Ep~~e~sy~es.,9(1991)11-18 Elsevier
11
EPIRES 00403
Intraperitoneal
phenytoin suppresses kindled responses: effects on motor and electrographic seizures
E.W. Lothman, J.M. Williamson and K.E. VanLandingham” Department of Neurology,
University of Virginia Health Sciences Center, Charlottesville, VA (U.S.A.)
(Received 27 June 1990; revision received 13 January 1991; accepted 16 January 1991) Key words: Phenytoin; Limbic seizures; Complex partial seizures; Pharmacokinetics;
Antiepileptic drug
C~ntrove~y has arisen about the effectiveness of phenytoin against kindled seizures. It has been suggested that the reports of ineffectiveness could be accounted for by phenytoin being given by an intraperitoneal (i.p.) route in those experiments so that adequate serum concentrations were not achieved. Another possibility for the different results was dissimilar stimulus protocols employed in the various studies. The present study examined these issues. Doses of i.p. phenytoin were studied for their actions against kindled responses elicited with short (1 s) and long (10 s) stimulus trains through hippocampal electrodes. Serial application of the stimuli determined time-action relationships. Dose-dependent effects were demonstrated for ah time points examined. There was a consistently greater suppression of kindled motor seizures than limbic behavioral seizures or electrographic seizures. Phenytoin either totally blocked or did not affect the duration of afterdischarges. Actions of phenytoin against responses by short duration stimuli were greater than against long duration stimuli. Additional pharmacokinetic studies compared i.p. versus intravenous (i.v.) phenytoin. After i.p. phenytoin, serum levels peaked later than after iv. delivery, but were maintained in the ‘therapeutic’ range longer. The present experiments provide additional support for the idea that kindled seizures are a useful model for complex partial seizures in humans. In addition, they show that major actions of phenytoin are to decrease seizure spread and to elevate afterdischarge thresholds and that the i.p. route is appropriate for assaying the effect of phenytoin against kindled seizures in rats.
INTRODUCTION Repetition of electrical stimulus trains at various sites in the limbic system leads to a progressive enhancement of epileptiform responses until stereotyped generalized convulsions appear7*8. Even after extended periods with no stimulation, the con* Present address: Dept. Neurology, University of ~ichig~, Ann Arbor, MI 48109, U.S.A. Correspondence to: E.W. Lothman, Dept. Neurology Box 394, University of Virginia Health Sciences Center, Charlottesville, VA 22908, U.S.A.
0920-1211/91/$03.50 @ 1991 Elsevier Science Publishers B.V.
vulsions are again elicited when the stimulus is reinstituted. Thus, the condition of enhanced responsiveness, referred to as a kindled state, is enduring. The dynamic period, when responses demonstrate increasing severity, is referred to as kindling. Kindling and seizures elicited in the kindled state are regarded as powerful models of complex partial seizures with generalization7,14. Morphological changes are found in the brains of kindled animals”‘, comparable to those encountered in patients with complex partial epilepsy”. Moreover, the sensitivity of kindled seizures to a battery of
12
antiepileptic drugs agrees well with the sensitivity of seizures in humans with complex partial epilepsy to these agents11~12~14. This agreement between the set of antiepileptic drugs effective against kindled responses with the set of drugs effective against complex partial seizures in humans supports the appropriateness and utility of kindled responses as a model system. However, prior reports on the effectiveness of phenytoin against kindled responses have presented problems for the concordance of drug effects between the animal model and the clinical condition of complex partial seizures 1-5,11,12,16,19. In particular, many investigators have reported phenytoin to be ineffective against limbic afterdischarges elicited in kindled animals. There is also controversy as to how effective the drug is against kindled motor seizures. Phenytoin is well-recognized for its efficacy in suppressing complex partial seizures in patients13. Recently, McNamara et al.” suggested that the observations that phenytoin was ineffective against kindled responses were a consequence of the drug being given intraperitoneally (i.p.) in those experiments, instead of intravenously (i.v.) as they had done. If so, then the ‘disparity’ of the effectiveness of phenytoin observed in kindled animals compared to humans could be related solely to pharmacokinetic factors. We have developed a stimulus protocol to test the effect of various drugs against kindled responses at multiple time points after a single drug injection, thereby giving a time-action profile in a single experimental session”. This protocol employs 50 Hz, 10 set trains at a stimulus intensity supramaximal to afterdischarge threshold. Using this protocol, we found that i.p. phenytoin did not affect kindled limbic afterdischarges over a period of i/2 to 6 h after drug delivery’l. The stimulus protocol employed by McNamara and his colleagues15 to demonstrate an effect of phenytoin against kindled limbic afterdischarges used trains of comparable (60 Hz) frequency, but shorter (1 s) duration and a stimulus intensity equal to the generalized seizure threshold (GST). Therefore, another explanation for the dissimilar results is the different stimulus protocols used in the various studies. The present experiments examined two related
issues. In the first set of experiments we determined the influence of i.p. phenytoin against kindled responses elicited in rats by 2 different types of stimulus train. One type of train resembled that used by McNamara et al.“, and the other was like the one we had previously usedl’. This study tested whether different stimulus parameters could account for the dissimilar results of previous studies. In the second set of experiments, we compared the pharmacokinetics of i.p. versus i.v. phenytoin in rats. This study examined the possibility that i.p. phenytoin was ineffective in achieving adequate blood levels to suppress seizures. METHODS Adult male (300-325 g) Sprague-Dawley rats were prepared according to methods detailed elsewhere’0,“,20. In brief, bipolar electrodes were implanted in the ventral hippocampus. One week later, the animals were kindled with 50 Hz, 10 set trains of 400 ,uA, 1 msec biphasic pulses until stable, fully kindled responses were established. Following completion of experiments, electrode placements were confirmed histologically. After kindling, animals were studied with a modified ‘mixed trial’ protocol*‘, consisting of two kinds of stimuli, selected for features of the different test protocols used in previous studies and introduced above. Both types of stimuli consisted of 50 Hz trains of 1 msec biphasic pulses. One kind of train replicated the type of stimuli used in our previous work and was 10 s long and 4OOpA in intensity, supramaximal to the afterdischarge threshold”. The other kind of train simulated that used by other workers3,12’15,including McNamara and co-workers, and was 1 set long, twice afterdischarge threshold. The use of these stimulus parameters allowed stimulus current and intratrain frequency to be matched across the 2 types of stimuli (see below). For convenience the first type of stimulus will be referred to hereafter as ‘long duration’ and the latter type of stimulus will be referred to as ‘short duration’. Each test day began with a long duration stimulus and the two types of stimuli were alternated as follows. Stable, kindled responses are triggered
13
