Role of NO in Choroidal Blood Flow Regulation during Isometric Exercise in Healthy Humans

Abstract

purpose. Nitric oxide (NO) is an important regulator of basal choroidal blood flow. Animal experiments indicate that NO is also involved in choroidal blood flow regulation during changes in ocular perfusion pressure and inhibition of NO synthase (NOS) has been reported to shift choroidal pressure–flow curves to the right. The hypothesis for the study was that inhibition of NOS may influence choroidal blood flow during isometric exercise.

methods. To test this hypothesis, a randomized, double-masked, placebo-controlled, three-way crossover study was performed in 12 healthy male volunteers. Subjects received on different study days intravenous infusions of N ^G-monomethyl-l-arginine (l-NMMA), phenylephrine, or placebo. During these infusion periods, subjects were asked to squat for 6 minutes. Choroidal blood flow was assessed with laser Doppler flowmetry, and ocular perfusion pressure (OPP) was calculated from mean arterial pressure and intraocular pressure.

results. l-NMMA and phenylephrine increased resting OPP by 10% and 13%, respectively, but only l-NMMA reduced resting choroidal blood flow (−17%, P < 0.001). The relative increase in OPP during isometric exercise was comparable with all drugs administered. Isometric exercise increased choroidal blood flow during administration of placebo and phenylephrine, but not during administration of l-NMMA (P < 0.001 vs. placebo).

conclusions. These data indicate that NO plays an important role in the regulation of choroidal blood flow during isometric exercise.

Nitric oxide (NO) is a potent vasodilator that plays a role in the maintenance of vascular tone ¹ and regulates basal blood pressure. ^{2 3} It also exerts a wide spectrum of biological actions in the eye including regulation of retinal, optic nerve, and choroidal blood flow (CBF). ⁴ In the choroid, several studies have shown that NO maintains basal choroidal vascular tone in rats, ^{5 6 7} rabbits, ⁸ cats, ⁹ dogs, ¹⁰ and humans. ^{11 12}  

Recent studies have also suggested that NO plays a role in choroidal autoregulation in rabbits ⁸ and pigs. ^{13 14} This notion is controversial, however, because a variety of animal studies have failed to detect choroidal autoregulation. ^{15 16 17 18} Potential differences between the results of these previous studies have been discussed by Kiel. ¹⁹ In humans, the investigation of choroidal autoregulation is difficult, because changes in ocular perfusion pressure cannot be induced without changing neural input or vascular tone. In principle ocular perfusion pressure may be altered by isometric exercise, by pharmacologic stimulation, or by changes in intraocular pressure (IOP). 

We hypothesized that inhibition of NO synthase (NOS) with N ^G-monomethyl-l-arginine (l-NMMA) may alter the choroidal pressure–flow relationship in response to isometric exercise in healthy humans. A problem that arises during such experiments is that systemic administration of l-NMMA increases systemic blood pressure. Hence, the pressure–flow relationship during isometric exercise obtained during infusion of saline or l-NMMA are difficult to compare, because they start at different ocular perfusion pressures (OPPs). The present study was conducted to overcome this problem by using the α-adrenoceptor agonist phenylephrine (PE) in a dose assumed to induce comparable effects on blood pressure in comparison to l-NMMA. 

