GEMINI: Carvedilol Safe in Patients with Diabetes and Hypertension

November 9, 2004 (eshonline.org) - Carvedilol had a neutral effect on glycemic control and improved components of the metabolic syndrome, including blood pressure, in the presence of renin-angiotensin system blockade. The Glycemic Effects in Diabetes Mellitus: Carvedilol-Metoprolol Comparison in Hypertensives study was reported today at the 2004 Scientific Sessions of the American Heart Association and simultaneously published by JAMA (2004;292:2227-2236).

The double-blind, placebo-controlled GEMINI trial randomized 1235 patients to carvedilol or metoprolol (average dose 17.5 mg twice daily carvedilol and 128 mg metoprolol tartrate) for 6 months follow-up. The target blood pressure for patients with a baseline systolic blood pressure (SBP) between 140 and 179 mm Hg was < 135 mm Hg and for a baseline SBP of 130-140 it was < 130, and for a baseline diastolic blood pressure (DBP) > 90 mm Hg the target was < 85 and for a baseline DBP between 80 and 89 it was = 80 mm Hg. A washout of all antihypertensive agents except for ACE inhibitors and angiotensin receptor blockers was performed before randomization. To achieve target blood pressures, 42% of patients required add-on diuretic therapy and 20% required a calcium channel blocker. Target blood pressures were achieved in 68% of the carvedilol group and 67% of the metoprolol group, and achieved levels were similar between groups. The patients were between 30 and 80 years old (mean age 61) and 40% of the carvedilol group and 48% of the metoprolol group were women.

The primary endpoint of the mean change in hemoglobin A1c (HbA1c) from baseline was 0.13%±0.05 (p=0.04). An increase in HbA1c > 1% occurred in 14.2% of the metoprolol group and 7% of the carvedilol group (odds ratio 0.46; p<0.001). An increase in HbA1c is associated with a clear and significant increase in cardiovascular risk.

HOMA-IR improved with carvedilol (-9.1%; p=0.004) but not with metoprolol (-2.0%; p=0.48). The treatment difference was -7.2% (p=0.04). Withdrawals for worsening glycemic control was higher in the metoprolol than the carvedilol group (2.2% vs 0.6%, respectively; p=0.04). New onset microalbuminuria, a cardiovascular risk marker which is also associated with inflammatory and vascular effects, was less in the carvedilol group than in the metoprolol group (6.4% vs 10.3%, odds ratio 0.36; p=0.04).

In discussing the clinical implications, George Bakris, who presented the trial at the AHA on behalf of the GEMINI investigators, stated that physicians should consider carvedilol for patients who are at risk of diabetes and switching to carvedilol for patients with diabetes and hypertension already taking a beta blocker. The results could also be applicable to patients who should be on a beta blocker for other cardiac reasons without concern about glycemic control and other components of the metabolic syndrome. The alpha blocking and particularly the antioxidant properties of carvedilol, a newer beta blocker, are likely responsible for its impact on the metabolic syndrome components. GEMINI was the first large-scale trial to evaluate the addition of beta blockers in this study population.