Influence of heart rate and vasoactive drugs on blood flow patterns at the canine ilio-femoral bifurcation.

OBJECTIVE: The aim was to study the effects of altered heart rate and vasoactive drugs on the blood velocity patterns in the region of an arterial bifurcation. METHODS: Blood velocity profiles were measured in an exposed iliofemoral bifurcation of paced dogs using a pulsed Doppler ultrasound velocimeter with high temporal and spatial resolution. RESULTS: Decrease of the heart rate from 120 beats.min-1 (2 Hz) to 60 beats.min-1 (1 Hz) increased the peak forward velocity (30%), the peak reverse velocity (20%), and the duration of reverse flow (25%). Each drug caused qualitatively similar changes in velocity patterns at both heart rates. The systemic administration of angiotensin II reduced peak forward velocity (-26% at 2 Hz and -33% at 1 Hz) and forward flow duration (-15% at 1 Hz), the peak reverse velocity (-30% at 1 Hz), and reverse flow duration (-20% at 2 Hz and -28% at 1 Hz). Glyceryl trinitrate also reduced the peak forward velocity (-19% at both 2 and 1 Hz) but prolonged forward flow duration (28% at 2 Hz and 17% at 1 Hz) and that of reverse flow (45% at 2 Hz and 24% at 1 Hz), and also decreased the degree of oscillation (-16% at 2 Hz). Barnidipine hydrochloride (a calcium channel antagonist) also increased the duration of forward flow (48% at 1 Hz) and of reverse flow (31% at 2 Hz) but reduced the peak reverse velocity (-29% at 1 Hz) and flow oscillation (-22% at 2 Hz and 20% at 1 Hz). CONCLUSIONS: These dramatic changes in the pattern of blood flow, including alterations in the amplitudes and durations of the different phases of the flow cycle, are expected to have important consequences on the shear dependent responses of endothelial cells in the region of the bifurcation.

Effect of packed cell volume on diastolic coronary artery pressure-flow relations in the dog.

To elucidate the role of the haemorheological properties of the perfusate in the coronary circulation, the diastolic pressure-flow relation was studied in nine open chest heart blocked dogs with minimal vasomotor tone when blood with various packed cell volumes (12-67%) was used as perfusate. An electrical analogue model with proximal resistance R1, capacitance C, distal resistance R2, and the zero flow pressure intercept Pint was derived from the observation of the pressure-flow relation to support the data analysis. The diastolic pressure decay was then determined after the perfusion line had been clamped to calculate stop flow coronary artery pressure (Psf). The stop flow coronary artery pressure decreased in relation to packed cell volume (r = 0.45, p less than 0.01), and the value for the lowest packed cell volume (10-29%) was slightly higher than the great cardiac vein pressure (about 3 mmHg). The zero flow pressure intercept of the steady state pressure-flow relation showed a close correlation with the stop flow coronary artery pressure (r = 0.87, p less than 0.001). The value of R1 + R2, which reflects the inverse of the steady state pressure-flow slope, decreased simultaneously with the packed cell volume (r = 0.62, p less than 0.001). The resistance ratio R2/(R1 + R2) by our model prediction decreased in relation to packed cell volume (r = 0.5, p less than 0.001). The values of stop flow coronary artery pressure, zero flow pressure intercept, and R1 + R2 for the highest packed cell volume (50-69%) were 17.8(1.1) mmHg, 25.1(1.3) mmHg, and 0.48(0.05) mmHg.ml-1.min.100 g-1 respectively, whereas those for the lowest packed cell volume (10-29%) were 13.4(0.8) mmHg, 19.7(1.0) mmHg, and 0.24(0.02) mmHg.ml-1.min.100 g-1. The pressure difference between the stop flow coronary artery pressure and the zero flow pressure intercept may be due to the non-linearity in the pressure-flow relation at a low perfusion pressure. The left ventricular end diastolic pressure and great cardiac vein pressure did not change in relation to the packed cell volume of the coronary perfusate. Thus it is concluded that packed cell volume is one factor determining the high zero flow pressure.

Velocity profiles and phasic flow patterns in the non-stenotic human left anterior descending coronary artery during cardiac surgery.

