Effects of unoprostone or placebo administration on mean brachial artery blood pressure (A), pulse rate (B), and intraocular pressure (C) in the presence of exogenous endothelin 1 (2.5 ng/kg per minute) in 24 patients. Asterisk indicates significant treatment effects vs placebo, as calculated using repeated-measures analysis of variance. Error bars represent SEM.
Effects of unoprostone or placebo administration on fundus pulsation amplitude (A) and choroidal blood flow(B) in the presence of exogenous endothelin 1 (2.5 ng/kg per minute) in 24 patients. Asterisks indicate significant treatment effects vs placebo, as calculated using repeated-measures analysis of variance. Error bars represent SEM.
Elzbieta Polska, Arno Doelemeyer, Alexandra Luksch, Paulina Ehrlich, Nils Kaehler, Christine L. Percicot, George N. Lambrou, Leopold Schmetterer. Partial Antagonism of Endothelin 1–Induced Vasoconstriction in the Human Choroid by Topical Unoprostone Isopropyl. Arch Ophthalmol. 2002;120(3):348–352. doi:10.1001/archopht.120.3.348
There is increasing evidence that reduced ocular blood flow plays a role in the pathogenesis of glaucoma. In patients with normal-tension glaucoma, ocular blood flow abnormalities may be associated with dysfunction of the endothelin 1 (ET-1) regulation system.
To test the hypothesis that unoprostone, a topical docosanoid, may affect ET-1–induced vasoconstriction in the human choroid.
In a placebo-controlled, randomized, double-masked, 2-way crossover design, ET-1 (2.5 ng/kg per minute for 150 minutes) was administered intravenously to 24 healthy individuals. Thirty minutes after the start of ET-1 infusion, 1 drop of unoprostone or placebo was instilled into the right eye. After another 30 minutes, 2 drops of unoprostone or placebo was topically administered. This procedure was continued and the dose was increased further until 4 drops of unoprostone or placebo was reached. Subfoveal and pulsatile choroidal blood flow were assessed using laser Doppler flowmetry and laser interferometric measurement of fundus pulsation amplitude, respectively.
Administration of exogenous ET-1 decreased choroidal blood flow (mean± SEM, 17% ± 2%; P<.001) and fundus pulsation amplitude (mean ± SEM, 19% ± 2%; P<.001). This effect was significantly blunted when topical unoprostone was coadministered(mean ± SEM decrease in choroidal blood flow, 7% ± 2%; P = .04 vs placebo; mean ± SEM decrease in fundus pulsation amplitude, 12% ± 2%; P<.001 vs placebo).
There is a functional antagonism between ET-1 and topical unoprostone in the choroidal vasculature.
Our findings of a functional antagonism between ET-1 and topical unoprostone in the choroidal vasculature may be important in vascular eye diseases associated with increased ET-1.
SIGNIFICANT EVIDENCE has been accumulated recently indicating that abnormal ocular blood flow plays a role in the pathogenesis of glaucoma.1- 4 Reduced ocular perfusion as observed in patients with primary open-angle glaucoma and normal-tension glaucoma is hypothesized to be related to an altered nitric oxide and endothelin system.5
Endothelin 1 (ET-1) is the most potent vasoconstrictor known, and it affects vascular tone in ocular vessels. Systemic and topical administration of ET-1 in rabbits reduced capillary blood flow in the optic nerve head (ONH) and the choroid.6 Intravitreal injection of ET-1 in rat eyes induced dose-dependent vasoconstriction of major retinal vessels as well as pericyte contraction and significantly prolonged retinal circulation times.7 The ET-1–induced contraction of retinal arteries depends on the influx of extracellular Ca2+ through membrane potential–operated calcium channels, also suggesting that ET-1 may participate in the regulation of retinal artery tone.8
Unoprostone isopropyl is a docosanoid therapeutic agent used for glaucoma and ocular hypertension. Its structure resembles the 22-carbon structure of the naturally occurring metabolites of docosahexaenoic acid (DHA).9,10 Twenty percent to 60% of the phospholipid fatty acid content of the neural tissue is composed of DHA, which makes it the most abundant endogenous polyunsaturated fatty acid in the retina and brain.11,12 Thirty percent to 40% of the fatty acids of rod photoreceptor outer segments of the human retina is DHA, which makes the photoreceptor cells the richest in DHA of all the human cells.13
The synthetic docosanoid unoprostone reduces intraocular pressure (IOP) by increasing the total aqueous outflow.14 Up to now, few investigators have examined the effect of unoprostone on ocular blood flow. Topically administered unoprostone had no effect on the ONH circulation in normal rabbit eyes.15 In humans, choroidal blood flow (ChBF) was significantly increased after topical application of unoprostone, but no changes in ONH circulation were observed.16 Intravitreal injection of unoprostone significantly inhibited an ET-1–induced decrease in ONH blood flow, indicating a functional antagonism between unoprostone and ET-1 in the rabbit eye.15
In the present study, a functional antagonism between ET-1 and unoprostone was investigated in the human choroid. For this purpose, we administered exogenous ET-1 to healthy individuals and investigated whether the vasoconstrictor effect in the human choroid could be altered by topical application of unoprostone. Choroidal blood flow was assessed using 2 techniques: laser Doppler flowmetry and laser interferometric measurement of fundus pulsation amplitude (FPA).
