Abstract
purpose. To determine the effect of several common general anesthetics on
intraocular pressure (IOP) after experimental aqueous outflow
obstruction in the rat.
methods. A single episcleral vein injection of hypertonic saline was used to
sclerose aqueous humor outflow pathways and produce elevated IOP in
Brown Norway rats. Animals were housed in either standard lighting or a
constant low-level light environment. Awake IOPs were determined using
a TonoPen (Mentor, Norwell, MA) immediately before induction of
anesthesia by either isoflurane, ketamine, or a mixture of injectable
anesthetics (xylazine, ketamine, and acepromazine). For each
anesthetic, IOPs were measured immediately after adequate sedation
(time 0) and at 5-minute intervals, up to 20 minutes.
results. Awake IOPs ranged from 18 to 52 mm Hg. All anesthetics resulted in a
statistically significant (P < 0.01) reduction in
measured IOP at every duration of anesthesia when compared with the
corresponding awake IOP. With increasing duration of anesthesia,
measured IOP decreased approximately linearly for both the anesthetic
mixture and isoflurane. However, with ketamine, IOP declined to 48% ±
11% (standard lighting) and 60% ± 7% (constant light) of awake
levels at 5 minutes of anesthesia, where it remained stable. In fellow
eyes, the SD of the mean IOP in animals under anesthesia was always
greater than the corresponding SD of the awake mean. Anesthesia’s
effects in normal eyes and eyes with elevated IOP were
indistinguishable.
conclusions. All anesthetics resulted in rapid and substantial decreases in IOP in
all eyes and increased the interanimal variability in IOPs. Measurement
of IOP in awake animals provides the most accurate documentation of
pressure histories for rat glaucoma model
studies.
Many risk factors can influence glaucoma. Of these, elevated IOP
is the best recognized and documented. Nearly all glaucoma therapy
relies on lowering IOP. Understanding the mechanisms by which IOP
damages optic nerve fibers is important for developing new, logical
glaucoma treatments designed to protect the optic nerve directly.
Fortunately, optic neuropathy similar to that occurring in human
glaucoma can be produced in otherwise normal animals by experimental
IOP elevation.
1 2 3 4 5 Such models offer the ability to
reproduce effects solely due to elevated IOP and allow identification
of early effects of elevated IOP, before extensive nerve damage. In
addition, the affected retinas and optic nerves can be evaluated
directly to obtain unequivocal answers regarding the extent of optic
nerve damage and accompanying cellular alterations.
Due to expense and logistic considerations, there has been an increased
interest in producing such models in laboratory rats.
2 3 4 5 Successful use of such models requires careful correlation of the
extent of IOP elevation with the resultant injury and accurate
determination of the IOP responsible for the optic nerve damage it
induces.
The introduction of the TonoPen tonometer (Mentor, Norwell, MA) made it
possible to measure IOP in the rat eye without direct, anterior chamber
manometry. This instrument provides an accurate, reproducible,
noninvasive assessment of IOP in the rat eye.
6 7 In
addition, when used on awake Brown Norway rats, the TonoPen allows
documentation of a natural circadian fluctuation of IOP of
approximately 10 mm Hg, which is exaggerated during the dark phase of
the circadian cycle after obstruction of aqueous humor
outflow.
8 9 These fluctuations can be stabilized by
housing the animals in constant low-level light, thereby simplifying
the documentation of IOP exposure history.
10
In spite of these observations, many laboratories measure IOP in rats
with the aid of general anesthetics, particularly when using strains
that are less tolerant of awake measurements than the Brown Norway rat.
However, frequent use of general anesthetics may not be well tolerated
by experimental animals,
7 and commonly used anesthetic
agents, such as ketamine, may have neuroprotective properties,
confounding experimental results.
11 12 13 14 Finally, various
anesthetic agents may alter IOP in humans and several experimental
animals,
7 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 and the effects of these agents on eyes
with experimentally elevated IOP are not well understood.
In this study, we have combined the ability to measure IOP in awake
Brown Norway rats with our model of aqueous outflow
obstruction.
2 The purpose was to determine the effect of
several commonly used general anesthetics on measured IOP in the rat
eye, because they would most likely be used in a routine laboratory
setting, and learn the extent of their effect on eyes with
experimentally elevated IOP.
All experiments complied with the ARVO Statement for the Use of
Animals in Ophthalmic and Vision Research. Twenty-three male Brown
Norway rats (Rattus norvegicus), weighing 300 to 400 g,
were used in this study.
All animals were housed initially in standard lighting with food and
water provided ad libitum and the room temperature maintained at
21°C. The room was lit by fluorescent lights (330 lux) that were
turned on and off automatically every 12 hours. The animals were
weighed weekly to monitor their general health.
