Measurement of Aqueous Flow in Rabbits With Corneal and

Document Sample
Measurement of Aqueous Flow in Rabbits With Corneal and Powered By Docstoc
					        Measurement of Aqueous Flow in Rabbits With
        Corneal and Vitreous Depots of Fluorescent Dye
                                           Gerald R. Gaul and Richard F. Brubaker

            Using pigmented rabbits, twofluorophotometricmethods for measuring aqueous flow were compared.
            A corneal depot of dye was used to measure aqueous flow in one eye of each rabbit while the vitreous
            depot method of Johnson and Maurice was used to measure aqueous flow over the same period in the
            fellow eye. After intravenous administration of acetazolamide or mannitol, during the first hour after
            drug delivery a decrease in aqueousflowwas observed in the eye with the corneal depot of dye that was
            undetected in the eye with the vitreous depot of dye. In rabbits which were water-loaded by orogastric
            tube, an increase in aqueous flow was observed by the corneal depot method, while the vitreous depot
            method indicated a decrease in aqueous flow. The vitreous depot method depends on the assumption
            that the rate of loss of dye from the vitreous remains stable. It is shown that movement of water into
            or out of the vitreous can cause large changes in the rate of movement of dye from the vitreous to the
            anterior chamber and can make interpretation of the vitreous method ambiguous. While the vitreous
            depot method is probably superior for measuring sustained changes of the rate of aqueous flow over
            tens of hours or days, it is concluded that it cannot be used for measuring changes over shorter periods.
            Invest Ophthalmol Vis Sci 27:1331-1335, 1986

   Johnson and Maurice have recently described an                         fluorescein-labeled dextran, we studied the effect of
ingenious and simple method of measuring changes in                       movement of vitreous water on the movement of vit-
the rate of flow of aqueous humor through the cham-                       reous dye. We show that the results of the vitreous
bers of the rabbit eye.1 Their method consists of mea-                    depot method are ambiguous.
suring the concentration of fluorescein-labeled dextran
in the anterior chamber many days after intravitreal                                     Materials and Methods
injection of the tracer. The vitreous depot method for                    Experiment 1
measuring flow in rabbits has several advantages: flu-                       This experiment was performed to determine the
orescence need only be measured in the anterior                           precision of the corneal depot method and the vitreous
chamber, the method can be used to compare flow for                       depot method of measuring flow. Four pigmented rab-
several days with only one injection of tracer, and the                   bits were studied. A corneal depot of fluorescein was
method eliminates concern about diffusional losses of                     employed to measure the rate of flow in each eye for
tracer.                                                                   8 hr on two separate days. Twelve hours before the
    In this study of rabbits, we compared the vitreous                    measurement began, two drops of 10% sodium fluo-
depot method with a corneal depot method devised in                       rescein (Funduscein; CooperVision Pharmaceuticals,
its original form by Jones and Maurice.2 Aqueous flow                     Carolina, Puerto Rico) were instilled into each eye
was calculated before and after intravenous acetazol-                     twice. The rate of flow was measured as described in
amide, intravenous mannitol, and orogastric water-                        the subsequent section entitled "Corneal Depot
loading. We used both methods in each rabbit. The                         Method." The results of measurements of flow per-
control value of aqueous flow in the eye studied by                       mitted comparisons of the right to the left eye and
corneal depot method was used to calculate the rate of                    comparisons of rates of flow measured on two separate
loss of dye from the fellow vitreous-injected eye.                        occasions.
   In rabbits that had received intravitreal injections of                   The same rabbits were studied using the vitreous
                                                                          depot method. Ten n\ of a 10% solution of fluorescein
                                                                          isothiocyanate-labeled dextran (average molecular
   From the Department of Ophthalmology, Mayo Clinic and Foun-            weight 70,000 Daltons) was injected into the mid-vit-
dation, Rochester, Minnesota.                                             reous of both eyes of each rabbit with a Hamilton
  This study was supported by NIH Grant EY 00634, Research to             (Reno, NV) syringe and a 27-gauge needle. The injec-
Prevent Blindness, Inc., and the Mayo Foundation.
  Submitted for publication: March 27, 1985.                              tion site was just anterior to the equator. No signs of
   Reprint requests: Richard F. Brubaker, MD, Department of Oph-          ocular inflammation were observed in any of the in-
thalmology, Mayo Clinic, Rochester, MN 55905.                             jected eyes. Gradually, the bright fluorescent depot dif-