when long duration stimuli are given 30 min after the preceding seizure l”,il. In preliminary experiments it was determined that the period after the preceding seizure had to be increased to 60 min to provide stable, serial kindled responses to short duration stimuli. Accordingly, in our protocol, half an hour elapsed after the previous seizure with long duration stimuli, and 1 h with short duration stimuli. Each test period began with long duration trains. In order to provide a 30 min postdrug test for short duration stimuli, the order of long duration and short duration stimuli was reversed after the first two pairs of responses (see Fig. 1 below). Animals were repeatedly tested with this protocol, with at least 1 stimulus-free day between each testing session. At the beginning of each test day afterdischarge thresholds for both the 50 Hz, 10 set and for the 50 Hz, 1 set trains were determined. Only animals that had stable afterdischarge thresholds for both types of stimuli, and showed only kindled motor (stages 4 and 5) seizures and consistent afterdischarge durations (~10% variance) to each of the 10 stimuli on 3 serial test days before drug study and on at least 1 test day after drug study, were used in the drug study. Phenytoin (25 (n = 5), 37.5 (n = 5), and 50 (n = 4) mg/kg) was injected i.p. after 2 pairs of responses had established that the animals under study were consistently expressing kindled responses on that test day. Drug test days were at least 1 week apart with drug-free test days interposed as described above. The drug was given 30 min before the first postdrug response was obtained. Phenytoin was prepared in a vehicle of 70% distilled water/30% polyethylene glycol as described elsewhere”. Pharmacokinetic comparisons between i.p. and i.v. phenytoin were carried out with a separate group of animals. Attention was focused on the 50 mg/kg dose for the following reasons. As will be detailed below (see Results), this dose was shown to suppress afterdischarges to short duration stimuli, but not to affect afterdischarges to long duration stimuli. Hence, the responses to this single dose reproduced the different results cited in the Introduction. In addition, 50 mg/kg was the dose studied by McNamara and co-workers” in their comparisons of phenytoin blood levels after i.p.
and i.v. administrations. For the pharmacokinetic study, one subgroup of rats received placement of intravascular lines in their femoral arteries and femoral veins using methods described elsewhere’. Catheterization was done under halothane anesthesia 24 h before the following experiment. These animals then received 50 mg/kg phenytoin through the femoral vein catheter. In accord with clinical practice, the delivery was carried out over a 30 min period. Another subgroup of animals, with catheters placed only in their femoral arteries, received 50 mg/kg i.p. phenytoin. In both subgroups 250 ~1 blood samples were withdrawn via the arterial catheter at multiple time points after drug administration. Serum concentrations of phenytoin for the various samples were then determined by the Clinical Toxicology Laboratory at the University of Virginia by means of an enzyme immunoassay. Severity of behavioral seizures was assessed with a conventional 5-point behavioral seizure scoring system and electrographic seizures were quantified by afterdischarge durations”. Statistical comparisons of behavioral seizure scores were carried out with Mann-Whitney tests. Statistical comparisons among afterdischarge durations were carried out with analysis of variance (ANOVA) of repeated measures and Neuman-Keuls post-hoc tests. Statistical comparisons of phenytoin levels were done with t-tests. Level of significance was set at P < 0.05 throughout. RESULTS Effects of phenytoin on kindled responses to short duration and long duration stimulus trains
Representative results are shown in Fig. 1, where responses of behavioral seizures and of electrographic seizures elicited by both short duration and long duration stimuli are charted. Absolute current intensities were matched for the two types of trains (450 f 87 ,uA for short duration and 400 f 0 ,BA for long duration trains, not statistically different; t-test). Relative to afterdischarge thresholds (45 f 5 PA for long duration and 255 + 43 ,uA for short duration trains), stimulus intensities were supramaximal (approx. 10 times threshold) for long duration trains while being supra-
14 Injection
phenytorn
50
mg/kg
!** 0
Short (1
2
1
A
set)
Long (10
train duration
xc)
:
1
duration
train
p*
I
4,” 120
I / i**
i**
l** li*
r
100
r i
-
z
_m
so 60 40
t 1 Hr
Fig. 1. Comparison of effects of phenytoin on kindled responses triggered by short duration and long duration trains. Rats with stable, fully kindled responses were tested with 5 pairs of short duration (circles) and long duration (triangles) trains as described in the text. After the first two pairs, the order of delivery was reversed and 50 mg/kg i.p. phenytoin was administered. Thirty minutes later, response testing was resumed. Top panel displays behavioral seizure scores (BSS) and lower panel displays afterdischarge durations (ADD, in set), both plotted as means f SEM. Significant differences between BSS responses in each pair (i.e. short versus long duration train comparison) indicated by filled stars (Mann-Whitney tests; *t, P < 0.01; symbols next to short duration response of each pair; responses compared for each pair connected by line); significant differences between first 3 predrug responses or 6 postdrug responses when compared to lastpredrug response (also Mann-Whitney tests) indicated by asterisks (**P < 0.01). There was a significant difference across the 10 set ADD responses(ANOVA of repeated measures, F, 38.5; P < 0.001). Differences between responses in each test pair for ADD indicated by open stars (Neuman-Keuls test; bQ, P < 0.01). Significant differences between each of first 3 predrug or 6 postdrug responses vs. last postdrug response indicated by asterisks (Neuman-Keuls test; P < 0.01).
threshold (2 times threshold) for short duration trains. Behavioral seizures to long duration stimuli were suppressed after phenytoin (Fig. lA), with peak effect at the 1 h time point. The afterdischarges to long duration stimuli were not effected
at any time point (Fig. 1B). These results are comparable to those reported in a previous study”. At this dose, phenytoin totally abolished epileptiform responses to short duration stimuli, both motor and electrographic, at all the time points examined (Fig. 1A and B). Data from all doses studied are summarized in Fig. 2, where effects on behavioral seizures are presented in the first 2 columns and effects on electrographic seizures are given in the last column. Information about different time points studied is given in separate rows. By examining each panel, the dose-response relationships for phenytoin for particular time points and responses can be appreciated. It can be appreciated that there was a greater suppressive effect on responses elicited by short duration trains (open symbols) than on those elicited by long duration trains (closed symbols). In general, there were clear dose-dependent effects of phenytoin, but under some circumstances there was no effect of phenytoin demonstrated. For example, none of the doses suppressed afterdischarges elicited with long duration stimuli at any of the time points examined. In addition, it should be noted that consistently throughout the experiments, the suppression of kindled motor seizures was greater than the suppression of limbic behavioral seizures. This was perhaps most striking in the long duration stimuli tests where the 50 mg/kg dose, at all time points, suppressed kindled motor seizures, at least to some degree. At any time point, this dose either suppressed limbic behavioral seizures or afterdischarges slightly or not at all when they were triggered by long duration stimuli (compare far lefthand panels to the middle and far right-hand panels in Fig. 2). For the short duration stimuli tests, a more potent suppression of kindled motor seizures than of limbic behavioral seizures or of afterdischarges was evident with the 37.5 mg/kg dose. As with the 50 mg/kg dose (see above), in individual animals the other doses of phenytoin either totally suppressed afterdischarges to short duration stimuli or they were not changed from predrug values. Neither the 25 nor 37.5 mg/kg doses changed afterdischarge durations elicited by long duration trains.