OBJECTIVE: The aim was to investigate the phasic characteristics of normal human left coronary artery flow and velocity profiles across the vessel. METHODS: The phasic characteristics of flow in the human left anterior descending coronary artery, the centreline flow velocities, and the velocity profiles were measured in 10 patients during corrective surgery for atrial septal defect after closure of the defect. None of these patients had any detectable coronary artery stenosis or left ventricular hypertrophy. Measurements were made with a 20 MHz 80 channel pulsed Doppler velocimeter. RESULTS: The velocity waveform displayed a diastolic-predominant pattern with a systolic to diastolic velocity ratio of 0.29(SD 0.17). Reverse flow was observed in early systole in five patients and in mid to late systole in six patients. The values of peak Reynolds number, unsteadiness parameter, and pulsatility index were 504(198), 2.5(0.6), and 5.9(4.4) respectively. The velocity profiles during diastole showed considerable variability in shape, ranging from symmetrical to skewed to M shaped patterns. The peak wall shear rate was 765(250) s-1 on the epicardial wall of the vessel and 712(301) s-1 on the myocardial wall; the difference was not statistically significant. CONCLUSIONS: The velocity waveform displayed a diastolic-predominant pattern. Considerable variability in shape of the velocity profile was found and was perhaps due to the time evolution of the velocity profile within the diastolic time period.

Atrial contractility affects phasic blood flow velocity of atrial small vessels in the dog.

OBJECTIVES: The aim was to evaluate the relative contribution of atrial muscle contraction and atrial pressure to the phasic patterns of left atrial arterial and venous flows. METHODS: Using a laser Doppler velocimeter, blood velocities were measured in the atrial small arteries and veins (outer diameter: 150-500 microns) in anaesthetised open chest dogs (n = 21). The velocity sensor was fixed on the vessel surface with a drop of cyanoacrylate glue when good quality Doppler signals were consistently observed. Left atrial pressure and the contractility of the left atrium were changed by premature ventricular contraction and by intracoronary injection of isoprenaline (0.5 microgram), respectively. RESULTS: Premature ventricular contraction increased left atrial pressure significantly during arterial velocity measurements from 8.1(SD 2.7) to 16.4(1.3) mm Hg and during venous measurements from 8.2(1.2) to 14.3(3.7) mm Hg. However, premature ventricular contraction did not change the blood velocity patterns, the maximum deceleration rate of the systolic velocity wave in arteries, or the maximum acceleration rate of the systolic velocity wave in veins. Although isoprenaline did not change the left atrial pressure, it decreased minimum arterial blood velocity during atrial systole, from 3.3(3.4) to -2.5(3.2) cm.s-1, and increased maximum venous blood velocity from 15.9(5.5) to 19.2(7.4) cm.s-1. Isoprenaline also increased both the maximum arterial systolic velocity deceleration rate, from 90(45) to 234(143) cm.s-2, and the maximum venous systolic velocity acceleration rate from 356(230) to 763(366) cm.s-2. CONCLUSIONS: (1) Left atrial pressure is not a major determinant of the blood flow patterns of the atrial arteries and veins, and therefore it may not closely reflect pressure around mural vessels. (2) Atrial contractility affects the blood flow patterns of the atrial arteries and veins.

Endothelium-derived nitric oxide enhances the effect of intraaortic balloon pumping on diastolic coronary flow.

BACKGROUND: High shear rate with pulsation is one of the major stimuli for the release of endothelium-derived nitric oxide leading to coronary arteriolar dilation. Intraaortic balloon pumping mechanically enhances shear rate and diastolic-to-systolic flow oscillation. We aimed to evaluate whether or not coronary blood flow augmentation during intraaortic balloon pumping is mediated by coronary arteriolar dilation through endothelium-derived nitric oxide release. METHODS: Using a charge-coupled device intravital videomicroscope, we observed epicardial coronary arterioles (40 to 220 microm in diameter) in anesthetized open-chest dogs (n = 10) during 2:1 mode of intraaortic balloon pumping. Endothelium-derived nitric oxide-mediated vasodilatory effects of intraaortic balloon pumping were evaluated by comparing end-diastolic arteriolar diameters between the coupled beats of on and off intraaortic balloon pumping before and after intracoronary endothelium-derived nitric oxide synthesis inhibition with Nomega-nitro-L-arginine (L-NNA, 2 micromol/min) administration. RESULTS: Intraaortic balloon pumping increased coronary arteriolar diameters and coronary blood flow by 11.4%+/-1.8% (p < 0.0001) and 33.4%+/-4.1% (p < 0.001), respectively. Vasodilation was greater in small arterioles (<110 microm; 15.4%+/-2.2%) than in large arterioles (> or =110 microm; 4.2%+/-1.2%, p < 0.0001). L-NNA attenuated the intraaortic balloon pumping-induced vasodilation and augmentation of coronary blood flow to 4.6%+/-1.0% (p < 0.001) and to 20.8%+/-2.1%, (p < 0.05), respectively. Attenuation of vasodilatory effect by L-NNA was observed mainly in small arterioles (from 15.4%+/-2.2% to 5.9%+/-1.2%). CONCLUSIONS: Intraaortic balloon pumping augmented coronary blood flow by dilating coronary arterioles in diastole, more significantly in small arterioles than in large arterioles. Endothelium-derived nitric oxide inhibition markedly attenuated these effects. We conclude that, in a canine model, endothelium-derived nitric oxide contributes to mechanical enhancement of the coronary blood flow with diastolic arteriolar vasodilation during intraaortic balloon pumping.

