The role of beta-adrenoceptors in coronary blood flow distribution in normal and ischemic canine myocardium.

beta-Adrenoceptor agonists increase myocardial ischemic injury, mainly by elevating myocardial oxygen consumption. Moreover, it has been shown that isoprenaline may "steal" regional myocardial blood flow (RMBF) from ischemic to non ischemic areas and from epicardium to endocardium. The mechanisms of these two isoprenaline-induced redistributions of RMBF have been investigated by the use of radioactive microspheres in an experimental model of canine myocardial ischemia with simultaneous measurement of ST-segment elevation. Isoprenaline increased RMBF in both epi- and endocardial non ischemic areas and in epicardial ischemic areas, leading to a significant decrease in the endo/epi ratio. After atenolol, isoprenaline still increased RMBF but to a lesser extent and the endo/epi ratio was still decreased. Salbutamol, in doses inducing no significant changes in cardiac parameters or myocardial oxygen consumption, produced effects similar to those of isoprenaline. These results indicate a non-homogeneous beta2-stimulation-induced vasodilation in endo- and epicardium, which might be due either to the higher epicardial coronary vasocilatory reserve or to a heterogeneous distribution of transmural beta2-adrenoceptors. Isoprenaline also decreased the ischemic/non ischemic total blood flow ratio (I/NI) and caused further increases in ST-segment elevation. These effects were abolished by atenolol pretreatment, indicating the deleterious effects of isoprenaline-induced tachycardia in this I/NI decrease and in the ischemic injury.     

Effects of propranolol on regional myocardial blood flow and function during severe coronary stenosis in dogs

The effects of propranolol alone or associated with atrial pacing were studied on regional myocardial blood flows (RMBF) and regional contractility (sonocardiometry) in non-ischemic, moderately and severely ischemic areas of the canine myocardium. In non-ischemic areas, propranolol reduced both epicardial and endocardial flows, increased the endo/epi ratio and decreased regional contractility. The reductions in subendocardial flow and function were correlated. In moderately and severely ischemic areas, propranolol increased subendocardial flow, reduced subepicardial flow, increased the endo/epi ratio and preserved or even slightly improved regional contractility. There was a good correlation between the propranolol-induced protective effects on regional contractility and the drug-induced increase in subendocardial flow since under atrial pacing subendocardial flow no longer increased and regional function dropped dramatically.     

Effect of diltiazem on coronary blood flow distribution in dog heart under ischemia

Effect of intracoronary infusion of diltiazem (1 microgram/min) on regional myocardial blood flow (RMBF) was studied using 15-microns radioactive microspheres in 11 excised cross-circulated canine left ventricles. With total coronary blood flow (CBF) and heart rate (HR) held constant, regional ischemia was induced by ligating the left anterior descending coronary artery (LAD). Diltiazem at the dose used had no effects on ventricular Emax before and after LAD ligation. RMBF expressed by the counts divided by the counts averaged in all segments in each layer significantly (p less than 0.05) increased under diltiazem only in the low-flow region that had less than 50% RMBF before diltiazem; from 21% (+/- 12%) to 35% (+/- 18%) in the epicardial, from 22% (+/- 12%) to 32% (+/- 18%) in the midwall, and from 24% (+/- 10%) to 31% (+/- 12%) in the endocardial layers. We conclude that the beneficial effect of diltiazem on the ischemic heart involves a direct action on the coronary vascular system and does not necessarily depend on the concomitant changes in hemodynamics.     

丹参酮、纳络酮对缺血再灌注心肌局部血流量的影响

目的：观察丹参酮、纳络酮对缺血再灌注心肌局部血流量的影响。方法：用结扎冠状动脉左前降支的方法复制犬心肌缺血再灌注模型，心肌局部血流量用氢气清除法测定。结果：丹参酮、纳络酮均能明显增加正常及部分再灌注心肌血流量，丹参酮主要增加缺血周围区血流量，纳络酮可增加缺血中心区及周围区血流量。结论：丹参酮、纳络酮均能缩小缺血区范围，纳络酮还能减轻缺血程度。     

Postjunctional alpha-adrenergic stimulation of inotropy in hypoperfused myocardium outside an acute infarct.