15 KIndled
Motor
Afterdischarge
Seizures
0 20
60
Postdrug Response
30
40
50
40
50
40
50
60
100 2
60 60
/
40 20
I 20
30
40
50
0 I* 20
60
Postdrug Response
[,
~, 30
60
100 3
60 60
,/
40 20
0‘
20
o-
20
30
40
Dose of Phenytoin
50
60
mg/kg
0- 1 20
/ 30
60
i.p.
Fig. 2. Dose-response relationships for effects of phenytoin on behavioral seizures and on electrographic afterdischarges. Shown are the amount of total suppression of responses expressed as percentages (ordinates) at different doses studied (abscissae) for kindled seizures (circles, left-hand column), limbic behavioral seizures (squares, middle column) and electrographic seizures (afterdischarges; triangles, right-hand column). Kindled motor seizures (BSS 4-5) and limbic behavioral seizures (BSS 1-3) were differentiated according to a scheme previously developed ll. Responses to short duration stimuli are plotted as open symbols while responses to long duration stimuli are plotted as closed symbols. From above down, in different rows, are responses to first, second, and third tests for each type of stimulus train.
In accord with previous work”, injection of vehicle did not change either behavior responses or afterdischarges (data not shown). Pharmacokinetic experiments Results from the pharmacokinetic study are shown in Fig. 3. Compared to the i.p. route, serum concentrations rose and decayed more rapidly after i.v. phenytoin. One hour after administration, both routes gave the same blood levels. As a consequence of these 2 facts, serum concentrations of
phenytoin remained in the range clinically accepted as ‘therapeutic’ for longer after the intraperitoneal administration. At the end of the experiment on the effect of 50 mg/kg phenytoin on kindled responses (previous results section), blood samples were obtained from two animals by means of a cardiac puncture after the animals had been sacrificed. Analysis showed the corresponding serum concentrations were as expected from the full pharmacokinetic studies (see squares in Fig. 3).
16
1 P
TIME
AFTER
INJECTION
(HOURS)
Fig. 3. Pharmacokinetics of intravenous and intraperitoneal phenytoin. Serum concentrations (ordinate; logarithmic scale) of phenytoin were determined at various times (abscissa) after administration of i.v. (circles, n = 3) or i.p. (squares, n = 4) 50 mgikg phenytoin. Data presented as means i SEM. At 60 min there was no difference between serum concentrations after i.v. and i.p. administrations (t-test). Squares represent measurements at the time indicated from 2 rats in the separate subgroup of rats used to study effect of phenytoin on kindled responses.
DISCUSSION The results of the studies presented here relate to 2 experimental issues. In the first set of experiments we examined pharmacodynamic aspects of the actions of phenytoin against kindled responses. Our results demonstrated dose-response and time-action profiles for the effect of phenytoin on behavioral (kindled motor and limbic behavioral) seizures and on electrographic seizures (afterdischarges). Phenytoin was most potent in suppressing kindled motor seizures, but also attenuated limbic behavioral seizures and afterdischarges if they were elicited by short duration stimuli. These findings support the concept that kindled responses are a faithful model of complex partial seizures in humans. In addition, our results point to mechanisms by which phenytoin may exert its antiepileptic effect. The second set of experiments was pharmacokinetic in nature and established that in rats, i.p. administration of phenytoin is as effective in achieving therapeutic blood levels as i.v. administration. In the pharmacodynamic experiments, differential effects of phenytoin were found, depending, on one hand, on the type of response examined, and, on the other hand, on the type of stimulus employed. For all doses and at all study time points, the effects of phenytoin were greater against responses triggered by short duration trains than by
long duration trains. This was most dramatic for the ability of the drug to alter afterdischarges. For short duration stimuli, effects of the doses studied ranged from no suppression to total suppression of afterdischarges. While the 37.5 mg/kg dose caused a partial suppression of afterdischarges when considered from a pooled population point of view (Fig. 2), the effect was all-or-none in individual animals. Either afterdischarges were totally abolished or their durations were not different from control, pre-drug values. For long duration stimuli, no effect on afterdischarge duration at all was detected. It seems most likely that this differential sensitivity relates to the length of the short duration and long duration trains - 10 vs 1 set, respectively. In the present study, both short duration and long duration stimuli were given at a single hippo~ampal site. Therefore, differences between our previous results” and those of McNamara et a1.15cannot be attributed to their using amygdala stimuli and our using hippocampal stimuli. McNamara and coworkers do not report absolute stimulus intensities or afterdischarge thresholds. Instead they determined the generalized seizure threshold (GST) and tested for drug effects with stimuli slightly above GST. Therefore, a comparison of the influence of stimulus intensity in their study and others cannot be made. Yet, the design of the present study does provide information on this issue. In
17 our experiment, stimulus intensities were matched. Thus, absolute stimulus intensity was not a factor in the different responses. However, the intensity of the stimuli relative to afterdischarge thresholds is likely to have played a role in the results we obtained. Different stimulus intensities, relative to afterdischarge threshold, may also account, at least in part, for some of the disagreements in previous studies on the effectiveness of phenytoin against kindled responses. We also point out that by design our experiments were extended in order to develop time-action curves. In comparison, other workers have examined only time points soon after i.p. phenytoin administration. These design issues may also account for some of the disparities in the results in different reports on the efficacy of phenytoin. The pharmacokinetic experiments demonstrated that the i.p. route was efficacious in achieving therapeutic blood levels in rats. In fact, the present study demonstrated that at a 60 min time point, serum concentrations after i.p. and after i.v. injections were the same, and that thereafter the i.p. route actually maintained therapeutic blood levels for a longer period. The differences in the time course of blood levels after i.p. compared to i.v. drug delivery were entirely as would be anticipated from basic pharmacokinetic principles. Because the i.p. route provides a ‘reservoir’ of the drug, one would predict that, relative to an i.v. administration, the associated serum concentrations would show a delay in reaching their peak and a slower decay from that peak. The results from a previous study that compared blood levels obtained at 30 min after intraperitoneal phenytoin administration to those obtained 60 min after intravenous administration15 are therefore not surprising. Altogether, the i.p. route is appropriate for studying phenytoin with kindled seizures in rats. It is informative to consider pharmacokinetic and pharmacodynamic data together. After 50 mg/kg i.p. phenytoin, there was a total suppression of all (both behavioral and electrographic) seizure responses to short duration stimuli whereas the responses to long duration stimuli were affected much less with only a partial suppression of