Effects of size and shape (aspect ratio) on the hemodynamics of saccular aneurysms: a possible index for surgical treatment of intracranial aneurysms.

OBJECTIVE: The present study was undertaken to explore the relationship between the characteristic geometry of aneurysms prone to rupture and the blood flow patterns therein, using microsurgically produced aneurysms that simulated human middle cerebral artery aneurysms in scale and shape. METHODS: We measured in vivo velocity profiles using our 20-MHz, 80-channel, Doppler ultrasound velocimeter. We produced small (< or =5 mm, 5 cases) and large (6-13 mm, 12 cases) aneurysms with round, dumbbell, or multilobular shapes. RESULTS: The fundamental patterns of intra-aneurysmal flow were composed of inflow, circulating flow, and outflow. The inflow, which entered the aneurysm only during the systolic phase, was strongly influenced by the position and size of the neck and the flow ratio into the distal branches. The outflow was usually nonpulsatile and of low velocity. The circulating flow depended on the aspect ratio (depth/neck width). A single recirculation zone was observed in aneurysms with aspect ratios of less than 1.6. This circulation did not seem to extend to areas with aspect ratios greater than this value; in aneurysms with aspect ratios of more than 1.6, a much slower circulation was observed near the dome. Furthermore, in the dome of dumbbell-shaped aneurysms and daughter aneurysms, no flow was detected. Intra-aneurysmal flow was determined by the aspect ratio, rather than the aneurysm size. CONCLUSION: The localized, extremely low-flow condition that was observed in the dome of aneurysms with aspect ratios of more than 1.6 is a common flow characteristic in the geometry of ruptured aneurysms, so great care should be taken for patients with unruptured intracranial aneurysms with aspect ratios of more than 1.6.

Endocardial coronary microcirculation of the beating heart.

Direct and continuous observation of subendocardial (deep myocardial) microcirculation provides essential information on coronary circulation, since cardiac contraction affects subendocardial vessels most vigorously. To achieve this aim, we developed a portable needle-probe video-microscope with a charge-coupled-device (CCD) camera to visualize the subendocardial microcirculation. Images of the subendocardial microcirculation of a porcine beating heart were successfully observed in all cases. The vascular compression by cardiac contraction decreased the diameter of subendocardial arterioles and venules by about 20%.

Dual-fiber laser Doppler velocimeter and its application to the measurements of coronary blood velocity.

To obtain a smaller sample volume and a suitable sample position for the measurement of blood velocity, we fabricated a laser Doppler velocimeter (LDV) with a dual-fiber pickup. The two fibers (clad: 62.5 micron and core: 50 micron) were placed side by side. An He-Ne laser was introduced into the blood through one fiber and the backscattered light was collected by the other fiber. The Doppler signal was analyzed by a spectrum analyzer. The spectrum of the Doppler shift frequency showed a sharp peaked pattern for both forward and reverse flows and exhibited an excellent correlation with the known blood velocity. The blood velocity in the poststenotic portion of canine coronary artery was successfully measured by the dual-fiber LDV. These results indicate that the dual-fiber LDV is useful for measuring blood velocity accurately with a small sample volume even in disturbed flow fields.

Noninvasive assessment of left ventricular force-frequency relationships by measuring carotid arterial wave intensity during exercise stress.

