, patients selected for CABG had more triple vessel disease and proximal left anterior descending (LAD) artery disease than patients selected for PCI. The number of patients in each treatment group is outlined in Figure 1.

Approximately 42% of patients who were randomized to medical therapy underwent delayed revascularization by study end.⁴ More patients in the insulin sensitization group (43.4%) required addition of insulin provision therapies compared to patients in the insulin provision group (11.8%) that required addition of insulin sensitization therapies. Over 80% of patients in the treatment groups achieved the LDL target and approximately 70 % of patients achieved the BP target. The mean glycated hemoglobin was significantly lower in the insulin sensitization group compared to the insulin provision group (7.0% vs. 7.5%, respectively; p=<0.001). Also, patients in the insulin sensitization group had statistically significantly higher high density lipoprotein (HDL) levels (42 mg/dl vs. 40 mg/dl) and lower body mass index (BMI) values (31.7 vs. 32.5) compared to patients in the insulin provision group.

The 5 year survival rate of patients randomized to prompt revascularization was 88.3% compared to 87.8% of patients who received medical therapy for their IHD (difference 0.5%; 95% confidence interval [CI] -2.0 to 3.1; p=0.97).⁴ Patients who received insulin sensitization demonstrated a 5 year survival rate of 88.2% compared to 87.9% of patients who received insulin-providing treatments (difference 0.3%; 95% CI -2.2 to 2.9; p=0.89). There was also no difference in the secondary outcome of freedom from major cardiovascular events between these groups.

Comparing IHD treatments within each stratum demonstrated no significant difference in survival rate. Within the PCI stratum, the 5 year survival rate for patients who received medical therapy was 89.8% compared to 89.2% in the group randomized to revascularization (p=0.48). For patients selected for the CABG stratum, the survival rates were 83.6% and 86.4 % for patients randomized to medical therapy and revascularization, respectively (p=0.33). However, significantly more patients remained free from major cardiovascular events (secondary endpoint) within the CABG stratum who underwent revascularization compared to those randomized to medical therapy (77.6% vs. 69.5%, respectively; p=0.01). The frequency of nonfatal MI was also reduced for patients within the CABG stratum who received revascularization compared to those who received medical therapy (7.4% vs. 14.6%). On the other hand, those patients within the PCI stratum assigned to revascularization or to medical therapy did not demonstrate a significant difference in freedom from major cardiovascular events (77% vs. 78.9%, respectively; p=0.15).

Comparison of the 4 treatment combinations of medical therapy plus insulin sensitization, medical therapy plus insulin provision, revascularization plus insulin sensitization, and revascularization plus insulin provision revealed no significant differences in the primary or secondary outcomes.⁴ However, when analyzed by stratum, the only significant finding was a reduction in major cardiovascular events for patients within the CABG stratum who received revascularization plus insulin sensitization compared to those who received medical therapy plus insulin sensitization (18.7% vs. 32%, respectively; p=0.0002).

Only 2 adverse events demonstrated significant differences among treatment groups.⁴ Severe hypoglycemia, defined as hypoglycemia requiring assistance and a blood glucose level of <50 mg/dL, occurred significantly more frequently in patients on insulin provision compared to those on insulin sensitizers (9.2% vs. 5.9%, respectively; p=0.003). On the other hand, peripheral edema was more common in patients on insulin sensitizers than those randomized to insulin provision (p=0.02). The incidence of heart failure was not significantly different between these treatment groups.

Conclusion
Overall survival and major cardiovascular events were not affected by the type of diabetes treatment. Due to the findings of less hypoglycemia, higher HDL levels, and less weight gain, the authors suggest that insulin sensitization may be a preferable treatment option in patients with type 2 diabetes and IHD.

Mortality, the primary outcome, was not found to be significantly different between treatments for ischemic heart disease. Patients selected for the CABG stratum that underwent the revascularization procedure fared better in terms of fewer major cardiovascular events than those who received medical therapy. According to the authors, patients with diabetes who have extensive vessel disease should undergo prompt revascularization whereas patients with diabetes with less extensive coronary artery disease who are considered candidates for PCI may not need revascularization as an initial treatment option.

The findings of this trial provide support to current recommendations that suggest CABG for patients with diabetes and multivessel disease can be effective. However, the lack of difference in mortality and occurrence of cardiovascular events between PCI and medical therapy in patients with diabetes and IHD is contrary to the current recommendations. Currently, clinicians should rely on current recommendations and clinical judgment for management of IHD in patients with diabetes until further evidence can establish the benefits of PCI and medical therapy in this patient population.