behavioral seizures and no change in afterdischarge durations. This differential effect persisted throughout the period of study. In a parallel set of pharmacokinetic experiments, it was determined that during the period when pharmacodynamic studies were carried out, blood concentrations of phenytoin were within therapeutic levels or above. While not determined in the current work, previous work from our laboratory6 has established that brain levels of phenytoin would have been achieved throughout the period of pharmacodynamic studies. Thus, differences in the responses of limbic afterdischarges triggered by the 2 types of trains cannot be attributed to pharmacokinetic factors, but can be explained on the basis of dissimilar sensitivity to phenytoin of kindled responses triggered by different types of stimuli. The responses to the 2 types of trains can be considered with respect to potential mechanisms of action of phenytoin. With both long and short duration stimuli, behavioral seizure scores were substantially lower after phenytoin, indicating that a major action of the drug is to attenuate the spread of seizures from the point where they are started. Phenytoin was also capable of completely suppressing seizures at their site of initiation, but this effect was only evident with short duration stimuli. This observation, along with the finding that phenytoin was more potent against kindled motor seizures than against electrographic afterdischarges, indicates that a second action of phenytoin, preventing seizures from being initiated, is less robust than the ability of the drug to retard seizure propagation. The spacing of the stimuli in the protocol used in the current report was based on other work that indicates that postictal inhibition, when assessed by long duration stimuli, abates by 30 min, but requires 60 min when assessed with short duration stimuli (Williamson and Lothman, in preparation). One might comment on the longer time preceding short duration stimuli than long duration stimuli in the protocol employed here. On the basis of postictal ‘inhibition’, this spacing would favor phenytoin’s having a greater effect on the long duration stimuli. However, the opposite was found. In addition, in recent studies we have found that the effect of various antiepileptic drugs is the
18
same when tested by the protocol described above as with a protocol with equal temporal spacing before short duration and long duration stimuli (Williamson and Lothman, unpublished).
ACKNOWLEDGEMENTS
REFERENCES
11 Lothman, E.W., Salerno, R.A., Perlin, J.B. and Kaiser, D.L., Screening and characterization of antiepileptic drugs with rapidly recurring hip~campal seizures in rats, Epifepsy Res., 2 (1988) 367-377. 12 Loscher, W., Jlckel, R. and Czuczwar, S.J., Is amygdala kindling in rats a model for drug-resistant partial epilepsy?, Exp. Neural., 93 (1986) 211-226. 13 Mattson, R.H., Cramer, J.A., Collins, J.F., Smith, D.B., Delgado-Escueta, A.V., Browne, T.R., Williamson, P.D., Treiman, D.M., McNamara, J.O., McCutcheon, C.B., Homer, R.W., Crill, W.E., Luboyznski, M.F., Rosenthal, N.P. and Majersdorf, A., Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures, N. En& J. Med., 313 (1985) 145-151. 14 McNamara, J.O., Bonhaus, D.W., Shin, C., Crain, B.J., Gellman, P.L. and Giacchino, J.L., The kindling model of epilepsy: a critical review, CRC Crit. Rev. Neurobiol., 1 (1985) 341-392. 15 McNamara, J.O., Rigsbee, L.C., Butler, L.S. and Shin, C., Intravenous phenytoin is an effective anticonvulsant in the kindling model, Ann. Neural., 26 (1989) 675-678. 16 Schmutz, M., Klebs, K. and Baltzer, V., Inhibition or enhancement of kindling evolution by antiepileptics, J. Neurol. Transm., 72 (1987) 245-257. 17 Sutula, T., Cascino, G., Cavazos, J., Parada, I. and Ramirez. L., Mossy fiber synaptic reorganization in the human temporal lobe, Ann. Neurof., 26 (1989) 321-330. 18 Sutula, T., Xiao-Xian, H., Cavazos, J. and Scott, G., Synaptic reorganization in the hippocampus induced by abnormal functional activity, Science, 239 (1988) 1147-1150. 19 Trabka, W., Trabka, E. and Trabka, J., Kindled seizures: anticonvulsant drugs and agents, Pol. J. Pharmacol. Pharm., 40 (1988) 87-115. 20 Williamson, J.M. and Lothman, E.W., The effect of MK801 on kindled seizures: implications for use and limitations as an antiepileptic drug, Ann. Neural., 26 (1989) 85-90.
1 Albertson,T.E., Peterson, S.L. and Stark, L.G., Anticonvulsant drugs and their antagonism of kindied amygdaloid seizures in rats, Neuropharmacology, 19 (1980) 643-652. 2 Albright, P.S., Effects of carbamazepine, clonazepam, and phenytoin on seizure threshold in amygdala and cortex, Exp. Neural., 79 (1983) 11-17. 3 Albright, P.S. and Burnham, W.M., Development of a new pharmacological seizure model: effects of anticonvulsants on cortical- and amygdala-kindled seizures in the rat, Epilepsia, 21(1980) 681-689. 4 Callaghan, D.A. and Schwark, W.S., Pharmacological modification of amygdaloid-kindled seizures, Neuropharmacology, 29 (1980) 1131-1136. 5 Ehle, A.L., Effects of phen~oin
zures in the rat, Electroenceph.
on amydaloid kindled seiC&n. Neurophysiof.,
48
(1980) 102-105. 6 Geary, W.A.,
II, Wooten, G.F., Perlin, J.B. and Lothman, E.W., In vitro and in vivo distribution and binding of phenytoin in rat brain, J. Pharmacol. Exp. Z’her., 241 (1987) 704-713.
7 Goddard,
G.V., Dragunow, M., Maru, E. and McLeod, E.K., Kindling and the forces that oppose it. In: B.K. Doane and K.E., Livingston (Eds.), The Limb& System: Functional Organization and Clinical Disorders, Raven Press, New York, 1986, pp. 95-121. 8 Goddard, G.V., McIntyre, D.C. and Leach, C.K., A permanent change in brain fun~ion resulting from daily electrical stimulation, Exp. NeuroZ., 25 (1969) 295-330. 9 Lothman, E.W. and Collins, R.C., Kainic acid induced limbit seizures: metabolic, behavioral, electroencephalographic, and neuropathological correlates, Bruin Rex, 218 (1981) 299-318. 10 Lothman, E.W., Perlin, J.B. and Salerno, R.A., Response properties of rapidly recurring hippocampal seizures in rats, Epilepsy Res., 2 (1988) 356-360.
