Role of Nitric Oxide in the Regulation of the Mechanical Properties of Peripheral Conduit Arteries in Humans
Abstract Whether nitric oxide (NO) contributes to the regulation of the mechanical properties of large arteries in humans is not known. We measured the effect of local administration of the inhibitor of NO synthesis NG-monomethyl-l-arginine (L-NMMA; 1 and 4 μmol · L−1 · min−1 for 5 minutes) and acetylcholine (3 and 30 nmol · L−1 · min−1 for 3 minutes) on radial artery diameter and wall thickness in 11 healthy volunteers using an echo-tracking system coupled to a measurement of radial blood flow (Doppler) and arterial pressure. At the highest dose, L-NMMA reduced radial blood flow but surprisingly decreased incremental elastic modulus (from 1.36±0.22 to 1.00±0.22 kPa · 103; P<.05) and increased arterial compliance (from 3.20±0.46 to 4.07±0.45 m2 · kPa · 10−8, P<.05), without affecting radial artery internal diameter, wall thickness or midwall stress, thus reflecting a decrease in vascular tone. Acetylcholine decreased incremental elastic modulus (from 1.27±0.08 to 0.88±0.07 kPa · 103; P<.05) and increased arterial diameter, radial blood flow, and compliance (from 2.82±0.16 to 5.30±0.62 m2 · kPa · 10−8; P<.05). These results demonstrate in vivo that NO is involved in the regulation of the mechanical properties of large arteries in humans. However, the effects of L-NMMA, ie, a decrease in arterial wall rigidity and an increase in arterial compliance, which occur in the absence of any changes in blood pressure or arterial geometry, suggest that inhibition of NO synthesis is associated in humans with a paradoxical isometric smooth muscle relaxation. This effect could be due to the development of compensatory vasodilating mechanisms after NO synthesis inhibition.
Arterial smooth muscle cell activation regulates mechanical properties of conduit arteries by increasing arterial wall stiffness through an increase in isometric tone.1–3 The endothelium modulates this arterial tone by the release of various vasodilating and constricting factors.4 Previous experiments demonstrated an NO–dependent vasodilatory tone at the arteriolar level in animals and in humans.5,6 Although the contribution of NO to flow-mediated vasodilatation has been clearly demonstrated both in animals7 and in humans,8 the role of basal release of NO at the level of large arteries is less clear. In humans, the existence of such a basal release is indirectly suggested by the increased sensitivity of large arteries to nitrovasodilators after inhibition of NO synthesis.9 Such a permanent release of NO should in theory decrease arterial stiffness and increase arterial compliance by means of a decrease in isometric arterial tone. However, in humans the effective contribution of NO to the regulation of arterial tone and the effect of NO synthesis inhibition on the mechanical properties of large arteries remain unknown. Thus, the present study was designed to assess whether NO contributes to the regulation of the mechanical properties of peripheral conduit arteries in humans.
The study was performed in 11 healthy volunteers (6 men and 5 women; age 25±1 years). All subjects were normotensive and nonsmokers. All had normal routine laboratory tests including blood glucose and total cholesterol. The protocol was approved by the ethics committee of the Basel University Hospital, the procedures followed were in accordance with institutional guidelines, and written informed consent was obtained from all participants.
Measurements were performed while subjects were supine in a quiet, air-conditioned room maintained at a constant temperature (22°C to 24°C). Under local anesthesia (lidocaine 1%), an 18-gauge catheter was inserted into the left brachial artery for continuous measurement of arterial pressure (Statham P23Pb)10 and infusion of L-NMMA.8,9 Radial artery internal diameter and wall thickness were continuously measured using a high-precision A-mode echo-tracking device with an axial resolution of 150 μm and a theoretical resolution of less than 1 μm on artery wall displacements (NIUS 02, Asulab).11,12 Briefly, a 10-MHz focused transducer was positioned over the radial artery. The probe was set perpendicularly to the artery without any skin contact using a stereotaxic arm with micrometric screws, while proper positioning was adjusted using a stereo Doppler mode. After switching to A-mode, the echoes reflected by the interfaces between blood and artery wall on one side and between artery wall and surrounding tissue on the other side could be identified for both the anterior and posterior walls and visualized on the screen display. After selection by the operator of the desired interfaces, the echoes reflected were tagged by electronic trackers, allowing continuous recording of the artery internal diameter and wall thickness.3 Radial artery internal cross-sectional area was then calculated from the measurement of internal diameter. Furthermore, radial artery blood flow velocity was continuously recorded using an 8 MHz Doppler probe (Doptek 2002, Deltex). Radial artery flow was calculated from the measurements of velocity and internal cross-sectional area. The method has been validated with target measurements and comparisons between in vitro and in vivo measurements.12 The reproducibility of the method in humans assessed by calculating the coefficient of variation between consecutive measurements was 2±1%, 3±1%, 7±2%, respectively, for arterial internal diameter, wall thickness, and blood flow.3–8
From the measured variables, arterial pressure (p), internal diameter (d), and wall thickness (h), the cross-section (S)–pressure curve was fitted using an arctangent model as proposed by Tardy et al11: S=α[π/2+tang−1([p−β]/γ)], S=πd2/4, where α, β, and γ are the three optimal fit parameters of the model.
