British Journal of Ophthalmology 1996; 80: 217-223

Effect of isoproterenol, phenylephrine, andsodium nitroprusside on fundus pulsations inhealthy volunteers

Leopold Schmetterer, Michael Wolzt, Alex Salomon, Alexander Rheinberger,Christian Unfried, Gabriele Zanaschka, Adolf Friedrich Fercher

AbstractAims/Background-Recently a laser inter-ferometric method for topical measure-ment of fundus pulsations has beendeveloped. Fundus pulsations in the macu-lar region are caused by the inflow and out-flow ofblood into the choroid. The purposeof this work was to study the influence ofaperipheral vasoconstricting (the a,x adreno-ceptor agonist phenylephrine), a predomi-nantly positive inotropic (the non-specificI adrenoceptor agonist isoproterenol), anda non-specific vasodilating (sodium nitro-prusside) model drug on ocular funduspulsations to determine reproducibilityand sensitivity ofthe method.Methods-In a double masked ran-domised crossover study the drugs wereadministered in stepwise increasing dosesto 10 male and nine female healthyvolunteers. Systemic haemodynamicvariables and fundus pulsations weremeasured at all infusion steps.Results-Fundus pulsation increasedduring infusion of isoproterenol withstatistical significance versus baseline atthe lowest dose of 0-1 uwg/min. Neitherperipheral vasoconstriction nor peripheralvasodilatation affected the ocular funduspulsations.Conclusions-Measurements of funduspulsations is a highly reproducible methodin healthy subjects with low ametropy.Changes oflocal pulsatile ocular blood flowwere detectable with our method followingthe infusion of isoproterenol. As systemicpharmacological vasodilatation or vaso-constriction did not change fundus pulsa-tions, further experimental work has to bedone to evaluate the sensitivity of the laserinterferometric fundus pulsation measure-ment in various eye diseases.(BrJ Ophthalmol 1996; 80: 217-223)

Although 85% of the blood volume in the eyecirculates in the choroid,' choroidal circulationis far less investigated than the retinal bloodflow. The outer layers of the retina arenourished by the choroid, and animal experi-ments suggest that 90% of the oxygenation ofthe photoreceptors are supplied by choroidalcirculation.2 The introduction of fluoresceinangiography,3 particularly of video fluoresceinangiography,4 has increased the understandingof choroidal circulation. However, systematicinvestigations of choroidal haemodynamics

have not yet been carried out and quantitativepressure flow relations in human choroidalvessels are as yet unknown. Linear choroidalpressure flow relations have been obtained inanimal experiments in different species.5-8 Inthe rabbit, the choroidal blood flow has beenshown to be pressure independent when IOPwas less than 20-25 mm Hg.9

Blood vessels can be considered as cylin-ders filled with fluid at a pressure greater thanthat outside the cylinders. The pressure dif-ference between the inside and the outside ofa vessel is called the transmural pressure P.The corresponding tension T in the vesselwall can be calculated by Laplace's lawP=T/R, where R is the radius of the cylinder.Any variation of the transmural pressurealters the tension of the vessel wall. Thus,pressure oscillations of cardiac pulse pressureoutput lead to a pulsation of the vessel wall.As blood is pumped into an artery, the systolicpressure increases and dilates the vessel wall.When cardiac ejection decreases, the pressurefalls and the vessel wall returns to its equilib-rium position, the diastolic pressure.10 Thearterial pressure contour becomes progres-sively more distorted as the wave is transmit-ted down the arterial system. The elasticproperties of an artery are described by thearterial compliance, the change in diameterover the change in pressure. The non-linearelastic response of arteries implies that theirmechanical properties depend on the meanarterial pressure.11

In retinal and choroidal vessels the meantransmural pressure at the arteries entering theeye nearly equals the mean perfusion pressure,defined as the pressure in the arteries enteringthe eye (Pa) minus the pressure in the veins(Pv) leaving the eye. Pf=Pa-Pv. The pressurein the veins, Pv, is significantly higher than theintraocular pressure (IOP) in the retinal ves-sels'2 whereas the difference in the choroidalvessels is small.'3 Therefore, the transmuralpressure in the veins is rather small or evenzero. Changes in transmural pressure and theconsecutive change in vessel diameter lead topulsations of the surrounding tissue.Our work is concerned with local tissue

pulsations at the foveola. This area of highestvisual acuity is approximately 350 ,um in dia-meter.'4 This is a little smaller than the retinalavascular zone measuring approximately500-600 ,um in diameter. In our study, wherethe area involved in the measurement is 20-50,um, the effect of tissue pulsations is only influ-enced by the blood flow in choroidal vessels.