A TonoPen XL tonometer was used to measure awake IOP, with one drop of
0.5% proparacaine hydrochloride applied to each eye, as described
previously.
8 Briefly, the mean of 10 valid IOP readings
was obtained from each experimental and fellow (control) eye for each
awake or anesthetized IOP measurement. All animals were acclimated to
daily handling for 1 week, and normal awake IOPs were confirmed before
episcleral injection.
Initially, one eye of each animal was injected with 50 μl of a 1.75-M
hypertonic saline solution through an episcleral vein, as described
previously.
2 After the injection, animals were housed in
either standard lighting conditions, as described, (12 animals, 12
injected eyes), or a constant low-level (40–90 lux) light environment
(11 animals, 16 injected eyes). In five constant-light animals, the
fellow eye was subsequently injected, so that more information
regarding eyes with elevated IOP could be obtained from this group.
Awake IOP was determined in both eyes daily, from the first day after
injection throughout the experiment.
Three different anesthetics were evaluated in every animal:
isoflurane, ketamine, and a mixture of injectable anesthetics. This
mixture consisted of a solution of 5 ml ketamine (100 mg/ml), 2.5 ml
xylazine (20 mg/ml), 1 ml acepromazine (10 mg/ml), and 1.5 ml sterile
water. Every animal was evaluated with all three anesthetic agents.
Anesthesia and IOP measurements were performed in the following manner.
For isoflurane, animals were anesthetized by placing them in a shoebox
with 3 maximum alveolar concentration (MAC) isoflurane (Forane;
Anaquest, Madison, WI), and 0.5 l/min oxygen from an anesthesia cart
(Ohmeda 8000; BOC Health Care, West Yorkshire, UK).
7 After
sedation, animals were transferred to a respirator, and isoflurane was
continually supplied at 2.5 MAC, the minimum required to maintain
anesthesia. Ketamine (Sanofi Winthrop, New York, NY) and the anesthetic
mixture were administered by intraperitoneal injection at a dose of 1.0
ml/kg (100 mg/ml) and 1.0 ml/kg, respectively.
Awake IOPs were measured immediately before induction of anesthesia.
The first postanesthesia IOP reading (time 0) was obtained as soon as
the animals reached a depth of anesthesia determined by the absence of
pain and palebral reflexes. Time from the administration of anesthesia
to time 0 varied, but subsequent measurements were obtained at 5-minute
intervals up to 20 minutes. IOP measurements in awake and anesthetized
animals were performed in both eyes by the same examiner. All IOP
measurements in anesthetized rats were obtained with identical animal
handling and without restraint or eyelid manipulation.
For each measurement time point, average IOPs were determined for
both the experimental and control eyes. Statistical analyses were
performed by computer (Excel 97; Microsoft, Redmond, WA, and Prism
2.01; GraphPad, San Diego, CA, statistical software packages).
Awake IOPs in eyes with normal and experimentally elevated
pressure ranged from 18 to 52 mm Hg. For animals in both lighting
conditions and at every duration of anesthesia, all anesthetics
produced a statistically significant (
P < 0.01)
reduction in measured IOP, compared with the corresponding awake IOP.
The IOP data (in millimeters of mercury) for uninjected fellow eyes are
illustrated in
Figure 1 . The findings for eyes with experimentally elevated IOP are shown in
Figure 2 , with values expressed as a percentage of the initial, awake IOP.
Awake IOPs in uninjected normal Brown Norway rat eyes were
approximately 20 and 29 mm Hg for rats housed in standard and constant
light conditions, respectively. For animals in standard housing, all
anesthetics lowered the measured IOP rapidly and dramatically to
pressures of between 9 and 11 mm Hg within 5 minutes
(Fig. 1A) . The IOP
reduction in constant light–exposed eyes was similar
(Fig. 1B) ,
although slightly less dramatic. In both lighting conditions, IOP after
isoflurane and the anesthetic mixture continued to decrease throughout
the period of observation. For ketamine, the IOP decrease appeared to
stabilize at approximately 50% and 60% of the awake IOP level for
animals housed in standard and constant lighting conditions,
respectively.
Figure 2 presents the effects of anesthetics on eyes with
experimentally elevated IOP for animals housed in both lighting
conditions. As with normal eyes, IOP reduction after isoflurane and the
anesthetic mixture progressed throughout the experiment. After
ketamine, IOP stabilized after 5 minutes at approximately 50% and 60%
of the awake IOP in animals housed in standard lighting and constant
light, respectively. For each of the three anesthetics, the pattern of
anesthesia-induced IOP reduction over time was unaffected by the
initial level of the awake IOP.
