The Effect of Brimonidine Tartrate on Retinal
`Blood Flow in Patients With Ocular
`Hypertension
`
`ANTHONY M. CARLSSON, BSC, BALWANTRAY C. CHAUHAN, PHD, A. ARLENE LEE, RN,
`AND RAYMOND P. LEBLANC, MD
`
`c PURPOSE: To study the effects of topical brimonidine
`tartrate 0.2%, an a2-agonist ocular hypotensive drug, on
`retinal capillary blood flow in patients with ocular hyper-
`tension.
`c METHODS: The study was a double-masked, random-
`ized, placebo-controlled trial set in a tertiary eye center.
`Ocular hypertensive patients with repeatable intraocular
`pressures greater than 21 mm Hg and normal visual fields
`and optic disks were consecutively recruited. After an
`eye examination, baseline retinal blood flow measure-
`ments were made with confocal scanning laser Doppler
`flowmetry in one study eye. Patients were then randomly
`assigned to receive either brimonidine or placebo (saline)
`twice daily for 8 weeks. Blood flow and intraocular
`pressure measurements were then repeated after 4 and 8
`weeks.
`c RESULTS: Seventeen patients were randomly assigned
`to receive brimonidine, and 14 received placebo. One
`patient in each group failed to complete the study. The
`mean group differences in baseline age and intraocular
`pressure were not
`statistically significant
`(59.23
`[610.24] and 52.23 [616.46] years, respectively, and
`24.84 [62.08] and 24.56 [62.85] mm Hg, respectively).
`Brimonidine reduced intraocular pressure by 17.90%
`and 16.17% at 4 and 8 weeks, respectively, with a
`significant difference in treatment effect compared with
`the placebo group (P < .007). The group difference in
`treatment effect in any of the three hemodynamic param-
`eters velocity, volume, and flow was within 8% and not
`
`Accepted for publication June 9, 1999.
`From the Faculty of Medicine (Mr Carlsson), Departments of Oph-
`thalmology (Drs Chauhan, Lee, and LeBlanc) and Physiology and
`Biophysics (Dr Chauhan), Dalhousie University, Halifax, Nova Scotia,
`Canada.
`This study was supported by the MacKeen Studentship from the
`Dalhousie Medical Research Foundation, Halifax, Nova Scotia, Canada
`(Mr Carlsson), and by a grant from Allergan Pharmaceuticals, Irvine,
`California (Dr Chauhan).
`Reprint requests to Balwantray C. Chauhan, PhD, Department of
`Ophthalmology, Dalhousie University, 2nd Floor Centennial Building,
`1278 Tower Rd, Halifax, Nova Scotia, Canada B3H 2Y9; fax: (902)
`473-2839; e-mail: bal@is.dal.ca
`
`significantly different at 4 or 8 weeks (P > .360). Based
`on a type I error of 0.05, our study had a power greater
`than or equal to 75% to detect group differences in
`treatment effect of greater than or equal to 15% to 20%.
`c CONCLUSIONS: Brimonidine reduces intraocular pres-
`sure without altering retinal capillary blood flow in
`patients with ocular hypertension.
`(Am J Ophthalmol
`2000;129:297–301. © 2000 by Elsevier Science Inc. All
`rights reserved.)
`
`B RIMONIDINE TARTRATE IS AN a2-AGONIST OCULAR
`
`hypotensive drug that exerts its effect by causing
`both a decrease in aqueous production and an
`increase in uveoscleral outflow.1 It is relatively new and
`has been proven to reduce increased intraocular pressures
`in glaucoma and ocular hypertension.2,3 As an a2- agonist,
`brimonidine belongs to the same class of drugs as clonidine
`and apraclonidine4; however, its molecular structure is
`sufficiently different to make it more selective for the
`a2-receptor than either clonidine or apraclonidine. Unlike
`clonidine, brimonidine does not appear to have an effect
`on the central nervous system and therefore does not cause
`sedation or systemic hypotension.5 Unlike apraclonidine,
`brimonidine causes less ocular irritation and allergy, prob-
`ably because it has a lower oxidation potential than
`apraclonidine.6,7 Brimonidine has few reported side effects,
`the most common being dry mouth.2,3,7 Studies have also
`shown that brimonidine is free of any cardiopulmonary
`side effects that are associated with the b-blocking ocular
`hypotensive agents.8,9
`Because brimonidine is an adrenergic agent, we wanted
`to determine whether topical application produced any
`significant effect on retinal microvascular blood flow.