Before the study, a sample size calculation was performed using an α level of .5 and a β level of .2. Accordingly, 24 healthy, nonsmoking men were studied (age range, 21-34 years; mean ± SD age, 26 ± 4 years) after approval from the ethics committee of the Vienna University School of Medicine was obtained. We anticipated a 20% effect of exogenous ET-1 on ChBF based on the results of a previous study.17 The reproducibility of measurements of ChBF using laser Doppler flowmetry in our laboratory is comparable to that previously reported for measurements of ONH blood flow using the same instrument.18 The reproducibility of laser interferometric measurement of FPA is higher19 and therefore is not critical. The sample size was calculated to allow for detecting changes in ChBF of 8%. This means that the present study had a power of detecting a 40% reduction in ET-1–induced changes in ChBF.
The nature of the study was explained, and all participants gave written consent to participate. All volunteers passed a prestudy screening during the 4 weeks before the first study day, which included a physical examination and medical history; a 12-lead electrocardiogram; a complete blood cell count; measurement of activated partial thromboplastin time, thrombin time, and fibrinogen levels; a clinical plasma histochemical analysis (measurement of sodium, potassium, creatinine, uric acid, glucose, cholesterol, triglyceride, alanine aminotransferase, aspartate aminotransferase, γ-glutamyltransferase, alkaline phosphatase, total bilirubin, and total protein levels); hepatitis A, B, and C testing; human immunodeficiency virus serologic analysis; and urine analysis. Furthermore, an ophthalmic examination, including slitlamp biomicroscopy and indirect funduscopy, was performed in each participant before the first study day. Inclusion criteria were normal findings in the screening examinations and ametropia of less than 3 diopters. Within 1 week of completion of the study, a follow-up safety investigation was scheduled, which included a physical examination; a complete blood cell count; measurement of activated partial thromboplastin time, thrombin time, and fibrinogen concentration; measurement of sodium, potassium, creatinine, uric acid, glucose, cholesterol, triglyceride, alanine aminotransferase, aspartate amino transferase, γ-glutamyltransferase, alkaline phosphatase, total bilirubin, and total protein levels; and a urine analysis.
The study was performed in a randomized, placebo-controlled, double-masked, 2-way crossover design. All participants were asked to refrain from alcohol and caffeine intake for at least 12 hours before trial days. After a 20-minute resting period in a sitting position, patients underwent baseline measurement of ChBF using laser Doppler flowmetry, FPA in the macula using laser interferometry, IOP (applanation tonometry), blood pressure, and pulse rate.
Thereafter, participants received a continuous intravenous infusion of ET-1 (Clinalfa AG, Läufelfingen, Switzerland) in a dose of 2.5 ng/kg per minute for 150 minutes. Thirty minutes after initiating ET-1 administration, 1 drop (35 µL using a micropipette) of unoprostone isopropyl (0.12%)(Rescula; CIBA-Vision, Basel, Switzerland) or placebo (physiologic saline solution) was instilled into the right eye. After another 30 minutes, 2 drops of unoprostone or placebo was topically administered. The dose was further increased every 30 minutes until 4 drops of unoprostone or placebo was instilled.
All hemodynamic measurements were performed in 30-minute intervals after the start of ET-1 infusion. Blood pressure was measured at 5-minute intervals during the entire study. Pulse rate and real-time electrocardiograms were monitored continuously. Participants were monitored throughout infusion until variables returned to baseline values. All measurements were performed with the pupil dilated using tropicamide (Mydriaticum; Agepha, Vienna). Two trial days were scheduled for each participant. The washout period between study days was at least 2 days.
Mean brachial artery blood pressure was monitored on the upper arm using an automated oscillometric device. Pulse rate was automatically recorded from a finger pulse-oxymetric device (HP-CMS patient monitor; Hewlett Packard, Palo Alto, Calif).
A slitlamp-mounted Goldmann applanation tonometer was used to measure IOP. Before each measurement, 1 drop of 0.4% benoxinate hydrochloride combined with 0.25% sodium fluorescein was used for local anesthesia of the cornea.