1332                    INVESTIGATIVE OPHTHALMOLOGY & VISUAL SCIENCE / September 1986                                         Vol. 27

fused in all directions until the vitreous appeared uni-    injection. Just before sacrifice, the concentration of
formly stained. One week after injection, the fluores-      fluorescein-labeled dextran was measured in vivo in
cence of fluorescein-labeled dextran in the anterior        the anterior chamber of each eye by fluorophotometry.
chamber of each eye was measured for 8 hr.                  After sacrifice, the contents of the chambers of the eyes
                                                            were aspirated with a tuberculin syringe and a 30-gauge
Experiment 2                                                needle. The volume and the concentration of fluores-
                                                            cein-labeled dextran in the aspirate was measured. Fol-
    This experiment was performed to determine the
                                                            lowing aspiration, the anterior chamber reforms. A
sensitivity of the two methods in detecting changes of
                                                            second aspiration was performed 30 min and a third
flow of aqueous humor. Eight pigmented rabbits of
                                                            one 90 min after the first aspiration. The concentration
either sex weighing 2-4 kg each were studied. One eye
                                                            of fluorescein-labeled dextran, the volume of each as-
of each rabbit was injected with fluorescein-labeled
                                                            pirate, and the interval between aspirations permitted
dextran as described in experiment 1. After 1-2 weeks,
                                                            calculation of the rate of efflux of fluorescein-labeled
the uninjected fellow eye was treated topically with
                                                            dextran from the vitreous into the posterior chamber.
two drops at 5-min intervals with 10% sodium fluo-
rescein (Funduscein; CooperVision Pharmaceuticals,
Carolina, Puerto Rico). Five minutes after the second                      Fluorophotometric Methods
drop, the cornea and the conjunctival cul-de-sac were            Corneal Depot Method: The rate of loss of fluorescein
irrigated to remove fluorescein from the surface of the     from the cornea was measured for each eye by mea-
eye. Twelve hours after topical application of fluores-     suring the concentration of fluorescein in the stroma,
cein to one eye, the intensity of fluorescence in the       c c , every hour and determining the rate of disappear-
anterior chamber of each eye and the intensity of flu-      ance of fluorescein from the stroma, dm c /dt. The rate
orescence in the stroma of the topically treated eye        of loss of fluorescein from the anterior chamber, d r n j
were measured with a slit-lamp fluorophotometer.3           dt was calculated from the concentration of fluores-
These measurements were repeated seven times at in-         cein in the anterior chamber, c a . To determine the rate
tervals of 1 hr. Measurements were made in unsedated        of loss of fluorescein, it wasassumed that the volume
animals comfortably restrained in a canvas bag atop a       of the corneal stroma was 70 ii\ and the volume of the
special holder. The holder permitted the examiner to        anterior chamber was 250 /*1.4
make measurements of either eye without assistance               Since the rate of loss of fluorescein from the eye is
and without touching the animal.                            equal to the clearance of fluorescein multiplied by the
   After the fourth measurement, the rate of flow of        concentration in the anterior chamber, then
aqueous humor through the anterior chamber was al-
tered in one of three ways:                                               clearance • ca = dm c /dt +             dmjdt

   1) By injection of 50 mg/kg acetazolamide into a           This equation is equivalent to equation 14 of Jones
marginal ear vein (eight animals).                          and Maurice:2
   2) By injection of 2 gm/kg mannitol as a 20% so-                            dm t
lution into a marginal ear vein (eight animals).                                df
   3) By delivery of 50 ml/kg water into the stomach           It has been shown that approximately 90% of the
via an orogastric tube (six animals).                       clearance of fluorescein from the anterior chamber of
   The rate of flow of aqueous humor was calculated         the normal eye is due to flow and 10% of the clearance
for each eye before and after treatment according to        due to diffusion.5
the methods described in the section below entitled                       flow s 0.9 • (dm c /dt + dma/dt)/c a
"Fluorophotometric Methods."
                                                            The mean concentration of tracer in the anterior
Experiment 3                                                chamber during an hour, c a , was determined by av-
                                                            eraging the concentration at the beginning and end of
   This experiment was performed to determine if the        the hour.*
rate of entry of fluorescein-labeled dextran into the an-
terior chamber is always diffusion limited as is assumed
in the vitreous depot method. Fluorescein-labeled dex-
tran was injected into the vitreous of both eyes of eight      * The average concentration in an interval c a , can be determined
                                                            either by averaging the concentrations at the start and end of the
rabbits according to the procedure described in exper-      interval, (ca( 1) + ca(2))/2, or ca can be determined from the exponential
iment 1. Half of these rabbits were sacrificed 10-12        formula ca = (c a (l) - ca(2))/(ln (ca( 1 )/ca(2))). For intervals of 1 hr or
days after injection, and the other half 23-24 days after   less the two methods produce almost identical values of c a .
No. 9                     INTRAVITREAL FLUORESCEIN AND AQUEOUS FLOW /        GQUI    and Brubaker                              1333