`
`PATIENTS AND METHODS
`
`THIS STUDY WAS A SINGLE-CENTERED, DOUBLE-MASKED,
`randomized, placebo-controlled trial that determined the
`effect of brimonidine tartrate 0.2% on retinal capillary
`
`0002-9394/00/$20.00
`PII S0002-9394(00)00389-5
`
`© 2000 BY ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED.
`
`297
`
`Page 1 of 5
`
`SLAYBACK EXHIBIT 1020
`
`

`

`blood flow in ocular hypertensive patients. Patients with
`increased intraocular pressure but normal visual fields and
`optic disks were recruited consecutively from the Eye Care
`Centre of the Queen Elizabeth II Health Sciences Centre,
`a tertiary care referral center. Informed consent was ob-
`tained from all patients. The study was approved by the
`Research Ethics Committee of the Queen Elizabeth II
`Health Sciences Centre.
`Patients were included in the study if they had all of the
`following: best-corrected visual acuity of 20/30 or better;
`normal visual fields in both eyes with a mean deviation
`index of better than 22.00 dB and a normal Glaucoma
`Hemifield Test10 using the STATPAC program for the
`Humphrey Field Analyzer (Humphrey Instruments, San
`Leandro, California); clinically normal funduscopic exam-
`ination,
`including a normal-appearing optic disk; and
`documented intraocular pressure greater than 21 mm Hg
`on three separate occasions without treatment.
`Patients were excluded if they had any of the following:
`diabetes; any disease known to affect blood flow,
`for
`example, polycythemia, temporal arteritis, and systemic
`vascular disease; use of oral calcium channel blockers, b-
`blockers, a-agonists, or angiotensin-converting enzyme
`inhibitors; refractive error greater than 6 diopters (spher-
`ical equivalent) or astigmatism greater than 2 diopters;
`aphakia or pseudophakia; and untreated intraocular pres-
`sure greater than 30 mm Hg.
`Retinal blood flow was measured noninvasively using
`confocal scanning laser Doppler flowmetry,11 a modifica-
`tion of the laser Doppler flowmetry technique described by
`Riva and associates.12 The technique relies on measuring
`time-related intensity variations of backscattered light
`from an illuminated spot on the fundus. These intensity
`variations are the result of interference between backscat-
`tered light from stationary structures, such as tissue and
`vessel walls, and from moving blood particles. The inten-
`sity variation measurements are subjected to a fast Fourier
`transform to obtain the power spectrum of the multiple
`frequency shift components. Thereafter, three hemody-
`namic variables, velocity, volume, and flow, are computed
`from the power spectrum in arbitrary units.13 Confocal
`scanning laser Doppler flowmetry was carried out using the
`Heidelberg Retina Flowmeter (Heidelberg Engineering
`GmbH, Heidelberg, Germany). The instrument and its
`operation have been detailed elsewhere.14 Briefly, a diode
`laser (wavelength, 780 nm) is used to scan an area of 10.0
`degrees by 2.5 degrees after it has been focused at the
`desired axial plane (for example, the retina). The image
`resolution is 256 by 64 picture elements (pixels). Each of
`the 64 horizontal lines is scanned 128 times, with a line
`repetition rate of 4 kHz over 2.05 seconds. The 128
`intensity measurements at each location are made by a
`photodiode behind a confocal pinhole. The result of each
`processed scan is a two-dimensional perfusion map of the
`imaged area (Figure 1). The measurements derived with
`this technique have been shown to have good reproduc-
`
`FIGURE 1. Confocal scanning laser Doppler flowmetry images
`obtained from a single scan of the temporal peripapillary retina
`of a subject in the study. (Top) Reflectivity image, akin to a
`funduscopic image and perfusion maps showing the (second)
`velocity, which is proportional to the mean Doppler shift;
`(third) volume, which is proportional to the concentration of
`particles moving through the sampling volume of tissue; and
`(bottom) flow, which is proportional to the blood flow through
`the sampling volume. The brightness of the pixels is scaled
`according to the respective hemodynamic values.
