Cerebral Protective, Metabolic, and Vascular
Effects of Phenytoin
ALAN A. ARTRU, M.D.
37?
AND JOHN D. MICHENFELDER, M.D.
SUMMARY In mice breathing 5 percent oxygen, pretreatment with the optimal dose of 200 mg/kg of
phenytoin increased survival time 123 percent This increase was somewhat less than that observed with certain
barbiturates using the same model but significantly greater than that obsened with diazepam which is more
effective than phenytoin in suppressing hypoxemlc convulsions in this model. In dogs maintained at an expired
halotfaane concentration of either 0.87 percent or < 0.1 percent, phenytoin tended to decrease cerebral blood
flow and had no effect on the cerebral metabolic rate for oxygen at 3 different doses. Assuming a similar effect
in mice, the cerebral protection during hypoxemia observed with phenytoin cannot be explained by a reduction
in metabolic rate, an increase in oxygen delivery, or by an anticonvulsant effect per se. In additional dog
studies, pretreatment with phenytoin decreased the rate of potassium accumulation in cisteraal cerebrospinai
fluid following 20 minutes of anoxia. We speculate that pbenytoin protection may be linked to this effect.
Stroke, Vol 11, No 4, 1980
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
AMONG ITS MANY therapeutic uses, phenytoin
has been reported to protect the brain during periods
of reduced oxygen availability. Using cats or guinea
pigs breathing only nitrogen as a model of cerebral
anoxia, Naiman and Williams1 showed that pretreatment with phenytoin prolonged survival as measured
by the duration of respiratory activity. In a model of
hypoxemia achieved by decompression, Hoff and
Yahn2 demonstrated that phenytoin pretreatment
prolonged survival in mice and rats at any given
hypoxic level and delayed the onset of clinical signs of
hypoxia. Baldy-Moulinier3 used a model of complete
global ischemia produced by aortic clamping to show
that survival time in cats was extended by phenytoin
pretreatment. Finally, in a model of incomplete global
ischemia for 15 minutes achieved via neck tourniquet
in rabbits, Cullen et al.4 found that administration of
phenytoin immediately postischemia reduced cerebral
histopathologic changes.
The present study was designed to examine the
comparative cerebral protective effects of phenytoin
using a mouse model which has already characterized
the protective effects of various barbiturates,6
anesthetic agents,* hypercarbia,7 and hypothermia.8
Employing a dog model, we also examined the
cerebral metabolic and vascular changes associated
with phenytoin administration.
Materials and Methods
Mouse Protection Studies. One hundred fifty male
ARS HA/ICR albino mice (Sprague Dawley,
Madison, WI) weighing 23-36 g were given free access
to food pellets and tap water. Groups of 5 mice were
weighed individually, then injected intraperitoneally
with either phenytoin in the commercial vehicle (25,
50, 100, 200, or 300 mg/kg) or an equal volume of 0.9
From the Department of Anesthesiology, Mayo Medical School
and Mayo Clinic, Rochester, MN. Dr. Artru is Fellow in
Anesthesiology and Dr. Michenfelder is Professor in
Anesthesiology, Mayo Medical School.
Reprints: Dr. Michenfelder, Dept. Anesthesiology, Mayo Clinic,
200 First St., SW, Rochester, MN 55901.
Supported in part by Research Grant NS-7507 from the National
Institutes of Health, Public Health Service.
percent saline (controls).* The injected volume of
phenytoin or saline ranged from 0.23 to 0.72 cc. Twenty minutes later one animal was placed in each of 5 interconnected airtight compartments, breathing room
air supplied at 4 L/min for an additional 10 minutes of
temperature equilibration. The compartments were
kept in a versa-range test chamber (Blue-M Engineering Co., Division of Blue-M Electric Co., Blue Island,
IL) which, at an ambient temperature of 35°C, maintained an intraperitoneal temperature of
36.9 ± 0.2°C. At the onset of each test period the
room air gas supply was replaced by a mixture of
nitrogen (N2) 6-8 Vi L/min and 5 percent oxygen (O2)
in Na 15 L/min. After 60 seconds the N, flow was
reduced to 0-0.5 L/min and after 120 seconds the
O J / N J flow reduced to 3-4 L/min. Since Haldane
analysis of our 5 percent O2 sources showed variation
between tanks (4.76-5.33 percent), gasflowrates were
appropriately adjusted to produce a rapid fall of Ot
concentration in all chambers to 5-5.25 percent (in
line Beckman O3 analyzer) with maintenance of that
concentration thereafter. Results from any chamber
where the O, concentration exceeded the specified
range were discarded. Survival time, defined as the
time from initiation of hypoxic gas flow delivery to
cessation of respiration, was recorded for each animal.