To eliminate the hysteresis loop in the diameter-pressure graph due to the distance between the sites of measurements of arterial diameter and pressure, we used a previously reported method that is based on the evaluation of the phase lag between the diameter and pressure signals from sectional pulse-wave velocity calculated at each level of pressure.11
By taking the derivate of the cross-section-pressure function, the cross-sectional compliance (C) was expressed as C=(α/γ)/[1+([p−β]/γ)2].
Assuming isotropic and incompressible material, the incremental elastic modulus (Ei) for a thick-walled artery was expressed as a function of pressure13 as Ei=3/8[d(d+2h)2/ h(d+h)]∂p/∂d.
With the same assumptions, the midwall stress (s) was calculated as3 ς=2p(rire/r)2/(re2−ri2), where re and ri, are the external and internal radii, respectively, and r the radius at midwall (r=(re+ri)/2).
Subjects were allowed to rest for at least 30 minutes after instrumentation. Then heart rate, arterial pressure, radial artery flow, diameter, and wall thickness were recorded for 10 minutes during saline infusion at a constant rate of 0.8 L · min−1 10−3. After control measurements (physiological saline), two doses of L-NMMA (1 and 4 μmol L−1 min−1, Clinalfa) were infused for 5 minutes at the same infusion rate as physiological saline in the 11 volunteers. In 7 subjects, after control measurements (physiological saline), two doses of acetylcholine (3 and 30 nmol L−1 min−1, Ciba Vision AG) were infused for 3 minutes (infusion rate of 0.8 L min−1 10−3) 40 minutes before L-NMMA infusion. All parameters were continuously recorded from baseline to the end of the infusion period of L-NMMA and acetylcholine. From this continuous recording mean values were calculated from 15 consecutive cardiac cycles during the last minute of each infusion period. As previously reported, when an oscillatory pattern was noted, we extended the calculations to 30 to 40 consecutive cardiac cycles in order to cover the half period of oscillation and to obtain stable mean values.9 At each time, parameters were calculated both at operational pressure (mean arterial pressure) and to avoid the passive effects of changes in systemic arterial pressure, at a constant pressure of 80 mm Hg. This level of pressure was chosen because it was reached at each time point of the experiment in all subjects.
Results are expressed as mean±SEM. The analysis of the effect of L-NMMA and acetylcholine on the different parameters investigated was done by repeated measures ANOVA, followed by Tukey’s test when applicable. The relations obtained after L-NMMA and acetylcholine at the maximal doses were compared with those obtained at baseline using an ANCOVA with midwall stress as covariate and subjects and periods as factors. The analysis was repeated adding arterial flow as covariate to consider the effect of associated change in blood flow. For the latter analysis, reported values were adjusted least square mean±SEM. A value of P<.05 was considered statistically significant.
Mean Hemodynamic and Mechanical Parameters Obtained at Baseline and After L-NMMA and Acetylcholine
Table 1⇓ summarizes the geometric and mechanical parameters of the radial artery obtained at baseline during the control, acetylcholine, and L-NMMA experiments. There were no significant differences between the mean values of the parameters at baseline.
Forearm infusion of L-NMMA or acetylcholine did not modify arterial pressure or heart rate (Table 2⇓). In addition, there was no significant change in the cross-sectional area of the arterial wall during the three periods (from 4.77±0.44 to 4.77±0.44 m2 · 10−6 during the control period, NS; from 4.78±0.37 to 4.79±0.42 m2 · 10−6 after 4 μmol · L−1 · min−1 L-NMMA, NS; from 4.93±0.28 to 4.82±0.28 m2 · 10−6 after 30 nmol · L−1 · min−1 acetylcholine, NS).