The functional significance of myocardial postjunctional alpha-adrenergic support of inotropy in the vicinity of an acute regional ischemic zone was addressed in pentobarbital-anesthetized, beta-adrenergic blocked cats with circumflex coronary artery occlusion. Regional myocardial performance was measured by ultrasonic crystals in the anterior wall perfused by the left anterior descending coronary artery (LAD) before and during postjunctional alpha-adrenergic antagonism (SK&F 104078 2 mg/kg). A group with unrestricted flow in the LAD (control group) was compared with a group perfused below the autoregulatory pressure range (stenosis group). End-systolic pressure-length relations during dynamic after-load elevation were calculated for assessment of regional contractility. Regional myocardial blood flow (RMBF) was measured by radioactive microspheres. SK&F 104078 did not alter regional myocardial shortening or the slope of end-systolic pressure-length relations in the control group. In the stenosis group, however, alpha-adrenergic antagonism produced significant deterioration of shortening as well as consistent reduction of the slope of the end-systolic pressure-length relations (p < 0.05). As a reflection of reduced demands for perfusion, impairment of midmyocardial and endocardial blood flow occurred in the stenosis group (p < 0.05). These findings imply a negative inotropic effect of SK&F 104078 in metabolically vasodilated myocardium in the vicinity of an acute ischemic region.     

Effects of nicardipine, a dihydropyridine calcium antagonist, on regional myocardial blood flow, myocardial oxygen tension, and electrical abnormalities during acute coronary artery occlusion in dogs

Effects of nicardipine, a dihydropyridine calcium antagonist, on regional myocardial blood flow (RMBF), myocardial oxygen tension (PO2), and excitation and conduction abnormalities during the occlusion of the left anterior descending coronary artery (LAD) were examined in anesthetized dogs, and compared with those of nifedipine and dipyridamole. RMBF was calculated from the H2 gas clearance curves, and PO2 was measured using a membrane-coated Pt wire. Excitation and conduction abnormalities during the LAD occlusion were represented in terms of the degree of ST-T alternans (STTA), TQ depression, and conduction delay, which appeared in epicardial electrograms. Nicardipine and nifedipine in a dose of 10 micrograms/kg increased RMBF and PO2 levels in nonischemic and mildly ischemic tissues, but not in severely ischemic tissues. Nicardipine in a dose of 100 micrograms/kg and nifedipine in a dose of 10 micrograms/kg attenuated the degree of STTA, TQ depression, and conduction delay observed in severely ischemic tissues. In mildly ischemic tissues where only TQ depression was observed without STTA, nicardipine in a dose of 30 micrograms/kg attenuated TQ depression. Dipyridamole in a dose of 1 mg/kg produced only a slight attenuation of STTA and conduction delay. These results suggest that the beneficial effects of nicardipine as well as of nifedipine on myocardial ischemia are due to the increase in the myocardial PO2 levels caused by the increased RMBF and also to direct protecting effects on ischemic myocardial cells. In the severely ischemic tissues, the latter is a main effect of the drugs. In increasing the PO2 level, nicardipine was similarly potent as nifedipine, but in the direct effect, nicardipine was less potent, and dipyridamole was almost ineffective.     

Effects of intracoronary nifedipine on blood flow and segment shortening in normal and ischemic myocardium: potentiation by ischemia of the negative inotropic effect

The effects of intracoronary nifedipine on myocardial blood flow (flow probe or microspheres) and regional function (ultrasonic crystals in subendocardium) were examined both in the normal myocardium and in myocardium made ischemic by a partial coronary occlusion in the open-chest anesthetized dog. In a first group of experiments (n = 7), without ischemia, nifedipine infused into the left anterior descending coronary artery (LAD) during a 1-min period (doses 0.75-8 nmol/kg body weight) decreased coronary vascular resistance with a maximal effect at 4 nmol/kg. Systolic segment shortening was decreased from 10.7 to 7.4% (p less than 0.05) by 6 nmol/kg, whereas lower doses had no effect. In a second experimental group (n = 7), a partial LAD occlusion was applied to decrease subendocardial segment shortening by about 50%. Nifedipine (2 nmol/kg) injected into the partially occluded LAD induced a marked segmental bulging during early systole and systolic segment shortening was eliminated (from 4.2 to -3.1%, p less than 0.02) in the LAD-dependent myocardium. Concomitant with the decreased regional function, nifedipine caused a transmural redistribution of myocardial blood flow in the ischemic area, the endocardial/epicardial blood flow ratio increasing from 0.49 to 0.61 (p less than 0.02). It is concluded that ischemia potentiates the direct depressant effect of nifedipine on myocardial regional function.     