In addition, behavioral observations were made at 30
minutes after drug injection as well as during hypoxia.
Terminal hypoxemic seizure activity was graded either
as present, reduced, or absent.
At least 15 animals were studied at each phenytoin
dose simultaneously with controls. For each group of
5 animals, one to 2 control animals were sequentially
rotated among compartments. Analysis of variance
was done on the entire survival data. Where variance
was confirmed, individual phenytoin doses were compared to controls with Student's /-test for unpaired
•The commercial vehicle for phenytoin is propylene glycol 40 percent and alcohol 10 percent in water buffered to pH 12 with sodium
hydroxide. Previous studies from this laboratory (authors' unpublished data) showed that similar doses of a propylene glycol
vehicle did not significantly alter survival time in the hypoxic mouse
model nor CMRO, in the canine model, and in the canine model it
only briefly altered CBF (increased 10 percent for 5 min).
378
STROKE
samples. A contingency table was constructed for the
hypoxemic seizure data and the chi-square value computed.
Canine Cerebral and Systemic Studies. Subjects were
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
12 unmedicated fasting mongrel dogs (weight 14 to 21
kg). Anesthesia was induced with halothane (1 percent) and nitrous oxide (60-70 percent) in oxygen.
Succinylcholine, 30 mg as a bolus injection, followed
by infusion of 100 mg/h facilitated endotracheal intubation and maintained relaxation. Ventilation was
controlled with a Harvard pump and adjusted along
with the inspired oxygen concentration to maintain
serial blood gases (IL electrodes 37°C) at a Pao2 near
150 mm Hg and a PacOi near 40 mm Hg.
A triple lumen catheter (Type 5F, child, 70 cm,
Instrumentation Laboratory Inc., 113 Hartwell
Avenue, Lexington, MA) was passed through the right
internal jugular vein and its distal end placed in the
pulmonary artery (PA), as verified by presence of the
characteristic wave form. Cardiac output (Q) was
calculated by a dye dilution technique* using a densitometer to measure the concentration of cardiogreen
dye (sterile indocyanine green, Hynson, Westcott, and
Donning, Inc., Baltimore, MD) in continuously
sampled femoral artery blood after injection of the
dye via the distal PA catheter port. A single lumen
cannula was placed in the left femoral vein for fluid
and drug administration. The left femoral artery was
also cannulated for arterial blood gas sampling, buffer
base (BB+) determination, and both continuous
systemic and mean arterial pressure (MAP) monitoring. With the animal in the prone position, the sagittal
sinus was exposed and cannulated as previously
described10 for direct cerebral blood flow (CBF)
measurement and cerebral venous blood gas sampling.
Brain temperature was monitored by a parietal
epidural thermistcr probe and maintained near 37.0°C
with heat lamps. The EEG was monitored via bifrontal electrodes. The cerebral metabolic rate for oxygen
(CMROa) was determined as the product of CBF and
the arterial-cerebral venous (sagittal sinus) blood oxygen content difference. Blood oxygen contents were
determined from measurements of oxygen tension (IL
electrodes, 37°C), and oxygen saturation and
hemoglobin (Hb) (IL 282 CO-oximeter) using 1.39
ml/g as the oxygen carrying capacity of Hb.
In four dogs the expired concentration of halothane
(Beckman LB-2 Medical Gas Analyzer) was lowered
to 0.87 percent and N2 substituted for N2O in the inspired gas mixture. When stable baseline
measurements had been established (at least 20
minutes later), an infusion of phenytoin in the commercial vehicle was begun and the response of the
systemic and cerebral parameters was determined. At
0.87 percent halothane, a fall of mean arterial pressure
(MAP) to near 50 torr was observed when phenytoin
was infused, and in the remaining eight dogs the expired concentration of halothane was lowered to < 0.1
percent and NaO continued in the inspired gas mixture. Again, when stable baseline measurements had
been established, a phenytoin infusion was begun and
the response of the systemic and cerebral parameters
was determined. In both groups phenytoin was infused
VOL 11, No 4, JULY-AUGUST
1980
at a fixed rate of 10 mg/min. A preceding pilot study
showed that a more rapid infusion rate was associated
with the development of a poorly tolerated metabolic
acidosis. At 0.87 percent halothane, 2 dogs received
12.5 mg/kg phenytoin and 2 received 25 mg/kg. At
< 0.1 percent halothane 3 dogs received a total phenytoin dose of 12.5 mg/kg, 3 received 25 mg/kg, and 2
received 50 mg/kg. In all groups, values for cerebral
and systemic parameters were measured during the infusion of phenytoin 12.5 mg/kg at 5, 10, and 20
minutes (end infusion). During the infusion of 25
mg/kg, measurements were done at 5, 10, 20, and 40
minutes (end infusion). During the infusion of 50
mg/kg, measurements were done at 5, 10, 20, 40, 60,
and 80 minutes (end infusion). After the administration of phenytoin had been completed, further
measurements were made at 5 and 10 minutes, then
every 15 to 20 minutes thereafter. In 4 animals, these
values were measured to 180 minutes after end infusion but since no change was noted after 110 minutes,
the remaining animals were observed only for that
time interval. Serum phenytoin levels were drawn at
end infusion and at 110 minutes.