Institute ofMedicalPhysicsL SchmettererC UnfriedA F Fercher

Department of ClinicalPharmacologyL SchmettererM WolztG Zanaschka

Department ofOphthalmology BA SalomonA Rheinberger

Correspondence to:L Schmetterer, Institut fMrMedizinische Physik,Wahringer Strasse 13,A-1090 Wien, Austria.Accepted for publication9 November 1995

217

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Schmetterer, Wolzt, Salomon, Rheinberger, Unfried, Zanaschka, Fercher

Hence the inflow andchoroid leads to pulsalfundus.These fundus pulsa

a recently describedwhich measures the dcornea and retina. TIwas to evaluate the siday to day variability,ity, and the sensitivitychanges in ocularvolunteers. We haveeffects of well charmodel drugs on fundimasked randomiseccrossover study.

Materials and methiIn a double maskestudy design, subjectsinfusions of stepwiseperipheral vasoconsvasodilating, a predondrug, and of placebo (tion) on different stipossible sequences of:way crossover design

DRUGS ADMINISTEREDThese were phenyleWinthrop Breon Ladose 0-5, 1, 2, 4, 8,nitroprusside (NipruInfusionsbereitung, SCGermany; dose 0 5, 1isoproterenol (IsupiLaboratories, NY, U0-8, 1P6, 3-2,g/min),tion. The drugs werglucose (for sodiumappropriate concentra

PATIENTS

Ten men (aged 20-33 7) and nine wommean 25-0 (3 6) gave

LCCD

vc

PBSC2

L4 =

BSC

L

nLuL/2

Eye

Figure 1 Optical scheme of the laser interferometer. (PBSC=BSC= beam splitter cube, I12=halfwave plate,L/4= quartVC=video camera, LCCD=linear charge coupled device a?

outflow of blood into the to participate in this study according totile movements ofthe eye procedures approved by the ethics committee

of Vienna University School of Medicine. Allitions are measured with subjects were drug-free, non-smokers andL laser interferometer,15 judged to have no evidence of any relevantlistance changes between physical disorder. All had normal laboratoryhe purpose of this study evaluations, normal results from ultrasoundhort term variability, the examinations of carotid arterial blood flow,the intersubject variabil- and normal findings from ophthalmic exami-of the method to detect nations with ametropy of less than 2-5 dioptresblood flow in healthy and intraocular pressure in the range of 11 to! therefore studied the 16 mm Hg. The testing of the women wasacterised cardiovascular scheduled between the third and eighth day ofas pulsations in a double menstrual cycles; pregnancy was excluded byi placebo controlled urine pregnancy test (hCG Urine+Plus,

Abbott, Wiesbaden, Germany).All subjects were asked to refrain from

alcohol and caffeine for at least 12 hours beforeiods study days. Initially, a 20 gauge plastic cannulai randomised crossover (Venflon, Viggo-Spectramed, Helsingborg,were assigned to receive Sweden) was inserted into a suitable antecubitalincreasing doses of a vein, and a continuous infusion (1 ml/min) of

-tricting, a peripheral physiological saline was started for baselineiinately positive inotropic measurements to the comfortably sitting(physiological saline solu- subjects. Each stepwise infusion period lastedudy days. From the 24 10 minutes unless there was an increase ofadministration in the four >40 mm Hg or decrease of >20 mm Hg inwe randomly chose 19. mean arterial pressure, or until an increase of

>40 per minute or decrease of >20 per minutein heart rate versus baseline infusion period, oruntil any systemic effects - for example, palpita-

phrine (Neosynephrine, tion, headache, or diz.ziness occurred. Measure-aboratories, NY, USA; ments of fundus pulsations were performed in16,ug/kg/min), sodium the last 5 minutes of each infusion step.

