OBJECTIVE The opening of potassium (K+) channels during regional ischaemia may, by inducing rapid contractile arrest, be an intrinsic energy sparing mechanism. Thus K+ channel openers (for example, lemakalim) exert significant anti-ischaemic effects, whereas glibenclamide exacerbates ischaemic contracture and limits postischaemic functional recovery. The aim of the study was to investigate the ability of these compounds to influence ischaemic injury when used either alone or in combination with rapid arrest induced by a high K+ cardioplegic solution. METHODS Contractile function of isolated Langendorff perfused rat hearts was assessed using an intraventricular balloon. Hearts were subjected to normothermic global ischaemia (20 min) or cardioplegic arrest (35 min) with and without glibenclamide or lemakalim. Lemakalim (10 mumol.litre-1) or glibenclamide (10 mumol.litre-1) was given, in the presence or absence of cardioplegia, for 2 min immediately prior to the onset of ischaemia. The rate of ischaemia induced contractile failure, the severity of ischaemic contracture, and the degree of postischaemic functional recovery were all measured. RESULTS In the absence of cardioplegia, the time to contractile arrest in control hearts was 133 (SEM 4) s. This was increased by glibenclamide, to 145(6) s, and decreased by lemakalim, to 112(6) s. The time to onset of ischaemic contracture [8(1) min] was accelerated by glibenclamide [4(1) min] and delayed by lemakalim [14(1) min]. Lemakalim significantly improved the recovery of left ventricular developed pressure from 49(7)% in control hearts to 65(3)%, and left ventricular end diastolic pressure from 41(3) to 21(4) mm Hg. Hearts pretreated with glibenclamide showed similar recoveries to control hearts. During reperfusion, lemakalim exerted a transient vasodilator effect whereas glibenclamide caused a transient vasoconstriction. When either glibenclamide or lemakalim was added to a high K+ cardioplegic solution and hearts rendered ischaemic for 35 min, the ability of both compounds to influence ischaemic contracture and postischaemic functional recovery was lost. In additional studies the effect of lemakalim on the relative times to ischaemia induced mechanical failure and electrical arrest was assessed. In control hearts the time to contractile failure was 128(5) s and the time to electrical arrest was 241(30) s, while in the lemakalim treated hearts the values were 103(2) s and 509(161) s, respectively. In the lemakalim group all the hearts showed sustained ventricular arrhythmias; in the control group there were no arrhythmias. CONCLUSIONS Lemakalim can exert a significant anti-ischaemic effect when given alone. This effect is lost when it is used in combination with high K+ cardioplegia. The anti-ischaemic properties of lemakalim may thus be limited to its ability to accelerate contractile arrest.