Methods

Subjects

The present study was performed in compliance with the Declaration of Helsinki and the Good Clinical Practice (GCP) guidelines. After approval of the study protocol by the ethics committee of the Vienna University School of Medicine and after written informed consent was obtained, 12 healthy nonsmoking male subjects were studied (age: 28 ± 4 years, mean ± SD). All subjects were drug free for at least 3 weeks before inclusion and passed a prestudy screening during the 4 weeks before the first study day that included medical history and physical examination, 12-lead electrocardiogram, complete blood count, activated partial thromboplastin time, thrombin time, clinical chemistry (sodium, potassium, creatinine, uric acid, glucose, cholesterol, triglycerides, alanine aminotransferase, aspartate transcarbamylase, γ-glutamyltransferase, alkaline phosphatase, total bilirubin, and total protein), hepatitis-A, -B, and -C and HIV serology, urinanalysis and an ophthalmic examination. Subjects were excluded if any abnormality was found as part of the pretreatment screening, unless the investigators considered an abnormality to be clinically irrelevant. In addition, subjects with ametropia of more than 3 D, anisometropia more than 1 D, or any evidence of eye disease that might have interfered with the purpose of the present trial were excluded. During the last week after completion of the study, a follow-up safety investigation was scheduled for all participating subjects. This follow-up investigation included complete blood count, activated partial thromboplastin time, thrombin time, clinical chemistry (sodium, potassium, creatinine, uric acid, glucose, cholesterol, triglycerides, alanine aminotransferase, aspartate transcarbamylase, γ-glutamyltransferase, alkaline phosphatase, total bilirubin, and total protein), and urinanalysis. All subjects had several training sessions with the laser Doppler flowmetry (LDF) system before inclusion in the study. In addition, all subjects participated in periods of squatting, during which LDF measurements were performed. Only subjects with excellent target fixation and highly reproducible blood flow results during isometric exercise were considered for inclusion in the study. 

Study Design

Subjects were asked to abstain from alcohol and caffeine for at least 12 hours before trial days and were studied after an overnight fast. The study had a randomized, double-masked, placebo-controlled, three-way crossover design with l-NMMA (Clinalfa, Läufelfingen, Switzerland; dose: bolus 6 mg/kg over 5 minutes followed by a continuous infusion of 60 μg/kg per minute over 15 minutes), PE (Neosynephrine, Winthrop Breon Laboratories, New York, NY; dose: 1 μg/kg per minute, infusion period 20 minutes), or placebo (physiologic saline solution). To maintain double-masked conditions, two syringes that were identical in appearance to the l-NMMA syringes were prepared on the PE and the placebo days. Accordingly, the two syringes on the PE day had similar drug concentrations, whereas the l-NMMA syringes had different concentrations (bolus and continuous infusion). All drugs were administered intravenously into an antecubital vein, with automated devices to ensure constant infusion rates. The dosages of l-NMMA and PE were based on previous clinical trials in our laboratory. ^{12 20}  

Description of Study Days

After a resting period of at least 20 minutes, baseline measurements of ocular and systemic hemodynamics were performed. CBF was measured continuously by LDF for 3 minutes at baseline. Thereafter, subjects performed squatting for 6 minutes, and CBF was measured continuously by LDF. Squatting was performed in a position in which the upper and the lower leg was as close as possible to a right angle. For the subjects’ security a nurse stood behind the subject during the squatting periods. Systemic hemodynamics were assessed every minute and IOP at baseline and at the end of the squatting period. Thereafter, a rest of at least 30 minutes was scheduled. In some patients the rest was extended up to 45 minutes, because subjects still felt exhausted after 30 minutes. When systemic hemodynamics had returned to baseline, drug administration was started. Fourteen minutes after the start of drug administration another squatting period was scheduled for all subjects. This second period of isometric exercise was identical with the one before administration of the drug. 

Systemic Hemodynamics

Systolic (SBP) and diastolic (DBP) blood pressure, and mean arterial pressure (MAP) were measured on the upper arm by an automated oscillometric device. Pulse rate (PR) was automatically recorded from a finger pulse oximeter (HP-CMS patient monitor; Hewlett Packard, Palo Alto, CA). 

Applanation Tonometry and Ocular Perfusion Pressure

The intraocular pressure (IOP) was measured with a Perkins applanation tonometer (Clement Clarke, Edinburgh, UK). Oxybuprocaine hydrochloride was used to anesthetize the cornea. Ocular perfusion pressure was calculated as OPP = 2/3 × MAP − IOP. ²¹ This formula is based on the evidence that the pressure in choroidal veins almost equals the IOP. ²² During isometric exercise, we observed only small changes of IOP over baseline after 6 minutes of squatting. Hence, we used a linear regression model to extrapolate the IOPs at the other time points of squatting. 