The values for cerebral and systemic parameters
during and following phenytoin administration were
subjected to analysis of variance. Where variance was
confirmed, values obtained during and following
phenytoin were compared to those obtained during the
control period using Student's r-test for paired data.
Canine CSF Electrolyte Studies. In 7 of our 12
animals, the rate of accumulation of CSF K+ after
both hypoxemia and circulatory arrest was determined. This group included 3 dogs from the < 0.1 percent halothane group and 4 from the 0.87 percent
halothane group. Four animals had received 25 mg/kg
phenytoin, 2 animals had received 12.5 mg/kg and one
had received 50 mg/kg. At 110 minutes, these dogs
were placed on N2 (60-70 percent) in oxygen. After a
stable baseline was achieved, the foramen magnum
was cannulated with a No. 17 Touhy needle and 0.5 cc
CSF withdrawn and centrifuged for sodium
([Na+]c8p) and potassium ([K+]csF) determination.
The animals were then made hypoxemic by reducing
the inspired oxygen content to 5 percent. When the
end tidal oxygen concentration reached 5 percent (in
line Beckman O2 analyzer), 10 minutes were allowed
to elapse before another CSF sample was drawn. Circulatory and cerebral arrest were then rapidly
produced by reducing the inspired oxygen to 0 percent.
Ten and 20 minutes after circulatory arrest, CSF
samples were again taken. During this period, brain
temperature was maintained at 37°C by heat lamps.
The changes in CSF electrolytes were compared to
the changes observed in 14 dogs similarly prepared
and monitored that did not receive phenytoin. Seven
of these dogs were anesthetized with 1100 mg/kg yhydroxybutyrate (GHB) and N, (60-70 percent) in
oxygen and 7 were anesthetized with N2O (60-70 percent) in oxygen. In the group which received GHB,
CSF sampling was begun after 110 min of anesthesia,
while in the N,O group sampling was begun after 120
min. Comparison of CSF electrolytes within groups
was made using Student's Mest for paired samples.
CEREBRAL EFFECTS OF PHENYTOIN/zirfrM and Michenfelder
Comparison of cerebral and systemic values, as well
as CSF electrolytes, between groups was made using
Student's Mest for unpaired samples.
Results
Mouse Protection Studies. In the hypoxic mouse
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
model, the mean survival times of the controls at each
phenytoin dose were not significantly different and
were therefore pooled. The resulting mean control
value of 4.1 ±0.1 min did not differ from the mean
control values reported in 2 previous studies that
employed this model: 4.3 ± 0.2 min11 and 3.8 ± 0.2
min.8
Survival time was significantly prolonged in the
groups pretreated with phenytoin 50, 100 and 200
mg/kg when compared to controls (table 1). When
these times were compared to survival time in a group
of mice pretreated with diazepam (7.5 mg/kg), an
effective anticonvulsant in this model, the groups
which received phenytoin 100 or 200 mg/kg showed
significantly prolonged survival time.
Phenytoin significantly altered hypoxic seizure activity (table 2). While doses of <. 100 mg/kg had little
effect, doses of ;> 200 mg/kg attenuated but did not
abolish seizure activity. With phenytoin ^ 100 mg/kg
ataxia was noted. With 200 mg/kg spontaneous activity was reduced. With 300 mg/kg no spontaneous
activity was observed, righting reflex was absent,
respiratory rate was slow, and response to stimulus
was reduced.
Canine Cerebral and Systemic
Studies.
