BARI 2D: A Reanalysis Focusing on Cardiovascular Events

Abstract

Objective: To reanalyze the Bypass Angioplasty Revascularization Investigation 2 Diabetes trial using a new composite cardiovascular disease (CVD) outcome to determine how best to treat patients with type 2 diabetes mellitus and stable coronary artery disease.

Patients and methods: From January 1, 2001, to November 30, 2008, 2368 patients with type 2 diabetes mellitus and angiographically proven coronary artery disease were randomly assigned to insulin-sensitizing (IS) or insulin-providing (IP) therapy and simultaneously to coronary revascularization (REV) or no or delayed REV (intensive medical therapy [MED]), with all patients receiving intensive medical treatment. The outcome of this analysis was a composite of 8 CVD events.

Results: Four-year Kaplan-Meier rates for the composite CVD outcome were 35.8% (95% CI, 33.1%-38.5%) with IS therapy and 41.6% (95% CI, 38.7%-44.5%) with IP therapy (P=.004). Much of this difference was associated with lower in-trial levels of fibrinogen, C-reactive protein, and hemoglobin A_1c with IS therapy. Four-year composite CVD rates were 32.7% (95% CI, 30.0%-35.4%) with REV and 44.7% (95% CI, 41.8%-47.6%) with MED (P<.001). A beneficial effect of IS vs IP therapy was present with REV (27.7%; 95% CI, 24.0%-31.4% vs 37.5%; 95% CI, 33.6%-41.4%; P<.001), but not with MED (43.6%; 95% CI, 39.5%-47.7% vs 45.7%; 95% CI, 41.6%-49.8%; P=.37) (homogeneity, P=.05). This interaction between IS therapy and REV was limited to participants preselected for coronary artery bypass grafting (CABG). The lowest composite CVD rates occurred in patients preselected for CABG and assigned to IS therapy and REV (17.3%; 95% CI, 11.8%-22.8%).

Conclusion: In the Bypass Angioplasty Revascularization Investigation 2 Diabetes trial, the IS treatment strategy and the REV treatment strategy each reduces cardiovascular events. The combination of IS drugs and CABG results in the lowest risk of subsequent CVD events.

Trial registration: clinicaltrials.gov Identifier: NCT00006305.

Figure 1:

Modified from figures 1 and 3 in reference . HbA1c and Plasma Insulin were measured at 6, 12, 24, 36 and 48 months. CRP and Fibrinogen were measured at 12 and 24 months. All IS levels after baseline were significantly lower than IP levels after baseline, P<.001.

Figure 2:

Kaplan-Meier cumulative incidence curves of the composite CVD outcome comparing IS versus IP treatment (panel A) and REV versus MED treatment (Panel B) over 4 years.

Figure 3:

Kaplan-Meier cumulative incidence curves of the composite CVD outcome for each randomized cohort of glycemic management (IS or IP) coupled with CAD management (REV or MED) in the upper panel. The same curves are displayed separately for those in the CABG stratum (left lower panel) and those in the PCI stratum (right lower panel).