BACKGROUND: Evaluation of the contractile state of the left ventricle during exercise is important in drawing up a protocol of cardiac rehabilitation. It has been demonstrated that color Doppler- and echo tracking-derived carotid arterial wave intensity is a sensitive index of global left ventricular (LV) contractility. OBJECTIVES: We assessed the feasibility of measuring carotid arterial wave intensity and determining force-frequency (contractility-heart rate) relationships (FFR's) during exercise totally noninvasively. METHODS: We measured carotid arterial wave intensity with a combined color Doppler and echo tracking system in 15 healthy young male volunteers (age 20.8 ± 1.3 years) at rest and during exercise. FFR's were constructed by plotting the maximum value of wave intensity (WD1) against heart rate (HR). RESULTS: WD1 increased linearly with an increase in HR. The goodness-of-fit of the regression line of WD1 on HR in each subject was very high (r2 0.67 ~ 0.91, p < 0.0001 respectively). The slope of the WD1-HR relation ranged from 0.31 to 1.52 [m/s(3)(beat/min)]. CONCLUSIONS: A global LV FFR can be generated in healthy young volunteers with an entirely noninvasive combination of exercise and wave intensity. These data should show the potential usefulness of FFR in the context of cardiac rehabilitation.

An optical-fiber laser Doppler velocimeter and its application to measurements of coronary blood flow velocities.

In this paper we describe a laser Doppler velocimeter (LDV) with an optical fiber that measures blood flow velocities accurately in a small sample volume. The principle, optical arrangement, spatial and the temporal resolutions and accuracy for blood flow measurements are delineated, followed by a report of the results of measurements of coronary artery and vein blood flow velocities in dogs. Finally, we touch upon some recent progress made in the LDV with an optical fiber pickup.

Blood velocity patterns in poststenotic regions and velocity waveforms for myocardial inflow associated with coronary artery stenosis in dogs.

Velocity profiles across a vessel were investigated in poststenotic regions of the canine left coronary artery by our 80-channel 20 MHz ultrasound velocimeter. The velocity waveform in a small artery just before its penetration into myocardium was measured by our laser Doppler method. The poststenotic velocity configuration was characterized by a narrow region of high velocity with diastolic reverse flow near the wall which may dissipate energy. The velocity waveform in the distal small arteries exhibited increased systolic reverse flow with decreased diastolic forward flow, resulting in a remarkable reduction of coronary inflow into the myocardium.

Analysis of the characteristics of the flow velocity waveforms in left atrial small arteries and veins in the dog.

To clarify the characteristics of the phasic blood velocity pattern in small arteries and veins on the left atrial surface, we used our newly developed fiber-optic laser Doppler velocimeter. We intended particularly to examine the influence of atrial contraction and relaxation on velocity waveforms to obtain some insight into the nature of the mechanical force acting on the atrial intramyocardial vascular beds. In 14 anesthetized open-chest dogs, the left atrial appendage was gently displaced to expose small branches of the artery and vein. Vessels with an outer diameter of about 150-500 microns were chosen for the measurements because their walls are transparent to laser light. The fiber tip (velocity sensor) was fixed on the vessel surface with a drop of cyanoacrylate when good-quality Doppler signals were consistently observed. Additional experiments with three dogs were performed to observe the blood velocities in the atrial artery and vein during arrhythmia. The blood velocity waveform in the artery was similar to the pattern of aortic pressure during ventricular ejection (peak velocity, 18.8 +/- 7.8 cm/sec) but was characterized by a pronounced dip during atrial contraction. The temporal coincidence between the dip formation and atrial contraction was confirmed during atrial flutter with an atrioventricular block. After isoproterenol administration (2 micrograms i.v.), the acceleration rate of the forward flow velocity increased by 176% (p less than 0.05), and reverse flow appeared during atrial contraction in five cases out of eight (p = 0.013). The blood flow velocity in atrial small veins, on the other hand, was predominant during atrial contraction (peak velocity, 15.6 +/- 5.8 cm/sec). Isoproterenol increased the acceleration rate of this forward flow velocity by 121% (p less than 0.01). Nitroglycerin did not change the blood velocity waveform significantly in atrial arteries or in veins. The phase opposition between arterial inflow into and venous outflow from the atrial myocardium indicates that a large portion of the coronary inflow to the atrial myocardium may be stored due to the presence of atrial myocardial vascular capacitance. We conclude that atrial myocardial contraction impedes atrial inflow and promotes venous outflow from atrial capacitance vessels.

Role of NO and K(+)(ATP) channels in adenosine-induced vasodilation on in vivo canine subendocardial arterioles.