While
the
halothane < 0.1 and 0.87 percent groups differed with
respect to certain control values, the relative cerebral
and systemic effects of phenytoin were similar (tables
3-4, fig.) In both groups these effects were not
different among the 3 phenytoin infusion doses used
(12.5, 25, and 50 mg/kg), and therefore these values
were combined for each reported time interval (end of
phenytoin infusion and 20, 60, and 110 minutes
thereafter). CBF tended to be reduced (25-30 percent)
during the infusion of phenytoin, then to recover in the
first 20 minutes after the phenytoin infusion was con-
379
TABLE 1 Survival Time in Hypoxic Mice
Phenytoin
dose (mg/kg)
Control (n = 55)
25 (n = 15)
50 (n = 15)
100 (n = 30)
200 (n - 20)
300 (n = 15)
Minutes
± SIM
4.1
4.2
7.3
8.4
9.3
3.7
=t o.l
± 0.2
=»> 0.5
± 0.3
± 0.5
* 0.2
Signifi canoe
vi oontrols
Significance
VB Diamepam*
—
—
N.S.
—
V < .001
N.S.
P < .01
V < .001
P < .001
V < .001
N.S.
—
•From a previous study where oontrols did not differ (survival time —
4.3 ± 0.2 min). Pre-treatment with Diaiepam 7.6 mg/kg significantly
prolonged survival to 6.2 ± 0.5 min (p < 0.001)."
TABLE 2 Hypoxic Seizure Activity in Mice
Phenytoin
dose (mg/kg)
Control (n = 55)
25 (n = 15)
50 (n = 15)
100 (n = 30)
200 (n = 20)
300 (n = 15)
Seiiura
present
Number of mioe
Seizures
attentuated
51
4
0
15
14
21
6
1
0
0
1
0
Seizures
absent
8
1
13
1
12
2
For x* analysis, p < 0.01.
eluded. Thereafter CBF was significantly reduced at
60-110 minutes. By comparison, there was no effect
on CMRO, at any time. In the animals at < 0.1 percent halothane, EEG amplitude increased and frequency decreased with phenytoin infusion while no
changes were seen at 0.87 percent halothane. The
serum levels of phenytoin (table 5) correlated appropriately with the dose of phenytoin and the elapsed
time between administration of phenytoin and blood
sampling.
The major systemic effect was an approximate 1/3
reduction in cardiac output (0)- This decrease, which
ranged from 15-50 percent during the study, was accompanied by a reduction in MAP — significant during the infusion period and for 20 to 60 minutes
DPH infusion
FIGURE. The mean values for certain cerebral and systemic parameters are plotted with
respect to time following intravenous infusion
of phenytoin (DPH). These values (representing all 12 dogs) are expressed as percent change
from control, SEM'S are given in tables 3-4.
110
End of
infusion
380
STROKE
VOL 11, N o
4, JULY-AUGUST
1980
TABLE 3 Effect of Phenytoin on Systemic and Cerebral Values at 0.87% Halothane (mean ± SEM), (n •» 4.)
Variable
Control
End of
phenytoin
infusion
20
Minutes
110
Minute*
so
Minutes
Paoi (mm Hg)
Pacoj (mm Hg)
pH
BB+ (mEq/L)
Hb (g/dl)
MAP (mm Hg)
Epidural temp (°C)
Q (L/min/m 1 )
PssOi (mm Hg)
CBF (ml/min/100 gm)
CMRO. (ml/min/100 gm)
EEG Amp. (iiW)
EEG Freq. (Hz)
143 ± 8
41 ± 1
7.33 ±
42 ±
14.7 ±
83 ±
36.9 ±
3.23 ±
57 ±
0.01
1
0.9
6
0.1
0.72
2
80 =b 4
3.94 ± 0.14
35 ± 2
11 ± 1
146 ±
36 ±
7.33 ±
40 ±
15.1 ±
51 ±
37.0 ±
2.32 ±
50 ±
57 ±
3.76 ±
35 ±
10 ±
11
1
0.01
1
0.6
4*
0.1
0.66*
1*
5*
0.11
2
1
142
39
7.30
39
14.7
57
37.0
2.73
54
*
±
±
±
±
*
±
±
±
9
1
0.01
1
0.6
6*
0.0
0.66*
2
67
3.56
35
11
±
±
±
*
8
0.08
2
1
139
42
7.28
40
14.6
67
36.9
±
±
±
±
±
±
±
9
2
0.03
1
0.6
5*
0.0
—
57 ± 2
69 ± 2
3.67 ± 0.09
—
141 ± 9
41 ± 9
7.29 ± 0.01
40 ± 1
14.9 ± 0.7
69 ± 7
37.0 ± 0.0
2.16 ± 0.48*
54 ± 1
57 ± 4*
3.66 ± 0.08
35 ± 2
11 ± 1
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
•Significant difference from oontrol. p <0.0fi.