兔心室肌各层细胞缺血后瞬间外向钾电流的变化

目的　研究兔心室肌内膜层、中层、外膜层细胞缺血后瞬间外向钾电流 (Ito)的变化 ,以探讨Ito与心律失常的关系。方法　应用膜片钳全细胞记录方法 ,定量观察缺血前及缺血后Ito活性的变化 ,分析比较缺血后三层细胞Ito的不同反应。结果　缺血前随着刺激电压的增大 ,三层细胞的Ito逐渐增大 ,70mV时达最大值。外膜层细胞的电流密度明显大于中层及内膜层细胞 ,具有显著差异 (P <0 0 5 ) ;随缺血时间的延长 ,外膜层细胞的Ito呈逐渐下降趋势 ,中层细胞Ito先增大后下降 ,内膜层细胞变化不明显。结论　缺血后三层细胞Ito的变化 ,间接影响到动作电位 ,引起跨室壁复极离散度增大 ,可能是诱发缺血性室性心律失常的原因     

Alpha-adrenoceptor control of norepinephrine release from acutely ischaemic myocardium: effects of blood flow, arrhythmias, and regional conduction delay

We studied the effects of yohimbine, and alpha-adrenoceptor blocker with selectivity for the alpha 2-subtype, on myocardial norepinephrine (NE) overflow, regional myocardial blood flow (RMBF), and patterns of epicardial conduction abnormalities during occlusion of the proximal left anterior coronary artery in an open-chest anaesthetised dog model. With a 12-min period of coronary occlusion (n = 9), spontaneous overflow of NE into ischaemic venous effluent was not observed either before or after yohimbine (1 mg/kg i.v.), but the drug significantly potentiated the enhanced NE overflow during supramaximal stimulation of the left stellate ganglion at low (1 Hz) and high (10 Hz) frequency [peak NE 4.3 +/- 0.4 pmol/ml control; 11.8 +/- 5.4 pmol/ml yohimbine (p less than 0.005)] with a delayed return towards prestimulation levels. Myocardial NE overflow on coronary reperfusion was also enhanced. Yohimbine increased arterial epinephrine two- to threefold but did not substantially alter myocardial lactate overflow during coronary occlusion. RMBF was reduced 24 and 36% to ischaemic endocardium and epicardium, respectively (p less than 0.01, compared with control occlusion). This contrasted with a 9 and 6% decrease in flow to the respective nonischaemic areas (p = NS, compared with control occlusion. Spontaneous ventricular fibrillation and the area and magnitude of epicardial conduction abnormalities in the ischaemic myocardium were both increased compared with the control occlusion. Thus, alpha-blockers with selectivity for the alpha 2-adrenoceptor may be detrimental to acutely ischaemic myocardium, presumably through increased local catecholamine release at the nerve terminal.     

Epicardial His bundle recordings in the guinea pig in vivo.

In small animals, His bundle activity is commonly registered using intracavitary electrodes in Langendorff-perfused hearts. The present study evaluates the feasibility of epicardial registration of His bundle activity in 30 guinea pigs in vivo, anesthetized with 60 mg/kg sodium pentobarbital i.p. Following median sternotomy, a bipolar electrode, mounted on a flexible stand, was placed into the aortic-right atrial groove. His bundle activity was easily detected in all animals. Mean PA, AH, and HV intervals were 16.7 +/- 1.4 msec (range: 6.0-34.0 ms), 36.4 +/- 1.5 ms (range: 22.0-54.0 msec) and 14.4 +/- 0.4 msec (range: 9.0-20 msec), respectively. Administration of verapamil and vagal stimulation induced significant prolongations of AH intervals without affecting HV intervals; lidocaine prolonged both AH and HV intervals. Both at baseline and following drug administration there was excellent agreement between the AH and HV intervals assessed from simultaneously registered epicardial and endocardial His bundle electrograms (n = 12). The presented model provides reproducible values for atrioventricular (AV) nodal and His-to-ventricular conduction intervals in the guinea-pig in vivo and will, thus, be a valuable tool in electrophysiologic drug evaluation.     

Effects of quinidine on arrhythmias and conduction in an isolated tissue model of ischemia and reperfusion.