L-NMMA decreased radial blood flow (Table 2⇑) but did not affect radial artery internal diameter, wall thickness, or midwall stress (Fig 1⇓). After inhibition of NO synthesis, the incremental elastic modulus of the arterial wall decreased (P<.05) and compliance increased (P<.05, Fig 1⇓).
Acetylcholine increased radial blood flow (Table 2⇑) and internal diameter (P<.05), and decreased radial artery wall thickness (P<.05, Fig 2⇓). Those changes in arterial geometry obtained without any change in arterial pressure were associated with an increase in midwall stress (P<.05). Despite this increase in midwall stress, acetycholine decreased incremental elastic modulus of the arterial wall (P<.05) and increased mean arterial compliance (P<.05, Fig 2⇓).
Effects of L-NMMA on Diameter–, Modulus–, and Compliance–Midwall Stress Curves
To assess the effect of L-NMMA at identical levels of wall-loading conditions, arterial diameter and elastic modulus were assessed as a function of midwall stress. Fig 3⇓ shows the diameter–and elastic modulus–midwall stress curves at baseline and during L-NMMA infusion. The diameter increased with wall stress. After L-NMMA the diameter–midwall stress curve was not shifted significantly.
The elastic modulus increased with midwall stress. After L-NMMA infusion the modulus–midwall stress curve was shifted downward (P<.01). Thus, NO synthesis inhibition was associated with a decrease in arterial wall stiffness at every level of midwall stress.
Fig 4⇓ shows the compliance–midwall stress curves at baseline and during L-NMMA infusion. Compliance decreased in a curvilinear manner when wall stress increased. During NO synthesis inhibition, compliance increased at every level of stress (P<.01).
The decreased arterial wall stiffness and increased compliance that occurred in the absence of any change in arterial pressure or geometry suggest that inhibition of NO synthesis was associated with a paradoxical isometric smooth muscle relaxation.
Effects of Acetylcholine on Diameter–, Modulus–, and Compliance–Midwall Stress Curves
Fig 5⇓ (panels a and b) shows the diameter–and elastic modulus–midwall stress curves at baseline and during acetylcholine infusion. After acetylcholine there was an increase in diameter at each level of midwall stress (P<.01) and a downward shift of the modulus–midwall stress curve toward the lower values of modulus, reflecting the decrease in arterial wall stiffness at every level of midwall stress (P<.01).
Fig 5c⇑ shows the compliance–midwall stress curves at baseline and during acetylcholine infusion. After acetylcholine, there was an upward shift of the compliance–midwall stress curve toward higher values of compliance (P<.01) reflecting, at the level of the artery chamber, the effect of the increase in artery diameter and the decrease in artery wall stiffness at every level of stress.
Effects of Arterial Blood Flow on Mechanical Changes Induced by L-NMMA and Acetylcholine
When the decrease in arterial flow during L-NMMA was taken as a covariate, the decrease in the elastic incremental modulus of the arterial wall (1.58±0.05 versus 1.44±0.05 kPa · 103; P<.025) and the increase in arterial compliance (3.99±0.07 versus 4.48±0.07 m2 · kPa · 10−8; P<.0001) were still significant. These results demonstrate that the isometric relaxation observed could not be fully explained by the decrease in arterial flow.
In addition, when the increase in arterial flow during acetycholine was taken as a covariate, the decrease in the elastic incremental modulus of the arterial wall (1.14±0.02 versus 0.57±0.02 kPa · 103; P<.0001), the increase in radial artery internal diameter (2.92±0.00 versus 3.10±0.00 m · 10−3; P<.0001) and the increase in arterial compliance (2.74±0.21 versus 9.21±0.21 m2 · kPa · 10−8; P<.0001) were still significant. These results demonstrate that the decrease in arterial tone and thus the vasodilatation observed after acetylcholine could not be fully explained by the increase in arterial flow but could be due in part to a direct flow-independent effect.