Pasos — sagittal ainua blood oxygen tension.
TABLE 4 Effect of Phenytoin on Systemic and Cerebral Values at < 0.1% Halothane {mean ^ BEM), (n = 5)
Variable
Control
End of
pbenytoin
infuEon
20
Minutes
Paoj (mm Hg)
Pacoj (mm Hg)
160 ± 5
pH
BB+ (mEq/L)
Hb (g/dl)
MAP (mm Hg)
Epidural temp (°C)
Q (L/min/m 1 )
PssOi (mm Hg)
CBF (ml/min/100 gm)
CMRO, (ml/min/100 gm)
EEG Amp. (^V)
EEG Freq. (Hz)
110
Minutes
60
Minutes
150 ± 5*
147 ± 6*
146 ± 8
142 ± 8*
41 =fa 1
39 * 1*
41 ± 1
38 ± 1
39 ± 0
7.33 ± 0.01
7.28 ± 0.03*
7.25 ± 0.03*
7.25 ± 0.02*
42 ± 1
38 ± 1*
37 ± 1*
36 ± 1*
36 ± 0*
17.0 ± 0.5
17.8 ± 0.6
17.3 ± 0.7
16.9 <* 0.8
16.4 ± 0.7
125 =* 6
104 * 8*
95 ± 11*
7.25 ± 0.01*
114 * 6
113 <*> 5
37.1 ± 0.1
37.0 ± 0.0
37.0 ± 0.0
37.0 ± 0.1
3.51 ± 0.78
2.34 ± 0.85
2.11 * 0.37
—
37.0 ± 0.1
65 ± 3
57 ± 4
60 ± 5
53 •*. 3*
49 ± 4*
58 ± 7*
1.77 ± 0.43*
105 * 15
79 ± 15
84 ± 13
66 ± 9*
4.91 ± 0.27
5.00 ± 0.34
4.84 ± 0.39
5.36 ± 0.32
5.53 ± 0.32
11 ± 0
16 ± 1*
16 ± 1*
—
16 ± 2*
14 ± 0
13 ± 1
12 ± 1*
—
13 ± 1*
•Significant difference from oontroL p < O.OA.
P a o i — sagittal sinus blood oxygen tension.
thereafter. The development of a moderate metabolic
acidosis was also observed, while heart rate (not
tabulated) did not change.
shown in table 7. While some differences did exist,
Pao,, Hb, MAP, and CBF were adequate, and Pacoj
and temperature normal in all groups. [Na+]CsF (not
Canine CSF Electrolyte Studies. Prior to inducing
hypoxemia/anoxia the [K + ] C S F was the same in
phenytoin treated dogs as in dogs not given phenytoin
and anesthetized with either GHB or N,O (table 6).
Cerebral and systemic values for these 3 groups are
TABLE 6 [K+ksr mEq/L ± SEM, S7°C
Time of
sampling
Control
TABLE 5
Serum
PhenyUnn Levels (jig/ml
± 8EM)
Phenytoin dose
(mg/kg)
End of
phenytoin infusion
100
Minutes
12.5
14.6 ± 2.0
16.3 ± 1.9
42.3 ± 11.0
11.7 *>• 0.9
25
50
6.0 =t 0.7
20.1 •>• 2.1
10 minutes at
circulatory arrest
20 minutes at
circulatory arrest
Phenytoin
Ni,Oi
(n - 7)
3.06
*0.17
4.21
±0.39*
5.29
±0.26*
GKB
Ni,O.
(n - 7)
NiO/Oi
(n - 7)
3.45
*0.08
4.94
±0.20*
7.16
±0.43*t
3.32
±0.20
4.05
±0.09*
6.64
*0.96*
•Significant difference from oontrol (paired (-test), p < 0.05.
tSignificant difference from Phenytoin (unpaired {-test), p < 0.05.