iss-Trockensubstanz zur In addition, we conducted measurementsanol-Schwarz, Monheim, to assess the influence of isoproterenol and[,23 4, 8, 16,ug/kg/min), phenylephrine on IOP of eight male volun-rel, Winthrop Breon teers in our study group. Baseline measure-JSA; dose 0O1, 0-2, 0 4, ments of IOP and fundus pulsations werephysiological saline solu- performed during a 10 minute infusion ofe diluted with saline or physiological saline solution. The drug effectnitroprusside) to yield was calculated from measurements of 10

itions. minute infusions of 0-8,ug/min isoproterenoland of 2 jig/kg/min phenylephrine, respec-tively. Two measurements ofIOP and funduspulsation measurements were performed

13 years, mean 25X6 (SD during the last 5 minutes of each infusion stepien (aged 20-30 years, with a Goldmann applanation tonometer. Nowritten informed consent measurements of the effect of sodium nitro-

prusside on IOP were performed, sinceprevious studies indicate that the influenceof systemic administration of sodium nitro-

E prusside16 on IOP is small. In contrast,C topical administration has been representedco to lowerIOP effectively.17

Systolic and diastolic blood pressures weremeasured by an automated oscillometricdevice (HP CMS patient monitor, HewlettPackard, Palo Alto, CA, USA). Mean arterial

670 nm pressure was calculated as two thirds diastolicpressure plus one third systolic pressure; pulse

PBSC1 pressure amplitude was calculated as systolicPBSblood pressure minus diastolic blood pressure.Pulse rate was registered automatically from afinger pulse oximetric device (HP CMS patientmonitor).Measurements of fundus pulsations were

=polarising beam splitter cube, performed with a laser interferometric tech-rer wave plate, L=lens, nique, which has been described in detail else-rray). where.15 The interferometer, schematically

218

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Effect of isoproterenol, phenylephrine, and sodium nitroprusside onfundus pulsations in healthy volunteers

Coefficients of varie

Short term variability

SBP

DBP

MAP

Pulse pressure

Pulse rate

Amplitude

10 20 30 40 0

ation (%) The detecting unit comprises a linear chargecoupled device (CCD) array and a video cam-

Day to day variability era. Owing to the high spatial coherence of theilluminating laser beam, the interferences of

10 20 30 40 the two re-emitted beams are not localised.Optimal visibility can only be achieved in aplane approximately 40 mm in front of the eye.

10 This plane is imaged by the lens onto the CCDarray and via a beam splitter cube (BSC) ontothe photo cathode of the video camera (VC).The video camera is used to supervise themeasurement procedure on a video monitor.The readout of the CCD array is a spatio-

temporal display of the fundus pulsations.Counting the number of fringes movinginwards and outwards the fundus pulsationamplitude (FPA), which is the maximum dis-tance change between cornea and retina duringthe cardiac cycle, can be calculated. The FPArepresents the difference of blood volume

a' during the systole and diastole at one particu-lar point of the fundus.A headrest is used to fix the subject's head.

The whole instrument is mounted on twotranslation stages and can be moved in x and ydirection perpendicular to the optical axis of

Figure 2 Short term (left panel) and day to day (right panel) variability ofourmeasurements in 19 healthy subjects. In the right panel coefficients of variation werecalculatedfor any subjectfrom thefour baseline recordings of the 4 study days. In the leftpanel they were calculatedfrom the seven recordings ofsaline infusion study day. Thebottom of the vertical box line marks 25%, the median 50%o, and the top of the box 75%.Whiskers represent 5% and 95%. Mean (square symbols), minimum (triangles), andmaximum values (circles) are presented. (SBP=systolic blood pressure, DBP=diastolicblood pressure, MAP= mean arterial pressure, pulse pressure=pulse pressure amplitudecalculatedfrom SBP-DBP, amplitude=fundus pulsation amplitude.