`
`ibility14 –16 and to have a linear relationship to actual flow
`rates in an experimental model system.17
`Patients who were receiving treatment for ocular hyper-
`tension underwent a 4-week washout period before enroll-
`ment. For safety purposes these patients had a full eye
`examination including intraocular pressure measurement
`halfway through the washout. At the baseline visit before
`randomization, patients underwent a full ophthalmic ex-
`amination including visual acuity measurement, slit-lamp
`evaluation of the anterior segment, intraocular pressure
`measurement, and funduscopic examination of the retina
`and optic disk. Automated central visual fields were then
`recorded in both eyes with the 24-2 program of the
`Humphrey Field Analyzer to ensure eligibility. All patients
`were previously exposed to automated perimetry on several
`occasions. If both eyes were eligible for the study, one eye
`was randomly selected as the study eye. Confocal scanning
`laser Doppler flowmetry was then carried out in the study
`eye. Five good-quality images of the temporal retina, with
`a portion of the optic disk, were recorded. Patients were
`then randomly assigned to receive either brimonidine
`tartrate 0.2% (Alphagan; Allergan Pharmaceuticals, Ir-
`vine, California) or saline twice daily in the study eye. The
`
`298
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`MARCH 2000
`
`Page 2 of 5
`
`SLAYBACK EXHIBIT 1020
`
`

`

`eye examination, tonometry, and flowmetry (with images
`recorded in the same retinal location as baseline) were
`repeated 4 and 8 weeks after baseline. The follow-up visits
`were scheduled at approximately the same time of day
`(within 1 hour) as baseline.
`After the final visit we analyzed the flowmetry data in
`each scan by averaging the hemodynamic values in each of
`five 100 3 100-mm areas (each containing 100 individual
`measurements) in the temporal peripapillary retina. We
`ensured that measurements were made only in capillary
`beds and that for a given patient they were made in the
`same locations by using landmarks on the fundus reflec-
`tivity image (and not the perfusion maps), which is also
`provided in the analysis. We determined the session mean
`for each of the five 100 mm by 100mm areas and after
`applying a log-transform determined the change in the
`hemodynamic parameters at
`the 4-week and 8-week
`points. After breaking the treatment code, we compared
`the treatment effect between the brimonidine and placebo
`group at the 4-week and 8-week points using a group t test.
`Intraocular pressure and age data were also compared using
`a group t test. We used group difference in treatment effect
`as the test statistic because it allowed control within
`subjects (treatment effect) and a direct comparison be-
`tween the brimonidine and placebo groups (group t test).
`
`RESULTS
`
`THE STUDY HAD 14 PATIENTS IN THE BRIMONIDINE GROUP
`and 17 patients in the placebo group. One patient,
`randomly assigned to receive placebo, was withdrawn after
`4 weeks, when her intraocular pressure exceeded 30 mm
`Hg. Another patient, randomly assigned to receive bri-
`monidine, withdrew voluntarily in the sixth week of the
`study after developing dizziness and headaches. Thus, 13
`patients in the brimonidine group and 16 patients in the
`placebo group completed the study.
`The mean age (61 SD) in the brimonidine and
`placebo groups was 59.23 (610.24) and 52.23 (616.46)
`years, respectively. The respective figures for baseline
`intraocular pressure were 24.84 (62.08) and 24.56
`(62.85) mm Hg. The group difference in neither
`baseline age nor intraocular pressure was statistically
`significantly different (P . .200).
`The intraocular pressure treatment effect between the
`brimonidine and placebo groups was significantly different
`at both the 4-week and 8-week points (P , .007, Figure 2),
`with the pressure reduction in the brimonidine group being
`17.90% and 16.17%, respectively. The difference in treat-
`ment effect between the brimonidine and placebo
`groups was within 8% for each of the three blood-flow
`parameters analyzed at either the 4-week or 8-week
`points (Figure 3). These differences were not signifi-
`cantly different (P . .360). There was also no relationship
`between the change in intraocular pressure and change in
`
`FIGURE 2. Change in intraocular pressure at the 4-week and
`8-week points from baseline in the brimonidine and placebo
`groups (error bar 5 61 SE).
`
`blood flow in either the brimonidine or placebo group at
`the 4-week (P . .790) or 8-week (P . .590) points.
`Based on the study sample size, we determined the
`statistical power to detect differences in treatment effect
`between the brimonidine and placebo groups. These cal-
`culations were done for 5%, 10%, 15%, 20%, 25%, and
`30% group differences in treatment effect (Figure 4) and
`show that our study was adequately powered (greater than
`or equal to 75%, with type I error [a] 5 0.05) to detect
`group differences of greater than or equal to 15% for the
`volume parameter and greater than or equal to 20% for the
`velocity and flow parameters.