The main result of the present study performed in the human radial artery is that L-NMMA induced a decrease in arterial wall stiffness and an increase in arterial compliance with no change in diameter, wall thickness, or midwall stress. This demonstrates that inhibition of basal release of NO is associated with a paradoxical isometric smooth muscle relaxation of the arterial wall. In contrast, stimulated release of NO by acetylcholine is associated with the expected decrease in artery wall stiffness. These results demonstrate that NO is involved in the regulation of the mechanical properties of the conductance arteries and suggest that the inhibition of NO synthesis is associated with the development of compensating vasodilator mechanisms.
Using the high resolution echo-tracking method, it was possible to obtain continuous and stable measurements of internal diameter and wall thickness. This was illustrated by the stable value of the calculated cross-sectional area of the arterial wall during control and infusion periods. Moreover, the high reproducibility of the measurements obtained with this method was previously demonstrated in our laboratory3,8,9 and was shown to be similar to that reported by others.14
In addition, the Doppler system enabled us to obtain simultaneous and continuous recordings of arterial blood flow at the same site at which mechanical parameters were measured. Thus, it was possible for the first time in humans to study the effects of NO synthesis inhibition on the mechanical properties of the arterial wall of the radial artery and to consider these effects in relation to the simultaneous changes in arterial flow. This method enables us to assess the direct effects of inhibition or stimulation of NO synthesis on the mechanical properties of the conduit arteries independently from those induced by the associated changes in flow.
In agreement with previous results obtained with plethysmography6,15 or with Doppler technique,9 the administration of L-NMMA decreased radial blood flow, demonstrating the existence of a basal release of NO in the forearm vascular bed. Moreover, this decrease in arterial flow after L-NMMA demonstrates that the doses we used were able to significantly affect the release of NO at least at the level of resistance arteries. However, at the level of conduit arteries, L-NMMA, administered at a dose that abolished the radial artery response to acetylcholine9 as well as the flow-dependent vasodilatation of the radial artery,8 did not significantly decrease radial diameter. Such a lack of effect on conduit artery is in agreement with previous results obtained in humans at the level of the pulmonary artery16 and the proximal site of coronary arteries17and stresses the fact that conduit artery can exhibit reactivity to vasoactive substances differently than resistance arteries.
In the present study, the relaxant effect of acetylcholine was directly assessed from the calculated decrease in wall stiffness, ie, the incremental elastic modulus of the arterial wall. This decrease in stiffness occurred despite the opposite effect of the increase in midwall stress, secondary to the increase in diameter at constant pressure.18 Moreover, since compliance is proportional to arterial diameter and inversely proportional to wall stiffness, the increase in arterial compliance after the administration of acetylcholine may be explained both by the increase in arterial diameter and by the decrease in arterial wall stiffness related to the decrease in arterial tone. This is in agreement with results previously reported with acetylcholine19 or with exogenous NO donors,20 which increase both arterial diameter and distensibility.
It must be noted that the NO-mediated relaxing effect of acetylcholine in vivo may occur through two distinct mechanisms, ie, direct effects and indirect flow-mediated vasorelaxation. However, the fact that the decrease in wall stiffness and the increase in compliance remained significant even when the changes in blood flow were taken into account suggests that at least part of the relaxing effect is flow-independent, confirming in humans the data previously obtained in animal models.21
The most surprising finding of our study was the demonstration that acute inhibition of NO synthesis was associated with a dose-dependent decrease in arterial wall stiffness of the radial artery, as evidenced by the decrease in incremental elastic modulus of the arterial wall. This occurred in the absence of changes in mean midwall stress, ie, in constant wall loading conditions, and was confirmed at each level of stress by the downward shift of the elastic modulus–midwall stress curve. In addition, this decrease in stiffness was observed without a change in mean arterial diameter and wall thickness, and thus represents a paradoxical isometric smooth muscle relaxation of the arterial wall after inhibition of NO synthesis. This decrease in wall stiffness induced an isometric increase in arterial compliance. A similar increase in arterial compliance was noted after endothelium removal at the level of the carotid artery in the rat.22 This change was explained by the increase in arterial diameter after endothelium removal.23 However, our results cannot be fully superposed with these previously described data, since endothelium removal not only suppresses NO-dependent tone but also other