Laser Doppler Flowmetry

Continuous measurement of CBF was performed with LDF. ²³ With this technique, the vascularized tissue is illuminated by coherent laser light. Light scattered by the moving red blood cells undergoes a frequency shift. In contrast, static tissue light-scattering cells do not change the light frequency, but lead to randomization of light directions, impinging on red blood cells. Hence, red blood cells receive light from numerous random directions. Because the frequency shift is dependent, not only on the velocity of the moving red blood cells, but also on the angle between the wave vectors of the incident and the scattered light, scattering of the light in tissue broadens the Doppler shift power spectrum. From this spectrum the average velocity of red blood cells (Vel), the volume of red blood cells (Vol), and the CBF (Vel × Vol) can be determined based on a theory of light scattering in tissue in relative units. ²⁴  

In the present study, a compact laser Doppler flowmeter, which has been described in detail previously, was used for the measurements of the CBF. ^{25 26} The laser beam of a single-mode laser diode (785 nm, 90 μW at the cornea) is delivered to the eye by a confocal optical system. The beam diameter at the fundus is nominally 12 μm. The light is collected by a bundle of six fibers with a core diameter of 110 μm that are arranged on a circle with a diameter of 180 μm. All measurements were performed in the fovea by asking the subject to directly fixate the beam, which appeared as a small red dot. The fovea was chosen, because the retina is avascular in this region. For statistical analysis, only the portions of the signal that were within 15% of the baseline direct current (DC) were taken for analysis. 

Compared with previous fundus camera–based systems ²³ for the assessment of CBF, the new system offers two major advantages: Adjustment of the detector relative to the measurement on the retina is omitted, because the system uses confocal optics, and the system is portable, which facilitates measurements during isometric exercises. 

Data Analysis

All statistical analyses were performed on computer. (Statistica software ver. 4.5; StatSoft, Inc., Tulsa, OK). All outcome variables were calculated for each subject individually and then averaged. The effect of exercise on the outcome parameters was assessed with repeated measures ANOVA versus baseline. In the repeated-measures ANOVA model, the significance of the time effect was used for this calculation. The effect of l-NMMA and PE on baseline parameters was characterized with two-way ANOVA. The relative change in hemodynamic parameters induced by isometric exercise was calculated. For the experiments during l-NMMA, PE, or placebo, the value immediately before the start of isometric exercise was taken as baseline. The effects of the administered drugs on exercise-induced changes in CBF were also calculated with a repeated-measures ANOVA. The interaction between time and treatment was taken as the level of significance to characterize effects of l-NMMA or PE on exercise-induced changes in CBF versus placebo. To gain information on the pressure–flow relationship, relative data were sorted according to ascending OPPs. ^{27 28} For each squatting period, we obtained a total of 72 OPP and CBF readings. These were divided into eight groups of nine OPP and CBF values. Hence, the first group consisted of the data with the lowest relative OPPs (n = 9) and the eighth group consisted of the data with the highest relative OPPs (n = 9). In contrast to the ANOVA, with this technique, subjects do not serve as their own control. In other words, it is possible with this technique to compare data from different subjects within each group. The mean values from these groups were used to determine the OPP at which the CBF significantly deviated from baseline. This was the case when the 95% confidence interval no longer intersected the baseline level. Data are presented as the mean ± SEM. A two-tailed P < 0.05 was considered the level of significance. 

Results

No adverse events were observed during the study. Compared with the prestudy screening, none of the subjects had any relevant changes in laboratory parameters at the follow-up investigation. 

There were no significant differences between the baseline values on the three trial days (Figs. 1 2 3 4 5) . As expected, isometric exercise induced a significant increase in MAP and PR during all pretreatment squatting periods (P < 0.001 vs. baseline, Figs. 1 and 2 ). This effect on systemic hemodynamics was comparable on all study days. IOP was unchanged during isometric exercise (11 ± 1 mm Hg at baseline on all study days; between 10 ± 2 and 12 ± 1 mm Hg at the end of the squatting period). Consequently, OPP was primarily influenced by changes in systemic blood pressure and showed a significant increase during all squatting periods. The maximum increase in OPP was between 56% and 62% during the squatting periods (Fig. 3) . This increase in OPP was paralleled by an increase in CBF, which was less pronounced, however, than the increase in OPP. Nevertheless, the increase in CBF, which was at its maximum between 11% and 15%, was significant versus baseline (P = 0.026 vs. baseline, Figs. 4 and 5 ). This increase in CBF was mainly caused by an increase in Vel (between 10% and 16%; P = 0.033 vs. baseline), whereas Vol remained almost unchanged during isometric exercise (between –1% and 3%; P = 0.62 vs. baseline). 