`
`DISCUSSION
`
`PREVIOUS STUDIES HAVE SHOWN THAT a2-ADRENORECEP-
`tors play an important role in microvascular autoregula-
`tion, particularly at the level of the terminal arteriole in
`vascular smooth muscle.18 –20 It is known that the retinal
`and optic nerve head microvascular network is under
`sensitive autoregulatory control,21,22 and that a2 binding
`sites are present in the retinal vasculature.23 Spada and
`associates24 studied the effect of various concentrations of
`brimonidine tartrate, an a2-agonist, on human retinal
`arteriolar diameter in retinal tissue that had been trans-
`planted into hamster cheek pouch membrane. They
`showed that brimonidine concentrations of up to 1025 M
`had no effect on vessel caliber. However, until very
`recently, no clinical studies had been published to deter-
`mine whether topical brimonidine had any effect on ocular
`blood flow.
`Lachkar and associates25 performed a double-masked
`randomized placebo-controlled crossover study that exam-
`ined the effect of brimonidine tartrate 0.2% on retrobulbar
`blood flow in patients with ocular hypertension. Using
`color Doppler ultrasound, they found no significant effect
`of brimonidine on the blood velocities in the central
`retinal, ophthalmic, nasal, and temporal ciliary arteries.
`
`VOL. 129, NO. 3
`
`BRIMONIDINE AND RETINAL BLOOD FLOW
`
`299
`
`Page 3 of 5
`
`SLAYBACK EXHIBIT 1020
`
`

`

`FIGURE 3. Change in the hemodynamic parameters, velocity, volume, and flow at the 4-week and 8-week points from baseline in
`the brimonidine and placebo groups (error bar 5 61 SE).
`
`than that reported by studies that included patients with
`glaucoma and patients with ocular hypertension.2,3,26 It is
`possible that the composition of the study population may
`yield different magnitudes of pressure reduction with bri-
`monidine.
`Because no difference in the treatment effect in retinal
`blood flow measurements between the brimonidine and
`placebo groups was found, we wanted to ensure that our
`study was adequately powered to determine a clinically
`significant difference had it existed. Given the variability
`characteristics of the hemodynamic measurements, we
`found that we would have detected a group difference in
`treatment effect of 15% to 20% 75 of 100 times had it
`existed and a difference of 17% to 23% 80 of 100 times had
`it existed (Figure 4). Scanning laser Doppler flowmetry has
`been used successfully to measure a statistically significant
`increase in retinal blood flow after subjects inhaled carbo-
`gen27 and reduction after they inhaled 100% oxygen.28
`Similar experiments have also been carried out in experi-
`mental animals.29 These data provide a positive control
`and suggest that confocal scanning laser Doppler flowmetry
`is capable of measuring changes in blood when they exist.
`Using a variety of techniques, investigators have shown
`that test–retest variability of blood flow measurements,
`estimated by the coefficient of variation, ranges from
`approximately 10% to 30%, even within single ses-
`sions.15,16,30 –32 Our earlier work suggests that with confocal
`scanning laser Doppler flowmetry, the largest component
`of measurement variability is physiologic variation in
`blood flow from one point in time to another.17 Given
`these findings, it is arguable whether detecting differences
`in treatment effect of less than 15% has much clinical
`significance, because any beneficial or detrimental drug
`effects on blood flow should clearly exceed the measured
`variability. Ideally, any functional consequences of altered
`blood flow, such as change in the visual field or optic disk,
`should ultimately be the variables of interest.
`The etiology of glaucoma is likely to be multifactorial,
`
`FIGURE 4. Power of the present study (based on a total sample
`size of 30 patients) to detect differences between the bri-
`monidine and placebo groups of 5% to 30% in treatment effect
`for the three hemodynamic parameters velocity, volume, and
`flow. Calculations are based on a type I error (a) of 0.05 and
`indicate that power greater than or equal to 75% to detect
`group differences of greater than or equal to 15% for the
`volume parameter and greater than or equal to 20% for the
`velocity and flow parameters.
`
`Despite the same general conclusions, their study differed
`with ours in that they measured blood velocity in the large
`vessels that supply the whole eye; in our study measure-
`ments were taken exclusively from retinal capillary beds.
`Studies have shown that a2-adrenoreceptors are distrib-
`uted differently across arteriolar, precapillary sphincter,
`and venular segments of various tissue microvascular
`beds.18,20 The fact that our results are similar to those of
`Lachkar and associates suggests that brimonidine has no
`apparent effect on either large-caliber or small-caliber
`vessels supplying the retina.