pathways involved in the regulation of vascular tone, and may markedly affect the balance between vasoconstrictor and vasodilator tone. In addition, our results on compliance cannot be explained by changes in arterial diameter since diameter did not change after L-NMMA. Our results could also suggest that the arterial diameter or tone of large arteries in humans may not be directly dependent on NO synthesis or that the basal release of NO may be absent or minimal at this level. However, the fact that the vasodilatory response to nitroprusside was increased in this vascular bed after L-NMMA9 suggests the presence of a basal release of NO, since potentiation of the response to exogenous NO after L-NMMA has been previously considered to be an index of basal NO release.24–26 Finally, our results are consistent with recent in vitro experiments, which demonstrated a rightward shift of the stress-strain relationship of the isolated rat coronary artery after incubation with L-NAME, reflecting a paradoxical decrease in the arterial wall stiffness after NO-synthase inhibition. These effects were partially reversed by the administration of l-arginine and were associated with a downward shift of the incremental elastic modulus–stress curve.27
Given the marked role of NO as an inhibitor of smooth muscle contraction, it would be expected that inhibition of NO release should be associated with an increase in the effects of vasoconstrictor influences (such as angiotensin II, catecholamines, or endothelin) and thus with an increased arterial isometric tone. One possibile explanation of these paradoxical effects would be that L-NMMA acts as a partial agonist on NO synthase. This is unlikely given the marked decrease in blood flow observed after L-NMMA. One more likely explanation is that the paradoxical decrease in arterial tone observed in our experiments is the consequence of compensating vasodilatory mechanisms occurring after the inhibition of NO synthesis. Such compensatory mechanisms could also explain why L-NMMA did not affect arterial diameter in our experiments or in previously published studies.16,17 Indeed, in the coronary circulation of dogs, chronic treatment with L-NMMA is associated with an increase in the basal release of vasodilator prostanoids. However, there was no evidence to support the existence of such a mechanism after acute NO-synthesis inhibition.28 In contrast, in isolated perfused rabbit carotid and porcine epicardial coronary arteries, acute administration of NO attenuates the release of EDHF.29 Therefore, these results suggest that the decrease in NO resulting from an acute NO-synthesis inhibition may be associated with a compensating increase in the release of EDHF. However, the role of EDHF cannot be easily assessed in humans in vivo, and thus whether compensatory release of EDHF may explain the decrease in wall stiffness after L-NMMA is not known and cannot be answered from the present study.
Finally, we evaluated the potential role of the modifications of flow in the L-NMMA–induced changes in the mechanical properties of the artery wall. Since we previously demonstrated that flow-dependent dilatation was converted to a vasoconstriction after the administration of L-NMMA8 (possibly mediated by an endothelium-dependent vasoconstrictor substance), it is possible that the observed relaxation is the consequence of the decrease in this flow-dependent vasoconstriction, secondary to the decrease in flow. However, the analysis of the effect of L-NMMA repeated with both midwall stress and arterial flow as cofactors demonstrated that inhibition of NO synthesis decreases isometric smooth muscle tone independently of the changes in flow. This lack of role of flow is also supported by the decrease in arterial wall stiffness observed after L-NAME in isolated, pressurized arteries in the absence of flow.27
The present investigation demonstrates that NO is involved in the regulation of the mechanical properties of a peripheral conduit artery and suggests that the inhibition of NO synthesis is associated at this level with the development of compensating vasodilator mechanisms. The effectiveness of this compensating mechanism in pathological states in which NO availability is decreased remains to be assessed.
Selected Abbreviations and Acronyms
|EDHF||=||endothelium-derived hyperpolarizing factor|
These studies were supported by the Swiss National Science Foundation (grant 32 to 49825.96), the Freiwillige Akademische Gesellschaft Basel, Switzerland, and the Association Charles Nicolle, Rouen, France.
Reprint requests to C. Thuillez, MD, PhD, Service de Pharmacologie, Hôpital de Bois-Guillaume, CHU de Rouen, 76031 Rouen Cedex, France.
- Received May 9, 1997.
- Revision received June 12, 1997.
- Accepted July 7, 1997.
Dobrin PB, Roveck AA. Influence of vascular smooth muscle on contractile mechanics and elasticity of arteries. Am J Physiol. 1969;217:1644–1651.
McDonald DA. The elastic properties of the arterial wall. In: London, UK: Edward Arnold. Blood Flow in Arteries. 1974:238–282.
Joannides R, Richard V, Moore N, Godin M, Thuillez C. Influence of sympathetic tone on mechanical properties of muscular arteries in humans. Am J Physiol. 1995;268:H794–H801.
Rees DD, Palmer RMJ, Moncada S. The role of endothelium-derived nitric oxide in the regulation of blood pressure. Proc Natl Acad Sci U S A. 1989;86:3375–3378.
