This study was performed to investigate possible effects of glibenclamide, insulin and metformin on the death latency and incidence caused by a cardiac glycoside, oubain. Mice of both sexes were injected with oubain (i.p. 20 mg kg(-1)), glibenclamide (s.c. 0.1-10 mg kg(-1)), insulin (s.c. 0.3-3 U kg(-1)) and metformin (i.p. 200 mg kg(-1)) and combinations of the last three drugs with oubain. Death latency was measured and lethality incidence was calculated. Death was assessed by visual observation. Plasma glucose level was evaluated from the tail blood. Glibenclamide (0.1 mg kg(-1)) prolonged the latency from 11.3 +/- 1.2 to 15.8 +/- 1.8 min but failed to decrease the incidence of death. At higher doses (1-10 mg kg(-1)) it had no effects on the latency or the incidence. 0.3 U kg(-1) insulin decreased the incidence from 73.7 to 33.3% (P < 0.05) without affecting the latency. However the higher dose (3 U kg(-1)) did not have any effects on the incidence or the latency. Oubain increased blood glucose level from 114.1 +/- 3.8 (control) to 152.1 +/- 5.3 mg dl(-1). Metformin (200 mg kg(-1)) did not affect either the latency or the incidence of death. While metformin did not decrease plasma glucose, insulin and higher doses of glibenclamide (1-10 mg kg(-1)) markedly lowered glucose in blood. However, at the dose of 0.1 mg kg(-1) glibenclamide did not alter the glucose level in the blood but prevented oubain from increasing it. Insulin (0.3 U kg(-1)) and, to some extent, glibenclamide (0.1 mg kg(-1)) but not metformin could be effective antiarrhythmic agents against oubain-induced arrhythmias. (C) 2002 Elsevier Science Ltd.