`Our study showed that brimonidine significantly re-
`duced intraocular pressure in patients with ocular hyper-
`tension. The magnitude of pressure reduction was similar
`to that reported by Lachkar and associates,25 whose sub-
`jects all had ocular hypertension; however, it was lower
`
`300
`
`AMERICAN JOURNAL OF OPHTHALMOLOGY
`
`MARCH 2000
`
`Page 4 of 5
`
`SLAYBACK EXHIBIT 1020
`
`

`

`with studies pointing to a possible role for impaired blood
`flow as one of the potentiating mechanisms of optic nerve
`damage.33 In view of the fact that a common initial line of
`treatment for glaucoma is intraocular pressure reduction
`with topical adrenergic drugs, it is important to determine
`whether these drugs have deleterious effects on blood flow.
`Our study suggests that topical brimonidine reduces in-
`traocular pressure without altering retinal blood flow.
`
`REFERENCES
`
`1. Toris CB, Gleason ML, Camras CB, Yablonski ME. Effects of
`brimonidine on aqueous humor dynamics in human eyes.
`Arch Ophthalmol 1995;113:1514 –1517.
`2. Schuman JS, Horwitz B, Choplin NT, et al. A 1-year study
`of brimonidine twice daily in glaucoma and ocular hyperten-
`sion. A controlled, randomized, multicenter clinical trial.
`Chronic Brimonidine Study Group. Arch Ophthalmol 1997;
`115:847– 852.
`3. Derick RJ, Robin AL, Walters TR, et al. Brimonidine
`tartrate: a one-month dose response study. Ophthalmology
`1997;104:131–136.
`4. Huang AS, Pollack IP. Apraclonidine and the treatment of
`glaucoma. Ophthalmol Clin North Am 1995;8:303–314.
`5. Wilensky JT. The role of brimonidine in the treatment of
`open-angle glaucoma. Surv Ophthalmol 1996;41(suppl 1):
`S3–S7.
`6. Munk SA, Wiese A, Thompson CD, MacDonald T. Oxida-
`tion potential and allergic response of a-2 agonists. Invest
`Ophthalmol Vis Sci 1996;37(suppl):S832.
`7. Walters TR. Development and use of brimonidine in treating
`acute and chronic elevations of intraocular pressure: a review
`of safety, efficacy, dose response, and dosing studies. Surv
`Ophthalmol 1996;41(suppl 1):S19 –S26.
`8. Nordlund JR, Pasquale LR, Robin AL, et al. The cardiovas-
`cular, pulmonary, and ocular hypotensive effects of 0.2%
`brimonidine. Arch Ophthalmol 1995;113:77– 83.
`9. Serle JB. A comparison of the safety and efficacy of twice
`daily brimonidine 0.2% versus betaxolol 0.25% in subjects
`with elevated intraocular pressure. The Brimonidine Study
`Group III. Surv Ophthalmol 1996;41(suppl 1):S39 –S47.
`10. Åman P, Heijl A. Glaucoma Hemifield Test. Automated
`visual field evaluation. Arch Ophthalmol 1992;110:812–
`819.
`11. Michelson G, Schmauss B. Two dimensional mapping of the
`perfusion of the retina and optic nerve head. Br J Ophthal-
`mol 1995;79:1126 –1132.
`12. Riva CE, Harino S, Petrig BL, Shonat RD. Laser Doppler
`flowmetry in the optic nerve. Exp Eye Res 1992;55:499 –506.
`13. Bonner RF, Nossal R. Principles of laser-Doppler flowmetry.
`In: Shepherd AP, O¨ dberg PÅ, editors. Laser-Doppler blood
`flowmetry. Boston: Kluwer Academic Publishers, 1990:17–
`45.
`14. Michelson G, Schmauss B, Langhans MJ, Harazny J, Groh
`MJ. Principle, validity, and reliability of scanning laser
`Doppler flowmetry. J Glaucoma 1996;5:99 –105.
`15. Nicolela MT, Hnik P, Schulzer M, Drance SM. Reproduc-
`ibility of retinal and optic nerve head blood flow measure-
`ments with scanning laser Doppler flowmetry. J Glaucoma
`1997;6:157–164.