Chu A, Chambers DE, Lin CC, Kuehl WD, Palmer RMJ, Moncada S, Cobb FR. Effects of inhibition of nitric oxide formation on basal vasomotion and endothelium-dependent responses of the coronary arteries in awake dogs. J Clin Invest. 1991;87:1964–1968.
Joannides R, Haefeli WE, Linder L, Richard V, Bakkali EH, Thuillez C, Lüscher TF. Nitric oxide is responsible for flow-dependent dilation of human peripheral conduit arteries in vivo. Circulation. 1995;91:1314–1319.
Joannides R, Richard V, Haefeli WE, Linder L, Lüscher TF, Thuillez C. Role of basal and stimulated release of nitric oxide in the regulation of radial artery caliber in Humans. Hypertension. 1995;26:327–331.
Linder L, Kiowski W, Bühler FR, Lüscher TF. Indirect evidence for release of endothelium-derived relaxing factor in human forearm circulation in vivo: blunted response in essential hypertension. Circulation. 1990;81:1762–1767.
Tardy Y, Hayoz D, Mignot JP, Richard PH, Brunner HR, Meister JJ. Dynamic non-invasive measurements of arterial diameter and wall thickness. J Hypertens. 1992;10(suppl 6):S105–S109.
Meister JJ, Tardy Y, Stergiopulos N, Hayoz D, Brunner HR, Etienne JD. Non-invasive method for the assessment of non-linear elastic properties and stress of forearm arteries in vivo. J Hypertens. 1992;10(suppl 6):S23–S26.
Kiowski W, Linder L, Stoschitzky K, Pfisterer M, Burckhardt D, Burkart F, Bühler FR. Diminished vascular response to inhibition of endothelium-derived nitric oxide and enhanced vasoconstriction to exogenously administered endothelin-1 in clinically healthy smokers. Circulation. 1994;90:27–34.
Celermajer DS, Dollery C, Burch M, Deanfield JE. Role of endothelium in the maintenance of low pulmonary vascular tone in normal children. Circulation. 1994;89:2041–2044.
Leffroy DC, Crake T, Uren NG, Davies GJ, Maseri A. Effect of inhibition of nitric oxide synthesis on epicardial coronary artery caliber and coronary blood flow in humans. Circulation. 1993;88:43–54.
Bader H. Dependence of wall stress in the human thoracic aorta on age and pressure. Circ Res. 1967;20:354–361.
Ramsay MW, Goodfellow J, Jones CJH, Luddington LA, Lewis MJ, Henderson AH. Endothelial control of arterial distensibility is impaired in chronic heart failure. Circulation. 1995;92:3212–3219.
Young MA, Vatner SF. Blood flow- and endothelium-mediated vasomotion of iliac arteries in conscious dogs. Circ Res. 1987;61(suppl II):II-88–II-93.
Caputo L, Tedgui A, Levy B. Control of carotid vasomotor tone by local renin-angiotensin system in normotensive and spontaneously hypertensive rats. Role of endothelium and flow. Circ Res. 1995;77:303–309.
Mourlon-Le Grand MC, Benessiano J, Levy BI. cGMP pathway and mechanical properties of carotid artery wall in WKY rats and SHR: role of endothelium. Am J Physiol. 1992;263(Heart Circ Physiol.32):H61–H67.
Moncada S, Rees DD, Schulz R, Palmer RMJ. Development and mechanism of a specific supersensitivity to nitrovasodilators after inhibition of vascular nitric oxide synthesis in vivo. Proc Natl Acad Sci U S A. 1991;88:2166–2170.
Lüscher TF, Richard V, Yang Z. Interaction between endothelium-derived nitric oxide and SIN-1 in human and porcine coronary arteries. J Cardiovasc Pharmacol. 1989;14(suppl 11):S76–S80.
Puybasset L, Bea ML, Ghaleh B, Giudicelli JF, Berdeaux A. Coronary and systemic hemodynamic effects of sustained inhibition of nitric oxide synthesis in conscious dogs - evidence for cross talk between nitric oxide and cyclooxygenase in coronary vessels. Circ Res. 1996;79:343–357.
Bauersachs J, Popp R, Hecker M, Sauer E, Fleming I, Busse R. Nitric oxide attenuates the release of endothelium-derived hyperpolarizing factor. Circulation. 1996;64:3341–3347.