shown in Figure 1, is a simplified version oftheinstrument described by Schmetterer"5 basedon the same principle. A single mode laserdiode (Sharp LT 023 MDO; X=783 nm) emitsthe beam used for the measurement. Thebeam of a second laser diode (Toshiba TOLD9211; X=670 nm) is used to adjust the beaminto the patient's eye and is switched off duringthe measurement. The polarising beam splitterPBSC1 combines these two beams to a singlebeam of approximately 1 mm diameter illumi-nating the subject's eye. The beam passes ahalf wave plate which defines the polarisingstate. Then the polarising beam splitterPBSC2 deviates the beam towards the sub-ject's eye. This beam splitter is used togetherwith a quarter wave plate to reduce reflectivelosses in the light re-emitted from the eye whenit passes towards the detecting unit. The beamis reflected both at the front of the cornea andat the retina.These two reflected beams generate non-

localised concentric circular interferencefringes. As the path difference between the twointerfering beams is twice the optical length Lof the eye, the interference order N=2xL/AX isa very large number. Distance variationsbetween cornea and retina lead to a corres-ponding variation AN(t) of the interferenceorder. AN(t) equals the number of fringes thathave moved up to the moment t through afixed point in the interferogram. The timecourse of ihe optical distance variation AL(t)is:

AL(t) =AN(t) x X/2

the subject's eye by computer controlledstepper motors. The observation of the visiblelaser beam reflected at the outer surfaces of theeye and the eyelid facilitates the gross adjust-ment of the instrument.

STATISTICAL ANALYSISStatistical analysis was done with Statistica soft-ware package (StatSoft Inc, Tulsa, OK, USA).Day to day reproducibility was calculated fromthe baseline registrations of the 4 study days.Using the seven measurements of the 10minute saline infusion periods we calculated theshort term variability. For each subject the

8 10

Amplitude

I

12 14 16

6

a1)Ce

C.)

0

oa)

.0

E

z

4

2

o2 4

Figure 3 Results offfundus pulsation measurements in 19healthy subjects under baseline conditions. The amplitude

(1) denotes thefundus pulsation amplitude in units of A/2.

I' II II

+° D

-CJ-HO {t

of{< >ID

H-EE

219

- - - I

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Schmetterer, Wolzt, Salomon, Rheinberger, Unfried, Zanaschka, Fercher

Systolic blood pressure Pulse rate

10

0

-10

-20

20

10 -

0

10

5

0

0 0.5 1.0 2.0 0.5 1.0 2.0

Phenylephrine (p.g/kg/min)Figure 4 Dose-response relation (solid line) of changes (%o) from baseline measurements ofsystemic haemodynamics and offundus pulsations during saline (0) and stepwise infusion ofphenylephrine at doses of 0 5, 1 0, and 2-0 ug/kg/min. Thebroken line presents the results from the placebo study day. The asterisks indicate significant treatment effects versus placeboas cakulated by repeated measure ANOVA (p<0 05). Results are presented as means (SEM).

coefficient of variation of measurements was

calculated.Results of fundus pulsation measurements

were expressed as percentage change from thebaseline. Significance was calculated versusbaseline by repeated measure ANOVA, andversus placebo by two way repeated measure

ANOVA. For these purposes only infusionsteps with less than 20% drop outs were takeninto consideration. The significance level wasset to p=0 05. Significances are presented as

p<0 05 or p<0O005.

ResultsResults for day to day and short term variabil-ity are shown in Figure 2. Interindividualdifferences are shown in Figure 3.

PHYSIOLOGICAL SALINE SOLUTIONNo time dependence of the seven short termreadouts was detected.

PHENYLEPHRINEIn three subjects the infusion was stopped at 2,ug/kg/min, in 10 at 4 ,ug/kg/min, in four at 8,ug/kg/min, and in two at 16 ,ug/kg/min. Therewas a small increase in systolic (p<0.05 versusplacebo and baseline), diastolic (NS versusplacebo, p<005 versus baseline), and meanarterial blood pressure (NS versus placebo,p<0 05 versus baseline). The pulse pressureamplitude was nearly constant during theinfusion steps. The pulse rate decreased dose

dependently (p<0.005 versus placebo andbaseline). The amplitude of fundus pulsationsshowed a small increase, which did not reachstatistical significance (Fig 4).