Transmembrane electrical activity from endo- and epicardium and a high-gain ECG were recorded from isolated segments of guinea pig right ventricles. Endocardium was stimulated. Tissues were exposed to ischemic conditions for 15 min and then reperfused with "normal" Tyrode's solution. Ventricular tachycardia, bigeminy, or trigeminy with characteristics of transmural reentry occurred in early reperfusion in 68% of control hearts. Arrhythmias were associated with prolongation of the transmural conduction time (CT) and abbreviation of the endocardial effective refractory period (EP). Quinidine significantly suppressed reperfusion arrhythmias at 1 and 5 microM, slightly increased the incidence of arrhythmias at 10 microM, and again suppressed arrhythmias at 50 and 100 microM. At 1 and 5 microM, quinidine prevented or attenuated prolongation of the transmural CT by ischemic conditions and reperfusion. The transmural CT was not significantly changed at 10 microM, and was further prolonged at 50 and 100 microM quinidine. The endocardial ERP was prolonged by 50 and 100 microM quinidine during ischemic conditions and reperfusion. In epicardial slices, 5 microM quinidine shortened the CT transverse to the fiber orientation during reperfusion but had no effect on the longitudinal CT. Thus, antiarrhythmic efficacy of low concentrations of quinidine may occur through differential effects dependent on tissue anisotropy.     

Effect of adenosine on transmural flow gradients in normal canine myocardium

Previous reports suggest that in comparison to epicardial flow the response of endocardial blood flow to systemic adenosine is limited. However, intravenous adenosine results in reflex tachycardia and hypotension, which we avoided in six anesthetized dogs by infusing adenosine into the left circumflex (LCx) artery to evaluate the endocardial/epicardial ratio during partial and full vasodilation. The mean (+/- SD) LCx endocardial/epicardial ratio was 1.37 +/- 0.18 with partial vasodilation and 0.81 +/- 0.06 with full vasodilation, which represents an initial increase followed by a decrease when compared with the normal left anterior descending values (p less than 0.05). This biphasic response in the endocardial/epicardial ratio with increasing vasodilation was a direct effect of adenosine, as no hemodynamic alterations occurred during the protocol. These data suggest that in an experimental model with a relative tachycardia the endocardial layer is more sensitive to adenosine but there is a greater capacity to increase flow in the epicardial layer.     

Beneficial effects of amiodarone pretreatment on early ischemic ventricular arrhythmias relative to infarct size and regional myocardial blood flow in the conscious dog

The effects of chronic pretreatment with amiodarone on ischemic ventricular arrhythmias were evaluated in fully conscious instrumented dogs. In control dogs (n = 14) with large myocardial infarcts, early (first 30 min) ventricular arrhythmias occurred in a bimodal distribution with peaks at 3-5 min and at 12-25 min, with only the former associated with epicardial conduction delay. Ventricular fibrillation occurred equally frequently during each peak of early ventricular arrhythmias. Amiodarone (30 mg/kg daily) for 3-4 weeks had no significant effect (n = 11) on anatomic infarct size (28 +/- 6 vs. 30 +/- 5% of left ventricular weight) nor on collateral blood flow in the center of the infarct (19 +/- 11 vs. 15 +/- 7 ml/min/100 g of tissue) or on the ratio of endocardial/epicardial perfusion (0.23 +/- 0.19 vs. 0.28 +/- 19). Despite significant lengthening of peak epicardial conduction delay (191 +/- 20 to 239 +/- 81 ms, p less than 0.05), the frequency of early ventricular arrhythmias, especially during the second peak of ectopic activity, were markedly attenuated by amiodarone pretreatment, with the extrasystole-free intervals often being as long as 6 h. The incidence of ventricular fibrillation was 9% in the treated animals compared with 29% in the controls. In the control animals, arrhythmias always supervened when epicardial fractionation was significant, and no ectopy-free interval was present in the first 6 h following coronary occlusion. The data indicate that chronic amiodarone pretreatment exerts a beneficial effect on the frequency and severity of such ventricular tachyarrhythmias, with reduction in the incidence of ventricular fibrillation and ectopic activity in the early phases following coronary occlusion.     