CEREBRAL EFFECTS OF PHENYTOIN/^r/ru and Michenfelder
TABLE 7 Cerebral and Systemic Values Prior to Inducing
Hypoxemia/Anoxia (mean =•= SEM). (n = 7 for each group)
Variable
PaOj (mm Hg)
Pacoi (mm Hg)
pH
BB+ (mEq/L)
Hb (g/dl)
MAP (mm Hg)
HR (beats/min)
T°(°C)
0 (L/min/m1)
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
CBF,
(ml/100 g/min)
CMRO,
(ml/100 g/min)
EEG Amp. (^V)
EEG Freq. (Hz)
Phenytoin
140 ± 5
40 ± 1
7.27
38
15.6
85
143
37
1.99
Group
GHB
166
39
=t 0.01 7.20
35
± 1
± 0.7 16.9
± 9
111
± 16
135
± 0
37
± 0.32 2.19
59 ± 7
NiO
150 *
± 13
41 ±
± 1
± 0.02 7.22 ±
=*• 4
35 ±
± 1.2 16.8 ±
± 14
110 ±
181 *
± 13
37 ±
± 0
**
± 0.17
39 ± 4
8
1
0.03
1*
0.6
4*
13
0
71 ± 7
4.23 ± 0.31 3.47 =*= 0.32 4.45 ± 0.28
26 ± 4
27 ± 4
15 ± 1*
12 ± 1
8 ± 0*
15 ± 2
•Significant differenoe from phenytoin, p <O.Ofi.
**Not measured.
tabulated) was higher in dogs given GHB (administered as the sodium salt in large doses). After 10
minutes of hypoxemia (5 percent O,), minimal
changes in [K + ] C S F were observed in the 3 groups
(not tabulated). After circulatory arrest, [K+]CSF
progressively increased at both 10 and 20 minutes
(table 6). After 10 minutes this increase was similar in
all 3 groups, but after 20 minutes the increase was
significantly less in the phenytoin treated animals (75
percent above control) than in the groups that did not
receive phenytoin (GHB = 108 percent above control,
NjO = 100 percent above control). [Na + ] C8F did not
change in any of the groups with either hypoxia or
anoxia.
Discussion
Using the hypoxic mouse model a number of drugs
(e.g. thiopental," pentobarbital," mephobarbital,11 and
diazepam8) and conditions (e.g. hypothermia8 and
hypercarbia7) have been shown to prolong survival.
The cerebral protective effects of each of these drugs
or conditions has been explained either by cerebral
metabolic depression,11 increased cerebral blood flow,7
suppression of convulsions,11 or by a combination of
such effects. Based upon pur results these mechanisms
do not conveniently explain the protective effects of
phenytoin. Assuming no major species differences,
phenytoin has no effect on CMRO3, tends to decrease
CBF, and is less effective in suppressing terminal convulsions than diazepam.
The lack of a cerebral metabolic effect in the dog
seems convincing for a number of reasons. There were
no metabolic alterations observed at any of 3 different
doses of phenytoin given at either anesthetic or subanesthetic concentrations of halothane. In the latter
group, phenytoin did induce modest EEG changes but
381
without a measurable metabolic component. Based
upon the behavioral response of our mice to protective
doses of phenytoin, it seems unlikely that major
cerebral metabolic changes were produced in that
species either. At the optimum protective dose (200
mg/kg), behavioral changes consisted only of mild
ataxia with some decrease in spontaneous activity. At
doses sufficient to suppress the righting reflex and
response to stimuli (300 mg/kg) no protection was
seen — presumably due to systemic toxic effects.
While we found no evidence of a CMRO2 depressant effect with phenytoin, some evidence favoring a
depressant effect has been reported. Broddle and
Nelson13 measured the rates of high energy phosphate
and glucose utilization in mouse brain following
decapitation. They reported that pretreatment with
200 mg/kg phenytoin reduced these utilization rates
by 40 and 60 percent respectively while pretreatment
with 20 mg/kg phenytoin produced no change. Using
a rat brain synaptosome-mitrochondrial preparation
Spector14 reported that the addition of phenytoin to
the incubation media reduced O, consumption 20 to
50 percent. The in vivo applicability of metabolic
measurements obtained in these ways remains unknown; lacking further data it seems reasonable to
conclude that phenytoin protection is likely not based
upon cerebral metabolic depression.
The possibility that phenytoin might protect the
brain by increasing CBF was not supported by our
canine studies. During drug infusion CBF consistently
fell simultaneous with reductions in both MAP and Q.
After the infusion was completed, CBF initially
returned toward control but then again decreased with
time. The rate of decline in CBF (10-12 percent/h) we
observed exceeds that previously reported for this
model in a group of dogs similarly prepared and
anesthetized with halothane <0.1 percent and N,O
(60 to 70 percent) in O2." Thus, while the decline in
CBF may represent primarily an effect of time only on
the preparation, a possible drug effect is suggested as
well. Previous studies of phenytoin's effect on CBF
have reported no uniform increase to above normal
levels. In man, Kennedy et al.18 reported that phenytoin returned the reduced CBF of epileptic children to
normal levels. In cats subjected to a period of global
ischemia, Baldy-Moulinier3 reported a persistent
decreased CBF unaltered by phenytoin. In rats,
Kennedy et al.17 used 14C-antipyrine autoradiography
and reported areas of both increased and decreased
CBF following doses of phenytoin ranging from 10 to
100 mg/kg. All of the evidence supports the conclusion that phenytoin does not increase O t delivery to
the brain and protection must be explained by some
other mechanisms.