   Vitreous Depot Method: It was assumed, on the time        Table 1. Flow measured by corneal depot method
scale of these experiments, that the rate of loss of flu-
                                                                                      Day 1                           Day 2
orescein-labeled dextran from the vitreous into the
posterior chamber was constant.                                                          Rate of aqueous flow (nl/min)

                    dmv                                      Rabbit         REye              LEye          REye              LEye
                          = constant
                    ~dT                                        1              1.86            2.00           1.82              1.67
                                                               2              1.88            2.11           1.34              1.87
   If a steady-state is assumed to exist, then the rate of     3              1.73            1.54           1.86              1.56
loss of dextran from the anterior chamber is also con-         4              2.09            2.30           1.95              1.88
stant and equal to the rate of entry into the aqueous        Mean             1.89            1.99           1.74              1.75
humor from the vitreous:
               ca • f = constant =
                                     ~dT                        Table 2 compares the measurement of flow in one
                                                             eye by means of a corneal depot to measurement of
where ca = concentration of dextran in the aqueous
                                                             changes of flow by means of a vitreous depot. The rate
humor, and f = rate of flow of aqueous humor.
                                                             offlowwas calculated for every hour of the experiment.
   In experiment 2, we assumed that the rates of flow
                                                             The mean rate of flow in the eyes labeled with an in-
in the two eyes are equal during the control period
                                                             trastromal fluorophore was 1.92 jd/min. We were un-
(hours 0-3). The concentration of dextran was mea-
                                                             able to demonstrate time-dependency of flow over the
sured in one eye and the rate offlowof aqueous humor
                                                             7 hr. The standard deviation of the means of the hourly
was measured by the corneal depot method in the other.
                                                             measurements in these animals was 9% of the mean
Thus, the rate of entry of dextran into the aqueous
                                                             of all the hours. The fluorescence of fluorescein-labeled
humor, dmv/dt, can be calculated for the control pe-
                                                             dextran in the anterior chamber of the intravitreally
riod. The rate of entry of dextran was assumed to be
                                                             injected eyes demonstrated very little scatter. Since
constant throughout the remainder of the experiment.
                                                             aqueous flow is assumed to be inversely proportional
The rate of aqueous flow of the dextran-injected eye
                                                             to anterior chamber fluorescence with this method, the
was calculated from the steady-state equation of John-
                                                             method produces very little scatter of calculated
son and Maurice:
                                                             aqueous flow values from hour to hour. The standard
                                                             deviation of fluorescence measurement from hour to
                        dt /
                                                             hour with this method was 2% of the mean of all the
                                                                The calculated rate of flow derived from the corneal
Statistical Methods                                          method depends on two measurements, the stromal
    All statistical comparisons were made with a two-        fluorescence and the cameral fluorescence. Of the two,
sided t-test for paired samples. A software package en-
titled STATPAC, version 52, copyright by David S.
Walonick, 1985, was employed.                                Table 2. Variability of the measurements
    All studies were done in accordance with the ARVO        in untreated animals
Resolution on the Use of Animals in Research.
                                                                                                             Vitreous depot
                                                                                 Corneal depot                  method
                        Results                                                    method
                                                                                                            Anterior chamber
   Table 1 is a summary of the rates of flow in the two                        Rate of aqueous*               fluorescence
                                                                 Time              flow ± SE              (equivalent ng/ml of
eyes of four rabbits, measured on two occasions with              (hr)          (fil/min, n = 8)           fluorescein, n = 8)
the aid of a stromal depot of fluorescein. The mean
rate of aqueous humor flow was 1.84 ± 0.24 (mean                 0-1                 1.90 ±.23                620.7   ± 40.8
                                                                  1-2                2.06 ± .25               595.9   ± 32.4
± SD) /il/min in these animals. The standard deviation           2-3                 1.91 ±.11                613.5   ±35.2
of the differences between the right and left eyes was           3-4                 2.10 ±.21                597.9   ± 30.3
0.28 /il/min (15% of the mean), and the standard de-             4-5                 1.54 ±.18                617.1   ±26.2
                                                                 5-6                 1.94 ±.12                619.9   ±26.9
viation of the differences between the determinations            6-7                 2.02 ±.14                605.0   ±31.8
on the two separate occasions was 0.19 /xl/min (10%              Mean                1.92 ±.07                610.0   ± 3.9
of the mean). In three animals, flows were lower on            * Rows at hours 3-4, 4-5, 5-6, or 6-7 are not significantly different
the second occasion, and in one the flow was higher.         (P > .05) fromflowbetween hours 0-3 ("control period").
1334                              INVESTIGATIVE OPHTHALMOLOGY 6 VISUAL SCIENCE / Seprember 1986                                                         Vol. 27