Placebo had no consistent effect on ocular or systemic baseline parameters. By contrast, PE induced a significant increase in MAP (+10.3% ± 2.2%, P < 0.001 vs. placebo) and decreased PR (−8.3% ± 5.0%, P = 0.017 vs. placebo). The systemic hemodynamic effects of l-NMMA were comparable to those of PE and were significant versus placebo. l-NMMA increased baseline MAP by 12.6% ± 1.7% (P < 0.001 vs. placebo) and decreased PR by 11.8% ± 3.4% (P < 0.001 vs. placebo). The effects of PE (+1.3% ± 4.0%) and l-NMMA (−5.2% ± 5.5%) on IOP were slight and not significant. As a consequence, baseline OPP increased parallel to MAP with both administered vasoconstrictors. Effects of PE (+12.3% ± 2.5%) and l-NMMA (+16.8% ± 1.9%) on baseline OPP were highly significant (each P < 0.001 vs. placebo) and comparable. PE had no effect on basal CBF (−3.5% ± 2.5%, Fig. 4 ). By contrast, l-NMMA decreased resting CBF (−16.6% ± 3.6%, Fig. 4 ), which was highly significant (P < 0.001 vs. placebo). This effect was mainly attributable to a decrease in resting Vel (−13.4% ± 3.2%; P = 0.012 vs. placebo), whereas only a tendency was shown toward reduced Vol (−5.6% ± 4.3%; P = 0.13 vs. placebo). 

The effect of squatting on OPP and CBF was not altered by infusion of placebo (Figs. 3 4 5) . PE and l-NMMA induced a significant increase in basal MAP and OPP, but did not alter the relative increase in MAP or OPP to isometric exercise. PE did not alter the response of CBF to squatting. By contrast, l-NMMA significantly altered the isometric exercise-induced changes in CBF. l-NMMA not only reduced basal CBF, but also attenuated the exercise-induced increase in CBF (Fig. 3 , P = 0.004 vs. placebo). Whereas CBF increased by a maximum of 13% and 12% during placebo and PE, respectively, the increase of CBF during squatting after administration of l-NMMA was only 5%. Again, these effects were mainly attributable to changes in Vel. During isometric exercise Vel increased by 12%, 13%, and 3% during infusion of placebo, PE and l-NMMA, respectively. The exercise-induced increase in Vel during infusion of l-NMMA was less pronounced than during placebo (P = 0.003). By contrast, isometric exercise induced only minor changes in Vol during all drug infusion periods (between 2% and 3%). 

The pressure–flow relationship during the squatting periods is presented in Figure 6 . Under basal conditions, CBF began to increase at OPPs between 48% and 55% above baseline. During placebo infusion, the pressure–flow curve was not altered, and CBF began to increase at an OPP that was 44% above baseline. The pressure–flow relationship was also unaltered when PE was administered. CBF increased at OPPs more than 55% above baseline. During infusion of l-NMMA the pressure–flow relationship was shifted to the right. In contrast to the other squatting periods, CBF did not increase, even at the highest OPPs (P < 0.001). 

Discussion

The present study clearly indicates that inhibition of NOS alters the response of CBF to isometric exercise. Resting CBF was significantly reduced after administration of l-NMMA. Considering that infusion of l-NMMA slightly increased MAP, this indicates an increase in choroidal vascular resistance, in keeping with our previous trials in healthy subjects ^{11 12} and supporting the concept that NO plays a major role in the maintenance of basal CBF. In extending these findings, the present trial indicates that inhibition of NOS significantly alters the pressure–flow relationship during isometric exercise, which is compatible with previous findings that inhibition of NOS induces a rightward shift of the pressure–flow curve during mechanical changes of MAP in the rabbit. ⁸ In contrast to placebo or equipotent doses of PE, l-NMMA did not increase CBF during isometric exercise, despite an increase in OPP of more than 70%. 