Although phenytoin is an effective anticonvulsant, it
was only marginally effective in suppressing the terminal hypoxemic convulsions observed in the mouse
model. A 50 mg/kg dose provided cerebral protection
yet did not alter hypoxemic convulsions. The 100 and
200 mg/kg doses provided a greater degree of protection than diazepam (which also does not reduce
CMRO,18) yet was clearly less effective than diazepam
STROKE
382
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
in suppressing hypoxemic convulsive activity. We
therefore conclude that although suppression of convulsions might contribute, this cannot be the major
mechanism by which phenytoin protects the hypoxemic mouse.
Another suggested mechanism for phenytoin
protection relates to possible effects on ion flux across
membranes. CSF K+ may increase due to failure of
active transport of K+ from CSF to blood or from
release of cerebral intracellular K + . Accumulation of
CSF K+ may aggravate the effects of reduced O2
delivery and thereby decrease the brain's tolerance to
hypoxemia/anoxia.3-1921 Phenytoin has been reported
to reduce ion flux during both cerebral hypoxemia/anoxia3> 22 and normoxia.23"28 It may do so by
enhancing the ATP dependent Na-K pump2428 and/or
by a direct membrane effect.23- "• 28 In any case, compared to our untreated anesthetized dogs, phenytoin
treatment resulted in a slower rate of accumulation of
cisternal [K+]CSF during anoxia. Reduced CSF K+ accumulation is not universally associated with all
agents which provide cerebral protection. Pentobarbital 33 mg/kg, which provides equivalent cerebral
protection to phenytoin 50 mg/kg in the hypoxic
mouse, has no effect on CSF K+ accumulation while
hypothermia to 35°C, also equiprotective to phenytoin 50 mg/kg, minimally reduces CSF K+ accumulation.*
The degree to which the sampled cisternal [K + ] C SF
reflects either intracellular K+ release or failure of
outward transport depends on a variety of factors —
the rate of CSF production and absorption, the size of
the compartment containing the CSF, and the percent
of total CSF volume taken with each sample, to name
a few. With the sampling technique we used, detection of any effect of phenytoin on either rate or magnitude of K+ release would have been delayed and
blunted. Thus we observed no [K+]CSF changes during
10 minutes of hypoxia, and only after 20 minutes of
anoxia were differences in the phenytoin treated animals apparent. In view of the lack of other obvious
mechanisms to explain phenytoin protection, it seems
reasonable to conclude that suppression of K+ accumulation may be linked to the protective effect.
References
1. Naiman JG, Williams HL: Effects of diphenylhydantoin on the
duration of respiratory activity during anoxia. J Pharmacol Exp
Ther 145: 34-41, 1964
2. Hoff EC, Yahn C: The effect of sodium 5, 5-Diphenylhydantoin
(Dilantin Sodium) upon the tolerance of rats and mice to
decompression. Am J Physiol 141: 7-16, 1944
3. Baldy-Moulinier M: Cerebral blood flow and membrane ionic
pump. Cerebral blood flow and intracranial pressure. Proc 5th
Int. Symp., Roma-Sienna 1971, Part I, EurNeurol 6: 107-113,
'Authors' unpublished data.