Supported in part by grants NS21671 and NS07199. We express our gratitude to Rose Powell for her assistance in preparation of the manuscript.
11
EPIRES 00403
Intraperitoneal
phenytoin suppresses kindled responses: effects on motor and electrographic seizures
E.W. Lothman, J.M. Williamson and K.E. VanLandingham” Department of Neurology,
University of Virginia Health Sciences Center, Charlottesville, VA (U.S.A.)
(Received 27 June 1990; revision received 13 January 1991; accepted 16 January 1991) Key words: Phenytoin; Limbic seizures; Complex partial seizures; Pharmacokinetics;
Antiepileptic drug
C~ntrove~y has arisen about the effectiveness of phenytoin against kindled seizures. It has been suggested that the reports of ineffectiveness could be accounted for by phenytoin being given by an intraperitoneal (i.p.) route in those experiments so that adequate serum concentrations were not achieved. Another possibility for the different results was dissimilar stimulus protocols employed in the various studies. The present study examined these issues. Doses of i.p. phenytoin were studied for their actions against kindled responses elicited with short (1 s) and long (10 s) stimulus trains through hippocampal electrodes. Serial application of the stimuli determined time-action relationships. Dose-dependent effects were demonstrated for ah time points examined. There was a consistently greater suppression of kindled motor seizures than limbic behavioral seizures or electrographic seizures. Phenytoin either totally blocked or did not affect the duration of afterdischarges. Actions of phenytoin against responses by short duration stimuli were greater than against long duration stimuli. Additional pharmacokinetic studies compared i.p. versus intravenous (i.v.) phenytoin. After i.p. phenytoin, serum levels peaked later than after iv. delivery, but were maintained in the ‘therapeutic’ range longer. The present experiments provide additional support for the idea that kindled seizures are a useful model for complex partial seizures in humans. In addition, they show that major actions of phenytoin are to decrease seizure spread and to elevate afterdischarge thresholds and that the i.p. route is appropriate for assaying the effect of phenytoin against kindled seizures in rats.
INTRODUCTION Repetition of electrical stimulus trains at various sites in the limbic system leads to a progressive enhancement of epileptiform responses until stereotyped generalized convulsions appear7*8. Even after extended periods with no stimulation, the con* Present address: Dept. Neurology, University of ~ichig~, Ann Arbor, MI 48109, U.S.A. Correspondence to: E.W. Lothman, Dept. Neurology Box 394, University of Virginia Health Sciences Center, Charlottesville, VA 22908, U.S.A.
0920-1211/91/$03.50 @ 1991 Elsevier Science Publishers B.V.
vulsions are again elicited when the stimulus is reinstituted. Thus, the condition of enhanced responsiveness, referred to as a kindled state, is enduring. The dynamic period, when responses demonstrate increasing severity, is referred to as kindling. Kindling and seizures elicited in the kindled state are regarded as powerful models of complex partial seizures with generalization7,14. Morphological changes are found in the brains of kindled animals”‘, comparable to those encountered in patients with complex partial epilepsy”. Moreover, the sensitivity of kindled seizures to a battery of
12
antiepileptic drugs agrees well with the sensitivity of seizures in humans with complex partial epilepsy to these agents11~12~14. This agreement between the set of antiepileptic drugs effective against kindled responses with the set of drugs effective against complex partial seizures in humans supports the appropriateness and utility of kindled responses as a model system. However, prior reports on the effectiveness of phenytoin against kindled responses have presented problems for the concordance of drug effects between the animal model and the clinical condition of complex partial seizures 1-5,11,12,16,19. In particular, many investigators have reported phenytoin to be ineffective against limbic afterdischarges elicited in kindled animals. There is also controversy as to how effective the drug is against kindled motor seizures. Phenytoin is well-recognized for its efficacy in suppressing complex partial seizures in patients13. Recently, McNamara et al.” suggested that the observations that phenytoin was ineffective against kindled responses were a consequence of the drug being given intraperitoneally (i.p.) in those experiments, instead of intravenously (i.v.) as they had done. If so, then the ‘disparity’ of the effectiveness of phenytoin observed in kindled animals compared to humans could be related solely to pharmacokinetic factors. We have developed a stimulus protocol to test the effect of various drugs against kindled responses at multiple time points after a single drug injection, thereby giving a time-action profile in a single experimental session”. This protocol employs 50 Hz, 10 set trains at a stimulus intensity supramaximal to afterdischarge threshold. Using this protocol, we found that i.p. phenytoin did not affect kindled limbic afterdischarges over a period of i/2 to 6 h after drug delivery’l. The stimulus protocol employed by McNamara and his colleagues15 to demonstrate an effect of phenytoin against kindled limbic afterdischarges used trains of comparable (60 Hz) frequency, but shorter (1 s) duration and a stimulus intensity equal to the generalized seizure threshold (GST). Therefore, another explanation for the dissimilar results is the different stimulus protocols used in the various studies. The present experiments examined two related
issues. In the first set of experiments we determined the influence of i.p. phenytoin against kindled responses elicited in rats by 2 different types of stimulus train. One type of train resembled that used by McNamara et al.“, and the other was like the one we had previously usedl’. This study tested whether different stimulus parameters could account for the dissimilar results of previous studies. In the second set of experiments, we compared the pharmacokinetics of i.p. versus i.v. phenytoin in rats. This study examined the possibility that i.p. phenytoin was ineffective in achieving adequate blood levels to suppress seizures. METHODS Adult male (300-325 g) Sprague-Dawley rats were prepared according to methods detailed elsewhere’0,“,20. In brief, bipolar electrodes were implanted in the ventral hippocampus. One week later, the animals were kindled with 50 Hz, 10 set trains of 400 ,uA, 1 msec biphasic pulses until stable, fully kindled responses were established. Following completion of experiments, electrode placements were confirmed histologically. After kindling, animals were studied with a modified ‘mixed trial’ protocol*‘, consisting of two kinds of stimuli, selected for features of the different test protocols used in previous studies and introduced above. Both types of stimuli consisted of 50 Hz trains of 1 msec biphasic pulses. One kind of train replicated the type of stimuli used in our previous work and was 10 s long and 4OOpA in intensity, supramaximal to the afterdischarge threshold”. The other kind of train simulated that used by other workers3,12’15,including McNamara and co-workers, and was 1 set long, twice afterdischarge threshold. The use of these stimulus parameters allowed stimulus current and intratrain frequency to be matched across the 2 types of stimuli (see below). For convenience the first type of stimulus will be referred to hereafter as ‘long duration’ and the latter type of stimulus will be referred to as ‘short duration’. Each test day began with a long duration stimulus and the two types of stimuli were alternated as follows. Stable, kindled responses are triggered