Preconditioning attenuates the shortening of recovery during coronary occlusion in isolated rabbit hearts with D-sotalol-induced long QT intervals

The effects of 20-min ligations of the anterior branch of the left coronary artery were studied in Langendorff-perfused rabbit hearts with 92 microM D-sotalol added to the perfusate to induce long QT intervals and triggered arrhythmias. Epicardial electrograms, a left ventricular endocardial monophasic action potential, and simulated X and Y lead electrocardiograms were used to characterize ventricular conduction and recovery. In contrast to previous work showing that global ischemia eliminated triggered activity, coronary occlusion did not alter its mean incidence. Although the anatomic distribution of earliest sites of epicardial activation by triggered beats was altered, triggered beats still appeared on the epicardial surface in the nonperfused regions. Coronary occlusion had a small and variable effect on epicardial conduction velocity but caused a significantly greater percent shortening of epicardial activation-recovery intervals in the nonperfused region of hearts given D-sotalol than in control hearts. In hearts given D-sotalol, preconditioning significantly attenuated the shortening of epicardial activation-recovery intervals in response to coronary occlusion. However, preconditioning had no effect on the mean incidence of triggered activity during coronary occlusion. Thus, the persistence of triggered activity and the shortened myocardial recovery time associated with coronary occlusion could contribute to increasing the likelihood of occurrence of malignant ventricular arrhythmias. Preconditioning by attenuating the shortening of recovery would be anti-arrhythmic.     

Effects of verapamil and lidocaine on changes in action potential characteristics and conduction time induced by combined hypoxia, hyperkalemia, and acidosis in canine ventricular myocardium

We examined the effects of verapamil and lidocaine on changes in action potential characteristics and conduction time in the isolated right ventricular myocardium of canine hearts during superfusion with altered Tyrode's solution imitating some of metabolic alterations that occur in acute myocardial ischemia (Po2 less than 50 mm Hg, KCl 8 mM, pH 6.80). Altered Tyrode's solution produced loss of resting membrane potential (RMP), action potential amplitude (APA), action potential duration, and upstroke velocity of action potential (Vmax), and prolonged conduction time (CT). In the presence of lidocaine (5 mg/L), altered Tyrode's solution aggravated the reductions of APA and Vmax and the prolongation of CT in both endocardial and epicardial muscle cells. On the other hand, in the presence of verapamil (1 mg/L), the degree of the reductions of APA and Vmax and of the prolongation of CT induced by altered Tyrode's solution was reduced in epicardial muscle cells, but not in endocardial ones. Neither verapamil nor lidocaine affected change in RMP. These results suggest that the improving effect of verapamil on ischemia-induced conduction delay can be partly explained by its salutary effect on the depressed fast channel.     

Effects of oral carbocromene pretreatment on infarct size following experimental coronary artery occlusion

This study examines the acute effects of the antianginal drug carbocromene (chromonar) in dogs (20 mg/kg p.o., twice daily for 8 weeks) on mortality, hemodynamics, coronary collateral blood flow, and myocardial infarct size. Following the chronic pretreatment and during acute phase of the experiments, the animals received an intravenous bolus of 4 mg/kg of carbocromene 15 min prior to left anterior descending coronary artery occlusion, and 40 micrograms/kg/min as an infusion during occlusion and reperfusion. Total mortality 2 days postocclusion was 50% in saline control experiments but 20% in carbocromene-treated animals (p less than 0.05). Hemodynamics were not significantly changed during drug administration except for a significant ST-segment elevation during vessel occlusion. Coronary collateral blood flow increased after carbocromene treatment by 30% (p less than 0.05) in the ischemic endocardial region and by 60% (p less than 0.02) in the border zone of the area at risk of infarction. Flow in nonischemic myocardium did not change so that "coronary steal" was not observed. At reperfusion, a flow increase occurred in the ischemic and border zones. Myocardial infarct size was 24% smaller after carbocromene than in control animals (p less than 0.02) when compared to the AR, and 46% smaller (p less than 0.01) in relation to the total left ventricle. We conclude that carbocromene administered orally before acute coronary artery occlusion and intravenously during occlusion and subsequent reperfusion can reduce infarct size by salvage of lateral and subepicardial border zones.     