Table 3. Acetazolamide 50 mg/kg at hour 3                                          Table 5. Water 60 ml/kg at hour 3
                                 Rate of aqueousflow± SE                                                           Rate of aqueousflow± SE
                                      (\il/min, n = 8)                                                                  (nl/min, n = 8)
       Time               Corneal depot                  Vitreous depot                  Time               Corneal depot                 Vitreous depot
       (hr)                 method                         method                        (hr)                 method                         method

       0-1                    .86 ± .30                     1.69   ±.19                   0-1                   1.95 ±.10                   1.81 ±.10
       1-2                    .59 ± .25                     1.55   ± .22                  1-2                   1.81 ±.29                   1.84 ±.10
       2-3                    .81 ±.20                      1.57   ±.10                   2-3                   1.68 ±.09                   1.79 ±.09
       3-4                  ().91 ±.13                       .44   ± .20                  3-4                  2.24 ± .09                   1.67 ±.08
                            (P=.001)                                                                           (P = .030)                   (P= .016)
       4-5                    .71 ±.27                      .42 ± . 1 5                   4-5                  1.58 ±.24                    1.61 ±.10
       5-6                    .93 ± .32                     .29 ± .24                                                                       (P = .004)
       6-7                    .51 ± 1.9                     .46 ± .20                     5-6                  1.61 ±.31                    1.59±.ll
                                                                                                                                            (P - .004)
  Probability that difference in flow at hours 3-4, 4-5, 5-6, 6-7 compared to             6-7                  1.47 ±.24                    1.68 ±.12
control period (hours 1-3) was due to chance is indicated in parentheses if less                                                            (P = .025)
Ihnn OOS
                                                                                      Probability that difference in flow at hours 3-4, 4-5, 5-6, 6-7 compared to
                                                                                   control period (hours 1-3) was due to chance is indicated in parentheses if less
                                                                                   than 0.05.
the stromal measurement is often difficult to perform,
especially in unsedated animals. The vitreous method
depends on the measurement of cameral fluorescence                                    Table 4 summarizes observations made after intra-
only. The vitreous method, however, as will be shown                               venous administration of 2 gm/kg mannitol 3 hr after
later, can produce ambiguous results following the ad-                             beginning measurements of fluorescence. A 50% re-
ministration of drugs having known effects on flow.                                duction of aqueous humorflow(P = .007) was observed
The problem seems to arise as a result of application                              during hour 3-4 in the eyes studied by means of a
of a steady-state equation when steady conditions do                               corneal depot which we attribute to the known effects
not exist.                                                                         of the drug. The flow at hour 6-7 was significantly
   Table 3 summarizes the effects of acetazolamide 50                              reduced in the same eyes by 13% (P = 0.036) for reasons
mg/kg given intravenously 3 hr after the first mea-                                which are not clear to us. In the fellow eyes studied by
surement offluorescence.A 50% decrease of the cal-                                 the vitreous depot method, the flow was not signifi-
culated rate of flow (P = .001) was observed during                                cantly different at hours 3-4 and 5-6, was increased
hour 3-4 in the eyes in which a corneal depot was                                   14% (P = 0.012) at hour 4-5, and decreased at hour