13
when long duration stimuli are given 30 min after the preceding seizure l”,il. In preliminary experiments it was determined that the period after the preceding seizure had to be increased to 60 min to provide stable, serial kindled responses to short duration stimuli. Accordingly, in our protocol, half an hour elapsed after the previous seizure with long duration stimuli, and 1 h with short duration stimuli. Each test period began with long duration trains. In order to provide a 30 min postdrug test for short duration stimuli, the order of long duration and short duration stimuli was reversed after the first two pairs of responses (see Fig. 1 below). Animals were repeatedly tested with this protocol, with at least 1 stimulus-free day between each testing session. At the beginning of each test day afterdischarge thresholds for both the 50 Hz, 10 set and for the 50 Hz, 1 set trains were determined. Only animals that had stable afterdischarge thresholds for both types of stimuli, and showed only kindled motor (stages 4 and 5) seizures and consistent afterdischarge durations (~10% variance) to each of the 10 stimuli on 3 serial test days before drug study and on at least 1 test day after drug study, were used in the drug study. Phenytoin (25 (n = 5), 37.5 (n = 5), and 50 (n = 4) mg/kg) was injected i.p. after 2 pairs of responses had established that the animals under study were consistently expressing kindled responses on that test day. Drug test days were at least 1 week apart with drug-free test days interposed as described above. The drug was given 30 min before the first postdrug response was obtained. Phenytoin was prepared in a vehicle of 70% distilled water/30% polyethylene glycol as described elsewhere”. Pharmacokinetic comparisons between i.p. and i.v. phenytoin were carried out with a separate group of animals. Attention was focused on the 50 mg/kg dose for the following reasons. As will be detailed below (see Results), this dose was shown to suppress afterdischarges to short duration stimuli, but not to affect afterdischarges to long duration stimuli. Hence, the responses to this single dose reproduced the different results cited in the Introduction. In addition, 50 mg/kg was the dose studied by McNamara and co-workers” in their comparisons of phenytoin blood levels after i.p.
and i.v. administrations. For the pharmacokinetic study, one subgroup of rats received placement of intravascular lines in their femoral arteries and femoral veins using methods described elsewhere’. Catheterization was done under halothane anesthesia 24 h before the following experiment. These animals then received 50 mg/kg phenytoin through the femoral vein catheter. In accord with clinical practice, the delivery was carried out over a 30 min period. Another subgroup of animals, with catheters placed only in their femoral arteries, received 50 mg/kg i.p. phenytoin. In both subgroups 250 ~1 blood samples were withdrawn via the arterial catheter at multiple time points after drug administration. Serum concentrations of phenytoin for the various samples were then determined by the Clinical Toxicology Laboratory at the University of Virginia by means of an enzyme immunoassay. Severity of behavioral seizures was assessed with a conventional 5-point behavioral seizure scoring system and electrographic seizures were quantified by afterdischarge durations”. Statistical comparisons of behavioral seizure scores were carried out with Mann-Whitney tests. Statistical comparisons among afterdischarge durations were carried out with analysis of variance (ANOVA) of repeated measures and Neuman-Keuls post-hoc tests. Statistical comparisons of phenytoin levels were done with t-tests. Level of significance was set at P < 0.05 throughout. RESULTS Effects of phenytoin on kindled responses to short duration and long duration stimulus trains
Representative results are shown in Fig. 1, where responses of behavioral seizures and of electrographic seizures elicited by both short duration and long duration stimuli are charted. Absolute current intensities were matched for the two types of trains (450 f 87 ,uA for short duration and 400 f 0 ,BA for long duration trains, not statistically different; t-test). Relative to afterdischarge thresholds (45 f 5 PA for long duration and 255 + 43 ,uA for short duration trains), stimulus intensities were supramaximal (approx. 10 times threshold) for long duration trains while being supra-
14 Injection
phenytorn
50
mg/kg
!** 0
Short (1
2
1
A
set)
Long (10
train duration
xc)
:
1
duration
train
p*
I
4,” 120
I / i**
i**
l** li*
r
100
r i
-
z
_m
so 60 40
t 1 Hr
Fig. 1. Comparison of effects of phenytoin on kindled responses triggered by short duration and long duration trains. Rats with stable, fully kindled responses were tested with 5 pairs of short duration (circles) and long duration (triangles) trains as described in the text. After the first two pairs, the order of delivery was reversed and 50 mg/kg i.p. phenytoin was administered. Thirty minutes later, response testing was resumed. Top panel displays behavioral seizure scores (BSS) and lower panel displays afterdischarge durations (ADD, in set), both plotted as means f SEM. Significant differences between BSS responses in each pair (i.e. short versus long duration train comparison) indicated by filled stars (Mann-Whitney tests; *t, P < 0.01; symbols next to short duration response of each pair; responses compared for each pair connected by line); significant differences between first 3 predrug responses or 6 postdrug responses when compared to lastpredrug response (also Mann-Whitney tests) indicated by asterisks (**P < 0.01). There was a significant difference across the 10 set ADD responses(ANOVA of repeated measures, F, 38.5; P < 0.001). Differences between responses in each test pair for ADD indicated by open stars (Neuman-Keuls test; bQ, P < 0.01). Significant differences between each of first 3 predrug or 6 postdrug responses vs. last postdrug response indicated by asterisks (Neuman-Keuls test; P < 0.01).