VOL 11, No 4, JULY-AUGUST
1980
1971/72
4. Cullen JP, Aldrete JA, Jankovsky L, et al: Protective action of
phenytoin in cerebral ischemic hypoxia. Anesth Analg 58:
165-169, 1979
5. Wilhjelm BJ, Jacobsen E: The protective action of different barbituric acid derivatives against anoxia in mice. Acta Pharmacol
et Toxicol 28: 203-208, 1970
6. Wilhjelm BJ, Arnfred I: Protective action of some anaesthetics
against anoxia. Acta Pharmacol et Toxicol 22: 93-98, 1965
7. Secher O, Wilhjelm B: The protective action of anesthetics
against hypoxia. Can Anaes Soc J 15: 423^*40, 1968
8. Steen PA, Michenfelder JD: Barbiturate protection in tolerant
and non-tolerant hypoxic mice: Comparison to hypothermic
protection. Anesthesiology 50: 404-408, 1979
9. Theye RA, Rehder K, Quesada RS, et al: Measurement of cardiac output by an indicator — dilution method. Anesthesiology
25: 71-74, 1964
10. Michenfelder JD, Messick JM, Theye RA: Simultaneous
cerebral blood flow measured by direct and indirect methods. J
Surg Res 8: 475-481, 1968
11. Steen PA, Michenfelder JD: Cerebral protection with barbiturates, relation to anesthetic effect. Stroke 9: 140-142, 1978
12. Michenfelder JD, Theye RA: Cerebral protection by thiopental
during hypoxia. Anesthesiology 39: 510-513, 1973
13. Broddle W, Nelson SR: The effect of diphenylhydantoin on
energy reserve levels in brain. Fed Proc 27: 751, 1968
14. Spector RG: Effects of formyl tetrahydrofolic acid and
noradrenaline on the oxygen consumption of rat brain
synaptosome-mitrochondrial preparations. Br J Pharm 44:
279-285, 1972
15. Takashita M, Michenfelder JD, Theye RA: The effects of
morphine and N-allylnormorphine on canine cerebral
metabolism and circulation. Anesthesiology 37: 605-712, 1972
16. Kennedy C, Anderson W, Sokoloff L: Cerebral blood flow in
epileptic children during the interseizure period. Neurology
(Minneap) 8: 100-105, 1958
17. Kennedy C, Grave GD, Jehle JW, et al: The effects of
diphenylhydantoin on local cerebral blood flow. Neurology
(Minneap) 22: 451-452, 1972
18. Carlsson C, Hagerdal M, Kaasik AE: The effects of diazepam
on cerebral blood flow and oxygen consumption in fats and its
synergistic interaction with nitrous oxide. Anesthesiology 45:
319-325, 1976
19. Bering EA: Effects of profound hypothermia and circulatory
arrest on cerebral oxygen metabolism and cerebrospinal fluid
electrolyte composition in dogs. J Neurosurg 40: 199*205, 1974
20. Meyer JS, Kanda T, Sinohara Y, et al: Effects of anoxia on
cerebrospinal fluid sodium and potassium concentrations.
Neurology (Minneap) 21: 889-895, 1971
21. Wade JG, Amtorp O, Sorensen SC: No-flow state following
cerebral ischemia. Arch Neurol 32: 381-384, 1975
22. Pincus JH, Grove I, Marino BB: Studies on the mechanism of
action of diphenylhydantoin. Arch Neurol 22: 566-571, 1970
23. Crane P, Swanson P: Diphenylhydantoin and the cations and
phosphates of electrically stimulated brain slices. Neurology
(Minneap) 20: 1119-1123, 1970
24. Escueta AV, Appel SH: Brain synapses: An in vitro model for
the study of seizures. Arch Intern Med 129: 333-344, 1972
25. Fertziger AP, Liuzzi SE, Dunham PB: Diphenylhydantoin:
Stimulation of potassium influx in lobster axons. Brain Res 33:
592-596, 1971
26. Festoff BW, Appel SH: Effect of diphenylhydantoin on
synaptosome sodium-potassium-ATPase. J Clin Invest 47:
2752-2758, 1968
27. Nasello AG, Montini EE, Astrada CA: Effects of veratrine,
tetraethylammonium, and diphenylhydantoin on potassium
release by rat hippocampus. Pharmacology 7: 89-95, 1972
28. Pincus JH: Diphenylhydantoin and ion flux in lobster nerve.
Arch Neurol 26: 4-10, 1972
Cerebral protective, metabolic, and vascular effects of phenytoin.
A A Artru and J D Michenfelder
Stroke. 1980;11:377-382
doi: 10.1161/01.STR.11.4.377
Downloaded from http://stroke.ahajournals.org/ by guest on June 14, 2017
Stroke is published by the American Heart Association, 7272 Greenville Avenue, Dallas, TX 75231
Copyright © 1980 American Heart Association, Inc. All rights reserved.
Print ISSN: 0039-2499. Online ISSN: 1524-4628
The online version of this article, along with updated information and services, is located on the
World Wide Web at:
http://stroke.ahajournals.org/content/11/4/377
Permissions: Requests for permissions to reproduce figures, tables, or portions of articles originally published in
Stroke can be obtained via RightsLink, a service of the Copyright Clearance Center, not the Editorial Office.
Once the online version of the published article for which permission is being requested is located, click Request
Permissions in the middle column of the Web page under Services. Further information about this process is
available in the Permissions and Rights Question and Answer document.
Reprints: Information about reprints can be found online at:
http://www.lww.com/reprints
Subscriptions: Information about subscribing to Stroke is online at:
http://stroke.ahajournals.org//subscriptions/