Adenosine (Ado) plays an important role in regulation of coronary vascular tone with nitric oxide (NO) and ATP-sensitive K(+) (K(+)(ATP)) channels. In vitro, it was reported that subendocardial (Endo) arterioles are more sensitive to Ado than subepicardial (Epi) arterioles. The purpose of this study was to observe enhanced vasodilation of Endo arterioles directly and to evaluate possible roles of K(+)(ATP) channels and NO in the different responses of Endo and Epi arterioles to Ado-induced vasodilation. We evaluated dilation of Endo and Epi arterioles (<120 micrometer) of beating canine hearts (n = 19) by Ado (20 and 50 microgram. kg(-1). min(-1) ic) before and after K(+)(ATP) channel blockade (glibenclamide; 200 microgram/kg ic), inhibition of NO synthase [N(G)-nitro-L-arginine methyl ester (L-NAME); 30 microgram. kg(-1). min(-1), 20 min ic], or glibenclamide + L-NAME using a novel needle-probe CCD intravital microscope. Ado induced dose-dependent vasodilation in both Epi and Endo arterioles, but vasodilation was greater in Endo arterioles, i.e., increase at 120 s (maximum dilation) after Ado (50 microgram. kg(-1). min(-1)) was 17% in Endo and 13% in Epi arterioles (P < 0.01). Endo arteriole dilation was attenuated by blockade of K(+)(ATP) channels from 18% (Ado) to 9% (Ado+glibenclamide) increase (P < 0.001) and by inhibition of NO synthase from 17% (Ado) to 9% (Ado+L-NAME) (P < 0.005). Epi arteriole vasodilation was attenuated by blockade of K(+)(ATP) channels from 15 to 9% (P < 0.005) and inhibition of NO from 16 to 10% (P < 0.005). Suppression of vascular response was additive (Endo, 14 to -1%; Epi, 12 to 3%) with glibenclamide + L-NAME. We conclude that 1) the degree of Ado-induced vasodilation was greater in Endo than in Epi arterioles, with higher sensitivity of smaller arterioles in both layers and 2) transmural difference of arteriolar sensitivity to adenosine was abolished or reversed by K(+)(ATP) channel blockade and/or by NO synthase inhibition, indicating crucial involvement of K(+)(ATP) and NO in transmural sensitivity difference.

A laser Doppler catheter for monitoring both phasic and mean coronary vein flow.

A new catheter-type laser Doppler velocimeter has been developed to monitor coronary vein flow. A thin graded-index multimode optical fiber (outer diameter of 125 microns) is set inside a 5-F catheter, and eight elastic silicon rubber spikes are arranged radially toward the vessel wall to fix the catheter tip in or near the axial region of the coronary vein. He-Ne laser light (wave length = 632.8nm) is introduced into the blood through the optical fiber, and reflected light is collected by the same fiber. The Doppler signal is detected by a spectrum analyzer. To avoid any effect by the spikes on flow, the fiber is extended from the catheter tip by 3 mm at the time of measurement. Straight and curved tubing was used to examine the accuracy of flow measurement. The flow velocities recorded by the catheter, which were measured by an electromagnetic flowmeter, exhibited excellent linearity (r = straight; 0.982, curved: 0.996). The blood flow velocity in the great cardiac vein was measured by this method in five dogs. The predominantly systolic waveform, which is a characteristic of the coronary vein flow, was observed in all of the dogs. The great cardiac vein velocity increased around the beginning of the ventricular ejection and decreased gradually after the peak formation at mid- or end-diastole. In addition to this main peak, small flow components were frequently observed during isovolumic contraction and the atrial contraction phase, although these flow components varied in individual dogs. Following left anterior descending artery occlusion, the great cardiac vein flow velocity decreased significantly.(ABSTRACT TRUNCATED AT 250 WORDS)

[A case of oral florid papillomatosis combined with gastroduodenal polyposis].

A 49-year-old man developing numerous papillary or partly cauliflower-like tumors over the lips, oral angles, gingiva and tongue starting 15 years earlier and associated with multiple gastroduodenal polyps of various sizes revealed by X-ray examination is presented. Histologically, the oral lesions consisted of marked proliferation of mature squamous cells without appreciable cellular atypism or submucosal invasion. Electron microscopic study failed to detect any virus-like particles, and negative reaction was obtained in immunohistological study for papilloma virus. The association of gastroduodenal polyposis with oral florid papillomatosis has appeared very rarely in the literature. Although the relationship between the two lesions remains unknown, it was assumed in the present case that all lesions would represent an entity of a hamartomatous nature.

