COMPARATIVE EFFECTS OF GLIBENCLAMIDE AND METFORMIN ON AMBULATORY BLOOD-PRESSURE AND CARDIOVASCULAR REACTIVITY IN NIDDM

OBJECTIVE - To compare the effects of chronic glibenclamide and metfor min therapy on blood pressure (BP) and cardiovascular responsiveness i n patients with NIDDM. RESEARCH DESIGN AND METHODS - Fourteen patients with NIDDM received metformin or glibenclamide for 1 month in a doubl e-blind, randomized crossover study At the end of each treatment perio d, patients were tested for forearm vascular responsiveness to intrabr achial arterial infusion of diazoxide (an ATP-sensitive potassium chan nel opener), acetylcholine, sodium nitroprusside, and norepinephrine, BP responses to intravenous infusions of NE and angiotensin II, BP res ponses to cold presser testing and isometric exercise, and 24-h ambula tory BP monitoring. RESULTS - Metformin and glibenclamide produced sim ilar glycemic control. Mean 24-h BPs did not differ between the two gr oups, but mean 24-h heart rates were significantly lower (75 +/- 6 bpm vs. 80 +/- 6 bpm) on glibenclamide therapy than on metformin. Plasma norepinephrine levels were significantly higher on glibenclamide (6.41 +/- 1.77 vs. 4.26 +/- 1.54 mmol/l, P < 0.01), and systolic BP respons es to intravenous norepinephrine and angiotensin II were significantly higher on glibenclamide than on metformin (P < 0.02 and P < 0.05, res pectively). Systolic BP responses to cold presser testing appeared hig her on glibenclamide than on metformin, but the difference did not qui te achieve statistical significance (P = 0.052). Baseline forearm vasc ular resistance did not differ between the two drugs, nor did forearm vascular resistance responses to diazoxide, acetylcholine, sodium nitr oprusside, and norepinephrine differ. CONCLUSIONS - Glibenclamide ther apy is accompanied by greater systolic BP responses to norepinephrine and angiotensin II and higher plasma norepinephrine levels than those that occur on metformin therapy. Lower heart rates on glibenclamide th erapy despite evidence of greater sympathetic activity suggests that g libenclamide may have negative chronotropic effects.