ISOPROTERENOLThe infusion was stopped at 1 6 ,ug/min ineight subjects and at 3-2 ,ug/min in sevensubjects. In the remaining four subjects allseven infusion steps were administered. Thesystolic blood pressure (p<0005 versusplacebo and baseline) and the pulse pressureamplitude (p<0 005 versus placebo and base-line) increased significantly. The mean arterialpressure and the diastolic blood pressure didnot change. The pulse rate increased signifi-cantly (p<0-005 versus placebo and baseline).The fundus pulsation amplitude increaseddose dependently (p<0 005 versus baselineand placebo, Fig 5).

SODIUM NITROPRUSSIDEAt 4 pug/kg/min the infusion was stopped in sixsubjects, at 8 ,ug/kg/min in eight subjects, andat 16 ,ug/kg/min in five subjects. Bloodpressure decreased during the infusion ofsodium nitroprusside: systolic pressure

(p<0 05 versus placebo and p<0 005 versusbaseline), diastolic pressure (p<0.005 versusbaseline and placebo), and mean arterial pres-sure (p<0005 versus baseline and placebo).Pulse rate increased significantly (NS versusplacebo, p<0 05 versus baseline), whereaspulse pressure amplitude decreased only

= t - X T

Diastolic blood pressure

151050

-5-10-15

151050

-5-10-15

10

L T T-

a)C

al)

.0

E20

a)

0)-c0

0)CDcoC.)

0~

Fundus pulsation amplitude

I_

Pulse pressure

0

-10

Mean arterial pressure

t7 ~~~~........i ... 1

=... .IT

220

-r,.............L

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Effect of isoproterenol, phenylephrine, and sodium nitroprusside on fundus pulsations in healthy volunteers

Systolic blood pressure Pulse rate

0i -0.1 0.2 0.4 0.8

20

10

0

-10

60

40

20

0

Fundus pulsation amplitude

'v.... *

Mean arterial pressure45

30-

15 H

-15 -

0-0 0.1 0.2 0.4 0.8

Isoproterenol (,ug/min)

Figure 5 Dose-response relation (solid line) of changes (%) from baseline measurements ofsystemic haemodynamics and offundus pulsations during saline (0) and stepwise infusion of isoproterenol at doses of 0-1, 0-2, 0 4, and 0-8 ,ig/min. Thebroken line presents the results from the placebo study day. The asterisks indicate significant treatment effects versus placeboas calculated by repeated measureANOVA (p<0 05). Results are presented as means (SEM).

slightly during the infusion steps. Funduspulsations did not show significant changesduring the infusion steps (Fig 6).

IOP MEASUREMENTSThe results of our IOP measurements duringisoproterenol and phenylephrine infusion are

summarised in Table 1. No changes in IOPduring administration of isoproterenol andphenylephrine were observed.

DiscussionOur results show that fundus pulsations innormal volunteers with ametropy of less thanplus or minus 2-5 dioptres are dose dependentlyincreased during infusion of isoproterenol butnot during infusion of phenylephrine or sodiumnitroprusside. Even at higher doses of theperipheral vasoconstrictor and the vasodilatorthe changes in the amplitude of funduspulsations are small.The intrasubject variability of this novel

method is generally small. The within day vari-ability, calculated from the saline infusion studyday, was slightly smaller than the day to dayvariability, calculated from the baseline registra-tions of the 4 study days. This finding was alsoobserved for the haemodynamic variables andargues for slightly different cardiovascular reac-tivity on different study days.The intersubject variability of our measure-

ments is much wider and may be caused byseveral ocular variables. James et a118 observeda significant change of the ocular pulse withaxial eye length and refractive state of the eye.