Effect of dietary control on plasma nitrate level and estimation of basal systemic nitric oxide production rate in humans.

It is of great interest and value to evaluate the systemic nitric oxide (NO) production rate in humans under various conditions. However, the currently available estimation methods are troublesome and time-consuming. We thus aimed at developing a simple method to estimate the basal systemic NO production rate in humans based on a steady-state analysis, i.e., a balance between the systemic NO production rate and the total nitrate elimination rate. Plasma nitrate concentrations of young healthy volunteers (n = 7 in group 1: n = 9 in group 2) were measured for 2 days. In group 1, all subjects had the same meals for 7 days prior to the plasma nitrate measurement. In group 2, all subjects were allowed free diets. The plasma nitrate concentrations were highly influenced by dietary nitrite/nitrate intake in both groups and reached the steady-state levels after 14-h fasting. Accordingly, the basal systemic NO production rates were estimated from the plasma nitrate concentrations after 14-h fasting (group 1, 630 +/- 37 nmol min(-1) = 0.78 +/- 0.03 micromol kg(-1) h(-1); group 2, 597 +/- 45 nmol min(-1) = 0.66 +/- 0.05 micromol kg(-1) h(-1), P = not significant vs group 1). These estimated values were comparable to the values obtained by other methods. In conclusion, the present estimation method with 14-h fasting using a single-compartment analysis was found to be a simple approach to quantitative evaluation and intra- and interindividual comparisons of the basal systemic NO production rates in humans.

Diameters of subendocardial arterioles and venules during prolonged diastole in canine left ventricles.

Using a needle-probe videomicroscope with a charge-coupled device (CCD) camera, we measured the diameter of subendocardial arterioles and venules during prolonged diastole beyond the time point at which coronary blood flow reached zero. In seven open-chest heart-blocked dogs, a sheathed needle probe with a doughnut-shaped balloon was introduced from the left atrial appendage and advanced into the left ventricle through the mitral valve. The tip of the probe was placed gently on the endocardial surface. Diameters of arterioles (n = 16) and venules (n = 16) at the beginning of long diastole ranged from 40 to 126 microns and from 32 to 192 microns, respectively. After cardiac arrest, the arteriolar diameter gradually declined with aortic pressure. Arteriolar diameters at zero flow decreased by 28 +/- 9% (mean +/- SD) compared with the initial diameter (P < .01). However, none of the subendocardial arterioles collapsed at zero flow or at 12 seconds after the beginning of prolonged diastole (8 to 9 seconds after zero flow) in an additional experiment (n = 5). In contrast to arteriolar diameter, venular diameter increased during prolonged diastole. Venular diameter at zero flow increased by 14 +/- 12% compared with the initial diameter (P < .01). We conclude that during prolonged diastole, when coronary arterial inflow ceases, subendocardial arteriolar diameter decreases without any visible collapse, whereas venular diameter increases.

Effects of nitroglycerin on diameter and pulsation amplitude of subendocardial arterioles in beating porcine heart.

We examined the effect of nitroglycerin (NTG) on the diameter and diastolic-to-systolic pulsation amplitude of large (ID > 100 microns) and small (ID < 100 microns) subendocardial arterioles with their segmental responses. In 10 open-chest, anesthetized pigs, subendocardial arterioles of beating hearts (n = 18) were videorecorded using a needle-probe videomicroscope. Subendocardial arteriolar diameter was determined before and 1-2 min after NTG administration (25 micrograms/kg i.v.). In an additional experiment using three pigs, we monitored the transient response of subendocardial small arterioles (n = 5) from the time before NTG administration until 3 min after NTG. NTG dilated large subendocardial arterioles by 13 +/- 4% (means +/- SD, n = 10, P < 0.001) at about 1.5 min after NTG, but not small subendocardial arterioles (2 +/- 2%, n = 8, not significant). However, the small arterioles responded transiently to NTG in an earlier phase, especially to a higher dose. The percent diameter change of large subendocardial arterioles between diastole and systole at 1.5 min after NTG administration was 32 +/- 4%, which was larger than that under control conditions (19 +/- 8%, P < 0.05). The pulsation amplitude of small subendocardial arterioles at this time was almost unchanged by NTG (16 +/- 8 vs. 17 +/- 6%, NS) but increased transiently in an earlier phase. In conclusion, NTG dilated large subendocardial arterioles on the plateau phase of its impulse (intravenously) response (approximately 1.5 min after NTG).(ABSTRACT TRUNCATED AT 250 WORDS)