We used the observed 52% and 40% reductions in IOP (for standard
and constant light, respectively) after 5 minutes of ketamine to
extrapolate an estimated “awake” IOP from the measured IOPs.
Figure 3 illustrates the correlation between actual awake IOP and these
extrapolated pressures for all eyes in both lighting groups. The
R 2s for these correlations were 0.61
and 0.67 for the standard lighting and constant light conditions,
respectively. The average difference between the actual measured IOP
and the extrapolated awake IOP was 4.3 mm Hg. The impact of this on
measurement variability is best seen among the 12 uninjected, fellow
eyes in standard lighting. Whereas the mean actual awake IOP for this
group was 20.0 ± 0.50 mm Hg, the extrapolated mean was 21.4 ± 5.1 mm Hg, representing a 10-fold increase in the SD.
The reliability of animal models for understanding the mechanism
of pressure-induced optic nerve damage relies heavily on our ability to
measure accurately the level of IOP to which the eye and optic nerve
are exposed. Previous studies strongly suggest that many general
anesthetics can lower IOP in humans,
15 16 17 18 19 20 21 22 monkeys,
23 24 25 26 27 28 29 30 dogs,
31 32 33 cats,
24 34 and rabbits.
24 35 Others report
that some agents, such as ketamine, can elevate IOP.
36 37 38
The present study provides a detailed, direct comparison between awake
and anesthetized IOP measurements in rat eyes and encompasses a wide
range of pressures after experimental outflow obstruction. The Brown
Norway rat is well suited for measuring awake IOP (using only topical
0.5% proparacaine HCl), in part because of its docile nature and
moderately prominent globes.
The present study demonstrates that all tested general anesthetics
significantly and dramatically reduced the measured IOP both in normal
eyes and in those with aqueous humor outflow obstruction in the rat.
The reduction in IOP with isoflurane and the anesthetic mixture was
relatively linear with time, whereas that with ketamine stabilized
after the first 5 minutes. Housing the animals in constant low-level
light, which minimizes large circadian IOP oscillations in the rat,
diminished the magnitude of the IOP reduction, but did not change the
overall characteristics of the response to anesthetics.
The use of general anesthetics resulted in a significant lowering of
IOP and a large increase in the SD in these measurements, suggesting
variable animal response to the anesthesia, even though all IOPs were
measured in a consistent manner by the same individual. In addition,
pressure measurements with both isoflurane and the anesthetic mixture
continued to decline over time. Both of these observations are contrary
to what would be expected if actual IOP were “uncovered” by the
anesthetics. These considerations, as well as the fact that
physiologic, circadian IOP oscillations can be measured in awake
animals,
8 argue that awake IOP, rather than IOP under
anesthesia, accurately reflects the actual IOP experienced by these
eyes.
Because IOP measured under ketamine stabilized after 5 minutes, its use
would appear to be the most reliable in situations in which an
anesthetic agent is unavoidable. However, the increased variability in
IOP induced by anesthetics limits this strategy, as shown in
Figure 3 .
Although the correlation between actual awake IOP and estimated awake
IOP extrapolated from pressures determined under ketamine was linear,
it is only marginally accurate. This is dramatically illustrated by the
group of normal, uninjected eyes in standard lighting, in which the
measured actual awake IOP ranged from 19 to 21 mm Hg. In this group of
eyes, there was a 10-fold increase in the SD of estimated awake IOPs
when extrapolated from pressures obtained in animals under ketamine
anesthesia, resulting in a range of 16 to 33 mm Hg. Similarly, if all
the animals with an estimated awake IOP of approximately 40 mm Hg are
considered, they corresponded to a very broad range in actual awake IOP
from 28 to 48 mm Hg.
Through careful comparison of IOP level and optic nerve damage, we have
determined that damage is linearly correlated with IOP in the range of
30 to 40 mm Hg.
10 That IOP measurement under anesthesia
produces estimated IOPs that can vary both above and below this range
strongly suggests that the use of anesthetics for measuring IOP rats
does not reliably represent the true IOP created in response to
experimental aqueous humor outflow obstruction.
In our experience, Brown Norway rats are easily acclimated to awake IOP
measurement, which is rapid and can be performed as frequently as
necessary. Avoiding general anesthesia eliminates the interanimal
variability in response to anesthetics, making the measurement of IOP
more reliable and accurate. It also minimizes concerns that anesthetics
may mask normal circadian fluctuations in IOP and it avoids any
confounding neuroprotective effects of the anesthetics
themselves.
11 12 13 14 We have found that measuring awake IOP
in Brown Norway rats provides the most reliable method for determining
IOP both in normal eyes and in those with experimentally elevated
pressures.