In our measurements the influence of bulbuslength is probably low owing to the smallrefractive errors of our study group. Moreover,the axial eye length mainly influences thepropagation of the pulse wave through the eye.Intersubject differences in the angioarchitec-ture of the choroid, especially the organisationof the vessels in the submacular region, asreported by Fryczowski et al,19 may have animportant effect on our measurements. Adecrease of the ocular pulse with an increasingheart rate was reported for patients with pace-makers.20 These findings suggest that, at anincrease of the heart rate, there may be a shiftfrom pulsatile to non-pulsatile blood flow inthe ocular circulation.20

During isoproterenol infusion the positiveinotropic effect was responsible for both thedose dependent increase in pulse pressureamplitude and fundus pulsation amplitude. Adirect comparison of the two variables is notpossible as we do not know the relationbetween changes in systemic blood pressureand the perfusion pressure in the ocular ves-sels. Although the presence of 0 adrenergicreceptors in choroidal vessels was recentlyassumed by showing an increased choroidalvascular tone after systemic administration oftimolol maleate,21 it is obvious that our obser-vations mainly depend on the cardiovasculareffects than on the confounding vasoconstrict-ing or vasodilating vascular responses.The mean arterial pressure was significantly

changed following administration of phenyl-ephrine and sodium nitroprusside, whereasboth drugs did not affect pulse pressure ampli-tude. When considering the results after the

*

.,T,,,,, T

Diastolic blood pressure

*

.T1... .: '

151050

-5-10-15

151050

-5-10-15

45

30

15

0

a)c

C,,a).0

E0

a)t-

4-

0)CD

a)0Ca)

0~

I .T I.. ...... ..i I ....

Pulse pressure

L' . .. T ....T

-i- I

0

221

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Schmetterer, Wolzt, Salomon, Rheinberger, Unfried, Zanaschka, Fercher

Systolic blood pressure

20

10

0

-10

60

Pulse rate

40k-

20

0

5

0

-5

-10

Ti.

I.-------'-'-' I---............-1

T.............

0 0.5 1.0 2.0 0 0.5 1.0 2.0

Sodium nitroprusside (,ug/kg/min)Figure 6 Dose-response relation (solid line) of changes (%o) from baseline measurements ofsystemic haemodynamics and offundus pulsations during saline (0) and stepwise infusion ofsodium nitroprusside at doses of 0 5, 1 0, and 2-0 ,ug/kg/min.The broken line presents the results from the placebo study day. The asterisks indicate significant treatment effects versusplacebo as calculated by repeated measure ANOVA (p<0 05). Results are presented as means (SEM).

administration of phenylephrine, the slightreduction of cardiac output and stroke volumeas reported by several authors must be takeninto account.22 23 Hence the lack of effect on

fundus pulsation amplitude may be caused byan increase of pulsatile blood flow comparedwith the total flow or by a local reactionin ocular vessels. The first assumption issupported by the unchanged pulse pressureamplitude in spite of the increased meanarterial pressure during infusion of phenyl-ephrine. Local autoregulative reactions in thechoroid are unlikely,9 whereas the observationthat posterior ciliary arteries spontaneouslydevelop tone24 indicates an autoregulatorycapacity in these prechoroidal arteries. Finally,our results could depend on the change inarterial compliance compared with increasedarterial pressure. In the radial artery there isexperimental evidence that the higher thearterial pressure, the smaller is the arterialcross section change due to a defined changein blood pressure.25 However, the stretchbehaviour of a vessel as large as the radialartery will be somewhat different from that ofthe much smaller choroidal vessels.6

Following the administration of sodiumnitroprusside no change in stroke volume hasbeen observed in healthy subjects.26 However,limited changes in arterial compliance as wellas autoregulation in the posterior ciliaryarteries again cannot be excluded in ourexperimental setup. Owing to the increasedmean arterial pressure the transmural pressureis high and the smooth muscles relaxed. Hencethe elastic behaviour and the tension shouldrely largely on the passive components of thevessel wall.

Phenylephrine, but not sodium nitroprus-side,16 slightly decreased IOP. Reduced IOPcauses a reduced venous pressure and there-fore an increased perfusion pressure in ocularvessels. Yet the transmural pressure changes inchoroidal arteries are not relevantly influencedby this mechanism. The small reduction ofIOP during infusion of isoproterenol, leadingto a decreased venous pressure, is negligible incomparison with the increase in pulse pressureamplitude.Whether absolute measurements of pul-

satile ocular blood flow are feasible with ourinstrument requires further investigation. Itis well known that, as a result of injection ofblood flow in the choroid, the ocular volumeas well as the IOP changes.27 This fact is usedfor the measurement of ocular blood flowwith a pneumatic tonometer.28 Hence, it islikely that the comea, which we use as a refer-ence surface, shows small pulsatile move-ments as well. Estimation of these movementsfrom the Friedenwald equation29 is difficultowing to the different elastic behaviour of thecornea and the sclera.30 We can only assumethat the influence of cornea pulsations issmall, as we observed big differences infundus pulsation amplitudes at variousretinal measurement points, when the corneal