Our results confirm those in previous studies showing that the choroid is capable of maintaining its perfusion level over a wide range of OPPs during squatting. This was first reported by Riva et al., ²⁷ who showed that, during isometric exercise, a 67% increase in OPP resulted in a 12% increase in CBF only. Although they used a squatting period of 90 seconds only, the data are in good agreement with the results of the present trial. Recently, we have shown in a small number of subjects (n = 6), by using the same experimental paradigm as in the present study, that squatting significantly increases CBF by 16%. ²⁸ This is again in good agreement with the results of the present study and indicates that isometric exercise is a sufficiently reproducible stimulus for the investigation of CBF regulation in humans. There is, however, some interindividual variability of the pressure–flow relationships, as indicated by the error bars in Figure 6 . This may be related to interindividual differences in the regulatory capacity of the choroid. More important, the time course and the magnitude of the systemic hypertensive response during squatting varied considerably between subjects. This may be related to the training status of the subjects, but also to the willingness of a subject to cooperate with the study personnel. We chose a crossover design rather than a parallel group design, because interindividual variability is less of a problem with this approach. 

Care must be taken in interpreting the results of the present study in relation to choroidal autoregulation. In its strictest sense, autoregulation refers to changes in vascular resistance during changes in perfusion pressure in isolated organs. Isometric exercise, however, stimulates the adrenergic system and may alter the neural input to the choroid. Several animal experiments indicate that ocular sympathetic vasomotor nerves may play a role in choroidal vasoconstriction in response to increased ocular perfusion pressure. ^{29 30 31} Moreover, it has been shown that ganglionic blockade with hexamethonium alters the pressure–flow in the choroid of the rabbit. ⁸ Nevertheless, the present study indicates that the choroid has some ability to regulate its blood flow during isometric exercise. Whether the choroid autoregulates in its strict sense should be studied further in animal experiments to clarify the discrepancies in findings in previously published studies. ^{8 13 15 16 17 18 19 27 28 32 33 34}  

The present results support the following hypothesis. During isometric exercise, a number of vasoactive substances are released inducing net vasoconstriction to keep CBF constant at increased OPP. When production of NO is reduced during infusion of l-NMMA, the potent vasodilator NO cannot counteract vasoconstriction in response to the increase in OPP. Consequently, the choroidal vasculature is capable of increasing vascular resistance to a higher degree than during physiological conditions. This is also compatible with the observation that excess production of NO narrows the autoregulatory plateau in the newborn piglet. ^{13 14} Recently, Koss ³⁴ reported, however, that the NOS inhibitor l-NAME reduces basal anterior segment ocular blood flow in cats, but does not alter autoregulatory responses. Several differences between these animal experiments and the present human study should be emphasized. To change OPP in the cat, the investigator cannulated the anterior chamber and increased IOP in steps of 10 to 15 mm Hg. Obviously, such an approach is not suitable for human experiments. In humans, we decided to use isometric exercise, because it induces a pronounced increase in systemic blood pressure. Direct comparison of our data with the previous feline data are therefore difficult, because we induced an increase in OPP, whereas elevating IOP obviously results in a decrease in OPP. Moreover, Koss ³⁴ measured anterior CBF, whereas we assessed submacular CBF. There are currently few data on possible local control mechanisms of CBF. In addition, possible species differences have to be taken into account. Finally, it must be considered that in all currently available animal experiments, CBF was studied under general anesthesia, which may itself influence the l-arginine–NO pathway. The absence of NO in mediating autoregulation was also reported in the retina. ³⁵ Again, a decrease in OPP was used, and the differences in regulatory mechanisms between the retina and the choroid hamper direct comparison with the present data. 