employed. The calculated rate of flow diminished                                   6-7 by \6%{P = 0.009).
slightly in the intravitreally injected eye after admin-                              The effect of water loading, 60 ml/kg, at hour 3 is
istration of acetazolamide, but the reduction was not                              summarized in Table 5. The calculated rate of flow
statistically significant for any subsequent hour. If hours                        through the anterior chamber increased 33% (P
3-7 together are compared to hours 0-3 by this method,                             = 0.030) in the eyes with the corneal depot between
a 13% reduction of flow (P = .034) is detectable.                                  hours 3-4. A paradoxial reduction of the calculated
                                                                                   rate offlowwas observed in the fellow eyes studied by
                                                                                   the vitreous method immediately following the water
Table 4. Mannitol 2 gm/kg at hour 3
                                                                                   loading (P = 0.016). In both eyes, the averageflowwas
                                 Rate of aqueousflow± SE                           lower between hours 4-7 (corneal method, P = 0.018;
                                        (fil/min, n •= 8)                          vitreous method, P = 0.004).
      Time               Corneal depot                 Vitreous depot                 These results led us to suspect that transient changes
       (hr)                 method                         method                  of intraocular pressure or of solvent flow between the
                                                                                   posterior chamber and the vitreous might effect the
                           1.85 ± . 2 7                  1.79      ±.13
       1-2                 1.63 ± . 1 7                  1.65      ±.18            rate of transfer offluorescein-labeleddextran from the
       2-3                 1.68 ± . 1 2                  1.80      ±.19            vitreous into the posterior chamber. Consequently, ex-
       3-4                 0.84 ± .34                    1.48      ±.15            periment 3, outlined under Materials and Methods,
                           (P = .007)
       4-5                 1.78 ± . 0 7                  2.00      ±.10            was conducted.
                                                         (P=       .012)              Table 6 summarizes the effect of sacrificing the an-
       5-6                 1.80 ± . 1 8                  1.75      ±.17            imal and draining the anterior and posterior chambers
       6-7                 1.50 ± . 1 3                  1.47      ±.19
                           (P = .036)                    (P =      .009)           on the rate of tracer efflux from the vitreous. We ob-
                                                                                   served a large increase in the rate of efflux of the flu-
   Probability that difference in flow at hours 3-4, 4-5, 5-6, 6-7 compared to
control period (hours 1-3) was due to chance is indicated in parentheses if less
                                                                                   orescent tracer, in the 10-12-day group by a factor of
than 0.05.                                                                          10 and in the 23-24-day group by a factor of 20. The
No. 9                    INTRAVITREAL FLUORESCEIN AND AQUEOUS FLOW / Goul and Drubaker                                             1335

results indicate that the watery fluid of the vitreous can    Table 6. Rate of movement of fluorescein-labeled
flow into the posterior chamber, bringing with it large       dextran from vitreous to chambers of
amounts of the labeled dextran. If the increase in the        rabbit eye (ng/hr, n = 8)
rate of efflux is proportional to the change of intraocular
                                                                                           10-12 days               23-24 days
pressure, even small changes of intraocular pressure                                      after injection          after injection
could interfere with measurements of flow for several              Hours ajter
hours.                                                               death                 Mean        SE            Mean     SE