threshold (2 times threshold) for short duration trains. Behavioral seizures to long duration stimuli were suppressed after phenytoin (Fig. lA), with peak effect at the 1 h time point. The afterdischarges to long duration stimuli were not effected
at any time point (Fig. 1B). These results are comparable to those reported in a previous study”. At this dose, phenytoin totally abolished epileptiform responses to short duration stimuli, both motor and electrographic, at all the time points examined (Fig. 1A and B). Data from all doses studied are summarized in Fig. 2, where effects on behavioral seizures are presented in the first 2 columns and effects on electrographic seizures are given in the last column. Information about different time points studied is given in separate rows. By examining each panel, the dose-response relationships for phenytoin for particular time points and responses can be appreciated. It can be appreciated that there was a greater suppressive effect on responses elicited by short duration trains (open symbols) than on those elicited by long duration trains (closed symbols). In general, there were clear dose-dependent effects of phenytoin, but under some circumstances there was no effect of phenytoin demonstrated. For example, none of the doses suppressed afterdischarges elicited with long duration stimuli at any of the time points examined. In addition, it should be noted that consistently throughout the experiments, the suppression of kindled motor seizures was greater than the suppression of limbic behavioral seizures. This was perhaps most striking in the long duration stimuli tests where the 50 mg/kg dose, at all time points, suppressed kindled motor seizures, at least to some degree. At any time point, this dose either suppressed limbic behavioral seizures or afterdischarges slightly or not at all when they were triggered by long duration stimuli (compare far lefthand panels to the middle and far right-hand panels in Fig. 2). For the short duration stimuli tests, a more potent suppression of kindled motor seizures than of limbic behavioral seizures or of afterdischarges was evident with the 37.5 mg/kg dose. As with the 50 mg/kg dose (see above), in individual animals the other doses of phenytoin either totally suppressed afterdischarges to short duration stimuli or they were not changed from predrug values. Neither the 25 nor 37.5 mg/kg doses changed afterdischarge durations elicited by long duration trains.
15 KIndled
Motor
Afterdischarge
Seizures
0 20
60
Postdrug Response
30
40
50
40
50
40
50
60
100 2
60 60
/
40 20
I 20
30
40
50
0 I* 20
60
Postdrug Response
[,
~, 30
60
100 3
60 60
,/
40 20
0‘
20
o-
20
30
40
Dose of Phenytoin
50
60
mg/kg
0- 1 20
/ 30
60
i.p.
Fig. 2. Dose-response relationships for effects of phenytoin on behavioral seizures and on electrographic afterdischarges. Shown are the amount of total suppression of responses expressed as percentages (ordinates) at different doses studied (abscissae) for kindled seizures (circles, left-hand column), limbic behavioral seizures (squares, middle column) and electrographic seizures (afterdischarges; triangles, right-hand column). Kindled motor seizures (BSS 4-5) and limbic behavioral seizures (BSS 1-3) were differentiated according to a scheme previously developed ll. Responses to short duration stimuli are plotted as open symbols while responses to long duration stimuli are plotted as closed symbols. From above down, in different rows, are responses to first, second, and third tests for each type of stimulus train.
In accord with previous work”, injection of vehicle did not change either behavior responses or afterdischarges (data not shown). Pharmacokinetic experiments Results from the pharmacokinetic study are shown in Fig. 3. Compared to the i.p. route, serum concentrations rose and decayed more rapidly after i.v. phenytoin. One hour after administration, both routes gave the same blood levels. As a consequence of these 2 facts, serum concentrations of
phenytoin remained in the range clinically accepted as ‘therapeutic’ for longer after the intraperitoneal administration. At the end of the experiment on the effect of 50 mg/kg phenytoin on kindled responses (previous results section), blood samples were obtained from two animals by means of a cardiac puncture after the animals had been sacrificed. Analysis showed the corresponding serum concentrations were as expected from the full pharmacokinetic studies (see squares in Fig. 3).
16
1 P
TIME
AFTER
INJECTION
(HOURS)
Fig. 3. Pharmacokinetics of intravenous and intraperitoneal phenytoin. Serum concentrations (ordinate; logarithmic scale) of phenytoin were determined at various times (abscissa) after administration of i.v. (circles, n = 3) or i.p. (squares, n = 4) 50 mgikg phenytoin. Data presented as means i SEM. At 60 min there was no difference between serum concentrations after i.v. and i.p. administrations (t-test). Squares represent measurements at the time indicated from 2 rats in the separate subgroup of rats used to study effect of phenytoin on kindled responses.
DISCUSSION The results of the studies presented here relate to 2 experimental issues. In the first set of experiments we examined pharmacodynamic aspects of the actions of phenytoin against kindled responses. Our results demonstrated dose-response and time-action profiles for the effect of phenytoin on behavioral (kindled motor and limbic behavioral) seizures and on electrographic seizures (afterdischarges). Phenytoin was most potent in suppressing kindled motor seizures, but also attenuated limbic behavioral seizures and afterdischarges if they were elicited by short duration stimuli. These findings support the concept that kindled responses are a faithful model of complex partial seizures in humans. In addition, our results point to mechanisms by which phenytoin may exert its antiepileptic effect. The second set of experiments was pharmacokinetic in nature and established that in rats, i.p. administration of phenytoin is as effective in achieving therapeutic blood levels as i.v. administration. In the pharmacodynamic experiments, differential effects of phenytoin were found, depending, on one hand, on the type of response examined, and, on the other hand, on the type of stimulus employed. For all doses and at all study time points, the effects of phenytoin were greater against responses triggered by short duration trains than by
long duration trains. This was most dramatic for the ability of the drug to alter afterdischarges. For short duration stimuli, effects of the doses studied ranged from no suppression to total suppression of afterdischarges. While the 37.5 mg/kg dose caused a partial suppression of afterdischarges when considered from a pooled population point of view (Fig. 2), the effect was all-or-none in individual animals. Either afterdischarges were totally abolished or their durations were not different from control, pre-drug values. For long duration stimuli, no effect on afterdischarge duration at all was detected. It seems most likely that this differential sensitivity relates to the length of the short duration and long duration trains - 10 vs 1 set, respectively. In the present study, both short duration and long duration stimuli were given at a single hippo~ampal site. Therefore, differences between our previous results” and those of McNamara et a1.15cannot be attributed to their using amygdala stimuli and our using hippocampal stimuli. McNamara and coworkers do not report absolute stimulus intensities or afterdischarge thresholds. Instead they determined the generalized seizure threshold (GST) and tested for drug effects with stimuli slightly above GST. Therefore, a comparison of the influence of stimulus intensity in their study and others cannot be made. Yet, the design of the present study does provide information on this issue. In