Effects of verapamil on regional myocardial blood flow and ST segment. Role of the induced bradycardia

The effects of verapamil on regional myocardial blood flow and on the $\text{ST}$ segment were studied in both normal and ischemic regions in dogs with and without cardiac pacing.In the absence of cardiac pacing, verapamil (0.05 mg/kg/min/10 min) induced marked bradycardia, a drop in blood pressure and an increase in epicardial and endocardial flows both in normal and ischemic regions of the heart. However, in ischemic regions, redistribution was favorable since the endo/epi ratio increased from 0.46 to 0.61 (p<0.01) whereas this ratio did not vary in normal regions (0.94 vs. 0.92). This increase is accompanied by less marked $\text{ST}$ segment elevation during coronary occlusion in dogs treated with verapamil.In dogs with cardiac pacing, administration of verapamil under the same conditions again induced a drop in blood pressure and, in normal regions only, a homogeneous but less marked increase in epicardial and endocardial flows, the endo/epi ratio varying from 0.94 to 0.98. In contrast, in ischemic regions, blood flow was not modified by verapamil, but the endo/epi ratio dropped from 0.46 to 0.31 (p<0.01), while the $\text{ST}$ segment elevation observed did not differ from that recorded during the control occlusion.These results demonstrate the major role of verapamil-induced bradycardia in the anti-anginal effects of this drug.     

Effects of lidocaine on conduction of extrasystoles in the normal canine heart

The effect of lidocaine on the conduction of extrasystoles was studied in 8 open-chest dogs after atrioventricular nodal block. Simultaneous recording of endocardial and epicardial activation provided separate measures of endocardial (Purkinje) conduction as well as myocardial (muscle) conduction. Lidocaine (1.25--10.0 mg/kg) caused a dose-dependent slowing of conduction of midrange extrasystoles (250--400 ms) in both the Purkinje system and the myocardium, which became statistically significant at doses larger than 1.25 mg/kg. On the other hand, low doses of lidocaine caused speeding of early extrasystoles, i.e., coupling intervals (less than 250 ms) in the Purkinje system but not in the myocardium. Measurement of transmural conduction time as a function of coupling interval revealed a period of "apparent" supernormal conduction through ventricular muscle that was eliminated at high doses of lidocaine.     

Pentaerythritol trinitrate and glyceryl trinitrate on intramyocardial oxygenation and perfusion in the dog. Krogh analysis of transmural metabolism

1. The effect of intravenous pentaerythritol trinitrate and glyceryl trinitrate on left ventricular subepicardial (epi) and subendocardial (endo) Po₂ and perfusion were compared in anaesthetized open-chest mongrel dogs. Tissue Po₂ was determined simultaneously at a depth of 3 mm (epicardial) and 9 mm (endocardial) with small platinum electrodes by polarography. In a separate series of dogs tissue perfusion of those regions was measured by hydrogen (H₂) clearance using similar electrodes. 2. Both nitrates increased endocardial Po₂ while epicardial Po₂ was not altered. Perfusion was determined at the point of the maximal rise in endocardial Po₂ (4–7 min after injection of either nitrate). At that period average coronary artery inflow and epicardial perfusion were decreased but endocardial perfusion was not significantly altered. 3. Using the data on Po₂, hydrogen clearance and intercapillary distance, the effect of the nitrates on transmural metabolism (oxygen consumption) was estimated by Krogh analysis. Basal endocardial metabolism was 20–30% higher than epicardial metabolism. The nitrates reduced metabolism in each region. The absolute decrease in oxygen consumption was greater in the endocardium. 4. The results show that both pentaerythritol trinitrate and glyceryl trinitrate improve endocardial oxygenation by producing a more favourable balance between perfusion and oxygen requirements in that region.     

Protective effect of diltiazem against ischemia-induced decreases in regional myocardial flow in rat heart

Diltiazem has cardioprotective properties following myocardial ischemic injury. However, there are controversial results regarding the beneficial effects of diltiazem on regional myocardial flow after ischemia. Therefore, we investigated the effect of diltiazem on changes in regional myocardial flow due to ischemia for different periods. Non-radioactive colored microspheres were used for this measurement in isolated rat heart. After 20 or 40 min of global ischemia and 40 min of reperfusion, regional myocardial flow was decreased, especially in the endocardial layer. The endocardial/epicardial ratio was also decreased. The decreases in endocardial flow and the endocardial/epicardial ratio were more remarkable after 40 min of ischemia than after 20 min of ischemia. Diltiazem (10(-6) M), which was administered 15 min before ischemia, prevented only the decrease in endocardial flow and endocardial/epicardial ratio after 20 min of ischemia, whereas it did not prevent that after 40 min of ischemia. Nifedipine (2x10(-6) M) did not exert a cardioprotective effect. These findings suggested that the effect of ischemia is marked in the endocardium and, also, that the protective effect of diltiazem is seen only during a decrease in endocardial flow following short-term and reversible ischemia.