The results of the present trial do not indicate the source of NO responsible for vasodilation in the human choroid. l-NMMA is not specific for the isoforms of NOS and therefore NO may either arise from neural (NOS1) and/or endothelial (NOS3) sources. NOS3 is present in the endothelium of choroidal blood vessels. ³⁶ A functional role for NO derived from nerves was proposed based on experiments in isolated external and internal canine ophthalmic arteries and porcine and bovine posterior ciliary arteries ^{37 38 39} and several animal experiments. Stimulation of the facial nerve in rabbits caused a pronounced increase in blood flow in all parts of the uvea, which was significantly attenuated by pretreatment with a NOS inhibitor. ⁴⁰ In the rat parasympathetic stimulation induced by electrical stimulation of the superior salivatory nucleus increased blood flow in the choroid, an effect that was partially blocked by 1-(2-trifluoromethyphenyl)imidazole, a relatively selective inhibitor of NOS1. ⁴¹ A definitive answer regarding the source of NO controlling CBF at rest and during isometric exercise in humans can be found only by using selective inhibitors of NO synthase in clinical trials. 

A limitation of the present study is that the pressure–flow relationships presented in Figure 6 are not directly comparable between the placebo-infusion period and the l-NMMA–infusion period, because the NOS-inhibitor increased systemic blood pressure and the pressure–flow relationships referred to baseline OPPs that are not comparable. In contrast, PE did not alter the response of CBF in response to an increase in OPP, although the systemic hypertensive response was comparable to that observed with l-NMMA. PE did not alter basal CBF in face of the increase in OPP. This indicates an increase in choroidal vascular resistance during PE administration, which may result either from direct vasoconstriction due to stimulation of vasoconstrictive α-adrenoceptors or from a potential autoregulatory response of the choroid. 

Several limitations of the LDF technique should be mentioned. Retinal red blood cell flux and not blood flow was measured. The relation between retinal red blood cell flux and retinal blood flow may be altered with changes in local hematocrit. We think it unlikely, however, that local hematocrit changed in the present experiments. This is supported by the observation that almost all changes in CBF in the present study were caused by changes in Vel, whereas Vol showed only minor changes. In addition, the present results apply only to submacular CBF, and little is known about potential local variations in regulation of CBF. Finally, the sampling depth of choroidal LDF is unclear. Thus, we could not distinguish which vessels of the choriocapillaris contributed most to the LDF signal. 

In conclusion, the present study indicates that locally produced NO plays an important role in regulation of CBF during isometric exercise. The relevance of this finding for regulation of CBF in systemic diseases associated with impaired formation of NO and ocular manifestations, such as diabetes or systemic hypertension, remains to be established. 

 Supported by a Grant P-14219 from the Austrian Science Foundation.

 Submitted for publication February 19, 2002; revised June 12 and August 30, 2002; accepted September 20, 2002.

 Commercial relationships policy: N.

 The publication costs of this article were defrayed in part by page charge payment. This article must therefore be marked “advertisement” in accordance with 18 U.S.C. §1734 solely to indicate this fact.

 

Figure 1.

 

The effect on MAP of squatting. The first period of squatting was done without drug (pretreatment). The second squatting period was performed during administration of placebo (no symbols), l-NMMA (solid up triangles) or PE (open down triangles). Data are presented as the mean ± SEM (n = 12). #Significant changes versus baseline; §significant effects of PE or l-NMMA on baseline parameters versus placebo.

Figure 2.

 

The effect of squatting on pulse rate (PR). The description of the experiment, symbols, and data are as in Figure 1 .

Figure 3.

 

The effect of squatting on OPP. The description of the experiment, symbols, and data are as in Figure 1 .

Figure 4.

 

The effect of isometric exercise on CBF. The description of the experiment, symbols, and data are the same as in Figure 1 , with the exception that the asterisk indicates significant effects of l-NMMA on exercise-induced changes in CBF versus placebo.

Figure 5.

 

Relative change of CBF over the preexercise value during squatting. Description of the experiment, symbols, and data are the same as in Figure 1 , with the exception the asterisk indicates significant effects of l-NMMA on exercise-induced changes in CBF versus placebo.

Figure 6.

 

Pressure–flow relationship using the categorized OPP and CBF during isometric exercise. Relative data were sorted into groups of nine values each, according to ascending OPPs. The first period of squatting was performed without drug administration (baseline; open down triangles). The second squatting period was performed during administration of placebo, l-NMMA, or PE (solid up triangles). The means and the lower limits of the 95% confidence intervals are shown (n = 12). The dotted line indicates 100% of baseline.