Table 1 Effect of 0-8 ,ug/min isoproterenol (Iso) and of2Agl/kg/min phenylephrine (Phe) onfundus pulsation ampli-tude (FPA in units ofA/2) and IOP (mm Hg) in eighthealthy volunteers. Results are presented as mean (SD).

Saline Iso Saline Phe

lOP 13-1 (2-2) 12-6 (2-5) 12-6 (2-7) 12-0 (2-8)FPA 8-8 (2-1) 13-4 (3-2) 8-9 (2-7) 8-8 (2-3)

*

Diastolic blood pressure

*~~~~~t 1, -r.....:......... ........

a)c

C',CO).0

E0

4J

e0)cu

0

0)CD

C.)00~

151050

-5-10-15

151050

-5-10-15

151050

-5-10-15

Fundus pulsation amplitude

TTI

*~~~~Pulse pressure Mean arterial pressure

1....

222

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from

Effect of isoproterenol, phenylephrine, and sodium nitroprusside on fiundus pulsations in healthy volunteers

reflection always originates from the samepoint.15We conclude that the measurements of

fundus pulsations in healthy volunteers is ahighly reproducible method. The sensitivity ofthe method to detect changes in ocular bloodflow following administration of isoproterenolwas excellent even at the lowest dose.However, in situations where the transmuralpressure is high and the smooth musclesrelaxed or the smooth muscle tension is high,we do not yet know whether changes inchoroidal blood flow can be recorded withour method with satisfactory sensitivity.Nevertheless, measurements of fundus pulsa-tions may prove to be a novel diagnostic toolfor the assessment of choroidal circulation invascular diseases of the eye as high transversalresolution is achieved.The authors thank Mr H Sattmann for constructing theelectronics of the laser interferometer. Excellent support duringthe measurements by H Breiteneder, RN, B Bruckner, RN, AMolnar, RN, and B Monitzer, RN, is acknowledged. Financialsupport was given by the Osterreichische Nationalbank grantNo 4733.

1 Bill A. Blood circulation and fluid dynamics in the eye.Physiol Rev 1975; 55: 383-417.

2 Ahmed J, Braun RD, Dunn R, Linsenmeier RA. Oxygendistribution in the macaque retina. Invest Ophthalmol VisSci 1993; 34: 516-21.

3 Novotny HR, Alvis DL. A method of photographingfluorescence in circulating blood in the human retina.Circulation 1961; 24: 82-6.

4 Richard G. Die klinische Anwendung der Videoangio-graphie der Retina. Klin Monatsbl Augenheilkd 1984; 185:119-22.

5 Alm A, Bill A. The oxygen supply ofthe retina: II. Effects ofhigh intraocular pressure and of increased arterial carbondioxide tension on uveal and retinal blood flow in cats: astudy with radioactively labelled microspheres includingflow determination in brain and some other tissues. ActaPhysiol Scand 1972; 84: 306-19.

6 Alm A, Bill A. Ocular and optic nerve blood flow at normaland increased intraocular pressure in monkeys (Macacairus): a study with radioactively labelled microspheresincluding flow determinations in brain and some othertissues. Exp Eye Res 1973; 15: 15-29.

7 Friedman E. Choroidal blood flow: pressure flow relation-ships. Arch Ophthalmol 1970; 83: 95-9.

8 Yu DY, Alder VA, Cringle SJ, Brown MJ. Choroidal bloodflow measured in the dog eye in vivo and in vitro by localhydrogen clearance polarography: validation of a tech-nique and effect to raised intraocular pressure. Exp EyeRes 1988; 46: 289-303.