Using this technique,20 the vascularized tissue is illuminated by coherent laser light, avoiding visible vessels in directing the laser beam. Scattering on moving red blood cells leads to a frequency shift in the scattered light. In contrast, static scatterers in tissue do not change light frequency but lead to randomization of light directions impinging on red blood cells. This serves as a reference signal. This diffusion of light in vascularized tissue leads to a broadening of the spectrum of scattered light, from which the mean red blood cell velocity, blood volume, and blood flow can be calculated in relative units.21 In the present study, laser Doppler flowmetry was performed in the fovea to assess ChBF. For this purpose, a fundus camera–based system was used(Oculix 4000; Oculix Sarl, Arbaz, Switzerland).
The eye is illuminated by the beam of a single-mode laser diode (λ= 783 nm) along the optical axis.22 The light is reflected at the front side of the cornea and at the retina. The 2 reemitted waves produce interference fringes from which the distance changes between the cornea and the retina during a cardiac cycle can be evaluated. The FPA, which is the maximum distance change between the cornea and the retina during the cardiac cycle, is an estimate of pulsatile ChBF.23 Again, measurements were performed in the fovea.
Statistical analysis was carried out using a statistical software program(CSS Statistica for Windows; Statsoft, Inc, Tulsa, Okla). For data description, outcome variables are expressed as percentage change from baseline. A 2-way analysis of variance model was used to analyze the data. For comparison vs baseline, we used the time effect to calculate the Pvalue. The effect vs baseline is given as the maximum percentage change over the trial period. P values comparing the effect of exogenous ET-1 in the absence or presence of topical unoprostone were calculated from the interaction of time by treatment. Data are presented as mean ± SEM. P<.05 was considered statistically significant.
Baseline values of mean brachial artery pressure, pulse rate, and IOP are given in Table 1. Changes in these variables during the study are illustrated in Figure 1. Mean brachial artery blood pressure was slightly elevated on both study days vs baseline (placebo day: 6% ± 2%, P = .007; unoprostone day: 9% ± 2%; P<.001), but this increase was not significantly different between study days. Pulse rate decreased by 11% ± 4% vs baseline on the placebo day (P = .004) and by 10% ± 3% vs baseline on the day of administration of unoprostone (P<.001). Again, the effect of ET-1 on pulse rate was not different between study days. Intraocular pressure decreased significantly by 7% ± 3% and 14% ± 4% vs baseline on the placebo and unoprostone days, respectively (P<.001 for both). The IOP-lowering effect was more pronounced on the unoprostone study day than on the placebo study day (P = .01).
Figure 2 shows the effects of ET-1 and unoprostone on ocular hemodynamic variables. Infusion of ET-1 led to a significant decrease in FPA. At 150 minutes, FPA was reduced by 19%± 2% from baseline (P<.001) when placebo was applied. Administration of unoprostone attenuated the FPA-lowering effect of ET-1: FPA was reduced by a maximum of 12% ± 2% from baseline (P<.001). Thus, 150 minutes after the start of ET-1 infusion, the effect on FPA was significantly blunted by coadministration of unoprostone(P = .003 vs placebo).
Measurement of ChBF exhibited comparable results; during administration of ET-1, ChBF was significantly decreased vs baseline by 17% ± 2% (P<.001) on the placebo day. As with FPA, coadministration of unoprostone diminished this effect: 90 minutes after the start of ET-1 administration, ChBF reached its lowest level (reduction of 11% ± 2% from baseline) and slightly increased thereafter (P<.001 vs baseline). At the end of ET-1 infusion, the reduction in ChBF from baseline was 7% ± 2% (P = .04 vs placebo).
Results of the present study indicate a functional antagonism between ET-1 and unoprostone on the level of the human choroid. This was evidenced from 2 independent measurements of ChBF using laser Doppler flowmetry and of FPA using laser interferometry. At the highest administered dose, unoprostone blunted the effect of ET-1 on FPA by more than 40% and on ChBF by more than 50%. The mechanisms underlying this functional antagonism are hitherto unidentified. In the human choroid, the potent vasoconstrictor effects of ET-1 are primarily mediated via the ET-A receptor,24 which is located directly on the smooth muscle cells. No data, however, are currently available, to our knowledge, to indicate that unoprostone has any affinity to this ET receptor subtype. Preliminary data suggest rather that unoprostone isopropyl does not bind to either ET-A or ET-B receptors (CIBA-Vision internal report, MDS Panlabs Pharmacology Services; 1999).