`16. Chauhan BC. Confocal scanning laser Doppler flowmetry of
`
`the retina and optic nerve head. In: Anderson DR, Drance
`SM, editors. Encounters in glaucoma research 3: how to
`ascertain progression and outcome. Amsterdam: Kugler,
`1996:263–276.
`17. Chauhan BC, Smith FM. Confocal scanning laser Doppler
`flowmetry: experiments in a model flow system. J Glaucoma
`1997;6:237–245.
`18. Faber JE. In situ analysis of alpha-adrenoceptors on arteriolar
`and venular smooth muscle in rat skeletal muscle microcir-
`culation. Circ Res 1988;62:37–50.
`19. Faber JE, Meininger GA. Selective interaction of alpha-
`adrenoceptors with myogenic regulation of microvascular
`smooth muscle. Am J Physiol 1990;259:H1126 –H1133.
`20. McGillivray-Anderson KM, Faber JE. Effect of reduced blood
`flow on alpha 1- and alpha 2-adrenoceptor constriction of rat
`skeletal muscle microvessels. Circ Res 1991;69:165–173.
`21. Orgu¨l S, Meyer P, Cioffi GA. Physiology of blood flow
`regulation and mechanisms involved in optic nerve perfu-
`sion. J Glaucoma 1995;4:427– 443.
`22. Harris A, Ciulla TA, Chung HS, Martin B. Regulation of
`retinal and optic nerve blood flow. Arch Ophthalmol 1998;
`116:1491–1495.
`23. Forster BA, Ferrari-Dileo G, Anderson DR. Adrenergic
`alpha 1 and alpha 2 binding sites are present in bovine
`retinal blood vessels. Invest Ophthalmol Vis Sci 1987;28:
`1741–1746.
`24. Spada CS, Nieves AL, Burke JA, Woodward DF, Wheeler
`LA. Comparative effects of a-2 adrenoreceptor agonists on
`microvessel caliber in human retinal tissue. In: Messmerk K,
`Kubler W, editors. Sixth World Congress for Microcircula-
`tion. Bologna: Monduzzi, 1996:511–514.
`25. Lachkar Y, Migdal C, Dhanjil S. Effect of brimonidine
`tartrate on ocular hemodynamic measurements. Arch Oph-
`thalmol 1998;116:1591–1594.
`26. LeBlanc RP. Twelve-month results of an ongoing random-
`ized trial comparing brimonidine tartrate 0.2% and timolol
`0.5% given twice daily in patients with glaucoma or ocular
`hypertension. Brimonidine Study Group 2. Ophthalmology
`1998;105:1960 –1967.
`27. Lietz A, Hendrickson P, Flammer J, Orgul S, Haefliger IO.
`Effect of carbogen, oxygen and intraocular pressure on
`Heidelberg retina flowmeter parameter “flow” measured at
`the papilla. Ophthalmologica 1998;212:149 –152.
`28. Strenn K, Menapace R, Rainer G, et al. Reproducibility and
`sensitivity of scanning laser Doppler flowmetry during graded
`changes in PO2. Br J Ophthalmol 1997;81:360 –364.
`29. Chauhan BC, Yu D-Y, Cringle SJ, Carlsson AM, Su E-N.
`Confocal scanning laser ophthalmoscopy and Doppler flow-
`metry of the rat retina. Invest Ophthalmol Vis Sci 1997;
`38(suppl):S274.
`30. Rankin SJ, Walman BE, Buckley AR, Drance SM. Color
`Doppler imaging and spectral analysis of the optic nerve
`vasculature in glaucoma. Am J Ophthalmol 1995;119:685–
`693.
`31. Harris A, Williamson TH, Martin B, et al. Test/retest
`reproducibility of color Doppler imaging assessment of blood
`flow velocity in orbital vessels. J Glaucoma 1995;281–286.
`32. Joos KM, Pillunat LE, Knighton RW, Anderson DR, Feuer
`WJ. Reproducibility of laser Doppler flowmetry in the human
`optic nerve head. J Glaucoma 1997;6:212–216.
`33. Nicolela MT, Drance SM, Rankin SJ, Buckley AR, Walman
`BE. Color Doppler imaging in patients with asymmetric
`glaucoma and unilateral visual field loss. Am J Ophthalmol
`1996;121:502–510.
`
`VOL. 129, NO. 3
`
`BRIMONIDINE AND RETINAL BLOOD FLOW
`
`301
`
`Page 5 of 5
`
`SLAYBACK EXHIBIT 1020
`
`