9 Kiel JW, Sheperd AP. Autoregulation of choroidal bloodflow in the rabbit. Invest Ophthalmol Vis Sci 1992; 33:2399-410.

10 Caro CG, Pedley TJ, Schroter RC, Seed WA. The mechanicsof circulation. Oxford: Oxford University Press, 1978.

11 Tardy Y, Meister JJ, Perret F, Brunner HR, Arditi M.Non-invasive estimate of the mechanical properties ofperipheral arteries from ultrasonic and photo-plethysmographic measurements. Clin Phys Physiol Meas1991; 12: 39-54.

12 Glucksberg MR, Dunn R. Direct measurement of retinalmicrovascular pressures in the live, anesthetized cat.Microvasc Res 1993; 45: 158-65.

13 Maepea 0. Pressures in the anterior ciliary arteries,choroidal veins and choriocapillaris. Exp Eye Res 1992;54: 731-6.

14 Fine BS. Ocular histology. Hagerstown: Harper and Row,1979.

15 Schmetterer L, Lexer F, Unfried C, Sattmann H, FercherA. Topical measurement of fundus pulsations. Opt Eng1995;34:711-6.

16 Elliott WJ, Karnezis TA, Silverman RA, Geanon J, TripathiRC, Murphy MB. Intraocular pressure increases withfenoldopam, but not nitroprusside, in hypertensivehumans. Clin Pharmacol Ther 1991; 49: 285-93.

17 Nathanson JA. Nitrovasodilators as a new class of ocularhypotensive agents. Y Pharmacol Exp Ther 1992; 260:956-65.

18 James CB, Trew DR, Clark K, Smith SE. Factors influenc-ing the ocular pulse-axial length. Graefes Arch Clin ExpOphthalmol 1991; 229: 341-4.

19 Fryczkowski AW, Sherman MD, Walker J. Observations onthe lobular organization of the human choriocapillaris.Int Ophthalmol 1991; 15: 109-20.

20 Trew DR, James CB, Thomas HL, Sutton R, Smith SE.Factors influencing the ocular pulse - the heart rate.Graefes Arch Clin Exp Ophthalmol 1991; 229: 553-6.

21 Grajewski AL, Ferrari-Dileo G, Feuer WJ, Anderson DR.Beta-adrenergic responsiveness of choroidal vasculature.Ophthalmology 1991; 89: 989-95.

22 Shenker Y, Bates E, Egan B, Hammoud J, Grekin R. Effectof vasopressors on arterial natriuretic factor and hemody-namic function in humans. Hypertension 1988; 12: 20-5.

23 Stamler JS, Loh E, Roddy M, Currie KE, Creager MA.Nitric oxide regulates basal systemic and pulmonary vas-cular resistance in healthy humans. Circulation 1994; 89:2035-40.

24 Nyborg NCB, Nielsen PJ. The level of spontaneous myo-genic tone in isolated human posterior ciliary arteriesdecreases with age. Exp Eye Res 1990; 51: 711-5.

25 Perret F, Mooser V, Hayoz D, Tardy Y, Meister JJ, EtienneJD, et al. Evaluation of arterial compliance-pressurecurves. Hypertension 1991; 18 (Suppl II): 77-83.

26 Merillon JP, Motte G, Aumont MC, Prasquier R, GourgonR. Study of left ventricular pressure-volume relationsduring nitroprusside infusion in human subjects withoutcoronary artery disease. Br HeartJY 1979; 41: 325-30.

27 SilverDM, Farrell RA, Langham ME, O'Brien V, Schilder P.Estimation of pulsatile ocular blood flow from intraocularpressure. Acta Ophthalmol 1989; 67 (Suppl 191): 14-25.

28 Langham ME, Farrell RA, O'Brien V, Silver DM, SchilderP. Blood flow in the human eye. Acta Ophthalmol 1989; 67(Suppl 191): 9-13.

29 Friedenwald JS. Retinal vascular dynamics. Am JOphthalmol 1934; 17: 387-95.

30 Edmund C. Corneal elasticity and ocular rigidity in normaland keratoconic eyes. Acta Ophthalmol 1988; 66: 134-40.

223

on October 19, 2020 by guest. P

rotected by copyright.http://bjo.bm

j.com/

Br J O

phthalmol: first published as 10.1136/bjo.80.3.217 on 1 M

arch 1996. Dow

nloaded from