Another possibility is that unoprostone interacts with the release of cytoplasmic Ca2+, which is responsible for many cellular mechanisms induced by ET-1.25 The effects of ET-1 on vascular tone in a specific vascular bed depend on the relative expression of ion channels and on the spatial and temporal pattern of the Ca2+ signals. It is obvious from the present study results that in the human choroid, ET-1–induced vasoconstriction predominates over vasodilatory effects of the peptide, which may be due to activation of nonselective cation channels, Ca2+-activated Cl−channels, and voltage-dependent L-type Ca2+channels. This mechanism would be consistent with previous results,26 where it was shown that in the human ONH, effects of ET-1 can be antagonized by administering low-dose nifedipine. Effects of unoprostone on ET-1–induced changes in intracellular calcium concentration in vascular smooth muscle cells, however, remain to be established.
In the present study, we cannot exclude effects on ocular perfusion pressure during drug administration because we observed significant changes in mean brachial artery blood pressure and IOP on both study days. However, a systemic hypertensive effect of ET-1 administration was observed on both study days and therefore does not limit the comparability of our ChBF variables on the 2 study days. Intraocular pressure was reduced on the placebo and unoprostone days, but this effect was more pronounced when unoprostone was coadministered with exogenous ET-1. Therefore, ocular perfusion pressure may have been slightly higher on the unoprostone day compared with the placebo day. The effect of unoprostone on IOP at less than 2 mm Hg was, however, small in the present study, and it is unlikely that this small effect had any detectable effect on choroidal perfusion. In the present study, administration of unoprostone blunted the ET-1–induced effect on ChBF by more than 50%, which cannot be caused by changes in ocular perfusion pressure of a few millimeters of mercury.
We observed a significant reduction in IOP on the placebo day when ET-1 was intravenously administered. Previous animal and in vitro studies on the effect of exogenous ET-1 on IOP are contradictory, and reduced27,28 and increased29 IOPs have been reported. One must be careful when interpreting the results of infusion of ET-1 on IOP in the present study because no placebo control was used with respect to ET-1, and a reduction in IOP could result from repeated applanation tonometry. On the other hand, systemic administration of ET-1 may induce effects on ciliary body blood flow, which in turn may decrease aqueous humor production. Our data, therefore, do not contradict the view that ET-1 is an important factor contributing to contraction of the trabecular meshwork.30
Our findings could be of interest in patients with glaucoma, in which ET-1 is assumed to play a pathogenic role.31 This may be related not only to the role of ET-1 in the control of IOP but also to the reduction in ONH blood flow as measured in patients with glaucoma2,32,33 potentially mediated by ET-1. The clinical significance of the results of the present trial, however, remains to be investigated. A recent study in patients with normal-tension glaucoma did not show an effect of topical unoprostone on ocular hemodynamic variables.34 Although patients included in this trial were tested for vasospastic diathesis, the local level of ET-1 in the eye was unknown. Moreover, short-term experiments applying high doses of unoprostone are not necessarily comparable to long-term effects of unoprostone in patients with glaucoma, and it is unclear whether a functional antagonism may also be achieved with lower doses of the drug. Results of in vitro experiments, however, indicate that already low doses of unoprostone exert vasodilation in ET-1–contracted pig retinal arterioles.35
Endothelin 1 has also been implicated in the pathogenesis of other eye diseases, including diabetic retinopathy,36,37 retinal vein occlusion,38 and human immunodeficiency virus–related retinopathy.39 Hence, there is significant need to further establish the potential role of drugs that interfere with the ET-1 system in the treatment of ocular vascular disease. The present study may establish unoprostone as a candidate for further investigations in this direction.
A limitation of all studies investigating ocular blood flow responses is that no technique is capable of directly quantifying perfusion in the eye. Kiel40 provided an excellent overview of all currently available techniques for the assessment of ocular blood flow. He concluded that no gold standard method is available, neither in humans nor in experimental animals. However, effects of unoprostone on ET-1–induced vasoconstriction in the present study were evidenced with 2 independent methods. In addition, results of previous studies indicate that the techniques used provide consistent results with several different stimuli.41,42
In conclusion, our data indicate that unoprostone isopropyl applied topically in the eye partially reverses the vasoconstrictor effect of infused exogenous ET-1 in the human choroid. Whether this effect is of clinical relevance for the treatment of patients with glaucoma remains to be shown.
Submitted for publication March 6, 2001; final revision received November 7, 2001; accepted November 16, 2001.
Unoprostone isopropyl (0.12%) (Rescula; Novartis Ophthalmics, Basel, Switzerland) was donated by Novartis Ophthalmics.
Corresponding author: Leopold Schmetterer, PhD, Department of Clinical Pharmacology, Allgemeines Krankenhaus Wien, Waehringer Guertel 18-20, A-1090 Vienna, Austria (e-mail: email@example.com).