                                                                  In vivo                  34.1*   ±    2.8          7.7*   ± A
                       Discussion                                 Aspirate, 0.0            39.1*   ±   10.2         10.0*   ± 1.9
                                                                            0.5           408.9    ±   38.5        200.4    ±39.9
   The applications and limitations of corneal depots                       1.5           165.5    ±   18.6         98.1    ±22.5
of tracer have been explored over years of use by many                       3             97.1    ±   15.4         69.1    ± 13.3
investigators. Many techniques are possible for staining        * Rate of aqueous humor flow just prior to sacrifice assumed to be 1.8 p\l
the cornea with tracer without significantly disturbing       min (Table 1).
the integrity of the eye. Corneal depot techniques per-
mit calculation of aqueous flow over a period of one
                                                              flow occurs more quickly or is not sustained, the con-
to several hours. The technique we have employed re-
                                                              centration of dye in the anterior chamber will not have
quires accurate measurements of fluorescence of a uni-
                                                              reached its steady state. In this case, the peak change
formly stained cornea. This procedure is simple to per-
                                                              in the rate of aqueous flow will be underestimated.
form in human subjects, but tricky to accomplish in
                                                              Miichi and Nagataki circumvented this problem with
rabbits, especially with a slit-lamp fluorophotometer,
                                                              a similar technique by predicting the steady-state from
as employed in this experiment. The variability of flows
                                                              the rate of change of ca during the transient.7 In prin-
measured by this technique in rabbits is evidence of
                                                              ciple, their mathematical technique could be applied
the difficulty of making this measurement accurately.
                                                              to the vitreous depot method. However, in Miichi and
   The applications and limitations of the vitreous de-       Nagataki's method, the tracer was infused at a constant
pot techniques are still not entirely clear. Its greatest     rate into the posterior chamber through a 27-gauge
advantages are its simplicity, its independence from          needle, and, thus, its rate of entry was known.
diffusional loss of tracer, and its capability of measuring
                                                                 The paradoxical effects demonstrated by this study
flow over many days with a single depot. The technique
                                                              of drugs having well-known effects on aqueous humor
depends on a steady rate of loss of dye both from the
                                                              flow demonstrate that the vitreous depot method can-
vitreous and from the aqueous. As long as the eye is
                                                              not be used uncritically.
not disturbed, the loss of dye from the vitreous into
the aqueous is demonstrably steady.                           Key words: fluorophotometry, rabbit eye, flow of aqueous
   When intraocular pressure is lowered, displacement         humor, intraocular pressure
of vitreous water causes increased amounts of the tracer
to enter the aqueous circulation. In the absence of a                                      References
change in the rate of aqueous formation, the concen-           1. Johnson F and Maurice D: A simple method of measuring
tration of fluorescein-labelled dextran in the anterior           aqueous humor flow with intravitreal fluoresceinated dextrans.
chamber increases. This change is indistinguishable               Exp Eye Res 39:791, 1984.
from what is expected to happen from suppression of            2. Jones RF and Maurice DM: New methods of measuring the rate
                                                                  of aqueous flow in man with fluorescein. Exp Eye Res 5:208,
the rate of aqueous formation. Recent work of Maurice
indicates that movement of aqueous water into the vit-         3. Brubaker RF and Coakes RL: Use of a xenon flash tube as the
reous compartment decreases the amount of tracer that             excitation source in a new slit-lamp fluorophotometer. Am J
enters the aqueous circulation,6 so the converse is also          Ophthalmol 86:474, 1978.
true. Maurice's recent work indicates that this property       4. Davis FA: The eye and orbit of the rabbit. Trans Am Ophthalmol
of dye movement can be exploited to measure the rate              Soc 27:401, 1929.
                                                               5. Goldmann H: Abflussdruck, Minutenvolumen und Widerstand
of water flow in the vitreous.                                    der Kammerwasserstromung des Menschen. Doc Ophthalmol
   We believe that the steady-state equation employed             56:278, 1951.
in the vitreous depot method correctly indicates the           6. Maurice D: Flow of water across aqueous-vitreous interface.
direction and magnitude of a change of aqueous flow               ARVO Abstracts. Invest Ophthalmol Vis Sci 26(Suppl):107,
through the anterior chamber only if the rate of               7. Miichi H and Nagataki S: Effects of pilocarpine, salbutamol, and
aqueous flow, as well as the intraocular pressure, has            timolol on aqueous humor formation in cynomolgus monkeys.
been stable for several hours. If the change in aqueous           Invest Ophthalmol Vis Sci 24:1269, 1983.

Shared By:
Tags: flow, water
Description: Typically, the flow of water in the system is very easy to do. If you eat only one meal a day flow of food or drink, you can in 8 months, lose 10 pounds. Flow of water should be diversified, so the lack of nutrition. Under the guidance of a doctor, or even two meals a day flow of food. This can be in five weeks lose 10 pounds. But to ensure that the selected flow of food to provide nutrients and protein your body needs, and to ensure three meals a day.