17 our experiment, stimulus intensities were matched. Thus, absolute stimulus intensity was not a factor in the different responses. However, the intensity of the stimuli relative to afterdischarge thresholds is likely to have played a role in the results we obtained. Different stimulus intensities, relative to afterdischarge threshold, may also account, at least in part, for some of the disagreements in previous studies on the effectiveness of phenytoin against kindled responses. We also point out that by design our experiments were extended in order to develop time-action curves. In comparison, other workers have examined only time points soon after i.p. phenytoin administration. These design issues may also account for some of the disparities in the results in different reports on the efficacy of phenytoin. The pharmacokinetic experiments demonstrated that the i.p. route was efficacious in achieving therapeutic blood levels in rats. In fact, the present study demonstrated that at a 60 min time point, serum concentrations after i.p. and after i.v. injections were the same, and that thereafter the i.p. route actually maintained therapeutic blood levels for a longer period. The differences in the time course of blood levels after i.p. compared to i.v. drug delivery were entirely as would be anticipated from basic pharmacokinetic principles. Because the i.p. route provides a ‘reservoir’ of the drug, one would predict that, relative to an i.v. administration, the associated serum concentrations would show a delay in reaching their peak and a slower decay from that peak. The results from a previous study that compared blood levels obtained at 30 min after intraperitoneal phenytoin administration to those obtained 60 min after intravenous administration15 are therefore not surprising. Altogether, the i.p. route is appropriate for studying phenytoin with kindled seizures in rats. It is informative to consider pharmacokinetic and pharmacodynamic data together. After 50 mg/kg i.p. phenytoin, there was a total suppression of all (both behavioral and electrographic) seizure responses to short duration stimuli whereas the responses to long duration stimuli were affected much less with only a partial suppression of
behavioral seizures and no change in afterdischarge durations. This differential effect persisted throughout the period of study. In a parallel set of pharmacokinetic experiments, it was determined that during the period when pharmacodynamic studies were carried out, blood concentrations of phenytoin were within therapeutic levels or above. While not determined in the current work, previous work from our laboratory6 has established that brain levels of phenytoin would have been achieved throughout the period of pharmacodynamic studies. Thus, differences in the responses of limbic afterdischarges triggered by the 2 types of trains cannot be attributed to pharmacokinetic factors, but can be explained on the basis of dissimilar sensitivity to phenytoin of kindled responses triggered by different types of stimuli. The responses to the 2 types of trains can be considered with respect to potential mechanisms of action of phenytoin. With both long and short duration stimuli, behavioral seizure scores were substantially lower after phenytoin, indicating that a major action of the drug is to attenuate the spread of seizures from the point where they are started. Phenytoin was also capable of completely suppressing seizures at their site of initiation, but this effect was only evident with short duration stimuli. This observation, along with the finding that phenytoin was more potent against kindled motor seizures than against electrographic afterdischarges, indicates that a second action of phenytoin, preventing seizures from being initiated, is less robust than the ability of the drug to retard seizure propagation. The spacing of the stimuli in the protocol used in the current report was based on other work that indicates that postictal inhibition, when assessed by long duration stimuli, abates by 30 min, but requires 60 min when assessed with short duration stimuli (Williamson and Lothman, in preparation). One might comment on the longer time preceding short duration stimuli than long duration stimuli in the protocol employed here. On the basis of postictal ‘inhibition’, this spacing would favor phenytoin’s having a greater effect on the long duration stimuli. However, the opposite was found. In addition, in recent studies we have found that the effect of various antiepileptic drugs is the
18
same when tested by the protocol described above as with a protocol with equal temporal spacing before short duration and long duration stimuli (Williamson and Lothman, unpublished).
ACKNOWLEDGEMENTS
REFERENCES
11 Lothman, E.W., Salerno, R.A., Perlin, J.B. and Kaiser, D.L., Screening and characterization of antiepileptic drugs with rapidly recurring hip~campal seizures in rats, Epifepsy Res., 2 (1988) 367-377. 12 Loscher, W., Jlckel, R. and Czuczwar, S.J., Is amygdala kindling in rats a model for drug-resistant partial epilepsy?, Exp. Neural., 93 (1986) 211-226. 13 Mattson, R.H., Cramer, J.A., Collins, J.F., Smith, D.B., Delgado-Escueta, A.V., Browne, T.R., Williamson, P.D., Treiman, D.M., McNamara, J.O., McCutcheon, C.B., Homer, R.W., Crill, W.E., Luboyznski, M.F., Rosenthal, N.P. and Majersdorf, A., Comparison of carbamazepine, phenobarbital, phenytoin, and primidone in partial and secondarily generalized tonic-clonic seizures, N. En& J. Med., 313 (1985) 145-151. 14 McNamara, J.O., Bonhaus, D.W., Shin, C., Crain, B.J., Gellman, P.L. and Giacchino, J.L., The kindling model of epilepsy: a critical review, CRC Crit. Rev. Neurobiol., 1 (1985) 341-392. 15 McNamara, J.O., Rigsbee, L.C., Butler, L.S. and Shin, C., Intravenous phenytoin is an effective anticonvulsant in the kindling model, Ann. Neural., 26 (1989) 675-678. 16 Schmutz, M., Klebs, K. and Baltzer, V., Inhibition or enhancement of kindling evolution by antiepileptics, J. Neurol. Transm., 72 (1987) 245-257. 17 Sutula, T., Cascino, G., Cavazos, J., Parada, I. and Ramirez. L., Mossy fiber synaptic reorganization in the human temporal lobe, Ann. Neurof., 26 (1989) 321-330. 18 Sutula, T., Xiao-Xian, H., Cavazos, J. and Scott, G., Synaptic reorganization in the hippocampus induced by abnormal functional activity, Science, 239 (1988) 1147-1150. 19 Trabka, W., Trabka, E. and Trabka, J., Kindled seizures: anticonvulsant drugs and agents, Pol. J. Pharmacol. Pharm., 40 (1988) 87-115. 20 Williamson, J.M. and Lothman, E.W., The effect of MK801 on kindled seizures: implications for use and limitations as an antiepileptic drug, Ann. Neural., 26 (1989) 85-90.
1 Albertson,T.E., Peterson, S.L. and Stark, L.G., Anticonvulsant drugs and their antagonism of kindied amygdaloid seizures in rats, Neuropharmacology, 19 (1980) 643-652. 2 Albright, P.S., Effects of carbamazepine, clonazepam, and phenytoin on seizure threshold in amygdala and cortex, Exp. Neural., 79 (1983) 11-17. 3 Albright, P.S. and Burnham, W.M., Development of a new pharmacological seizure model: effects of anticonvulsants on cortical- and amygdala-kindled seizures in the rat, Epilepsia, 21(1980) 681-689. 4 Callaghan, D.A. and Schwark, W.S., Pharmacological modification of amygdaloid-kindled seizures, Neuropharmacology, 29 (1980) 1131-1136. 5 Ehle, A.L., Effects of phen~oin
zures in the rat, Electroenceph.
on amydaloid kindled seiC&n. Neurophysiof.,
48
(1980) 102-105. 6 Geary, W.A.,
II, Wooten, G.F., Perlin, J.B. and Lothman, E.W., In vitro and in vivo distribution and binding of phenytoin in rat brain, J. Pharmacol. Exp. Z’her., 241 (1987) 704-713.
7 Goddard,
G.V., Dragunow, M., Maru, E. and McLeod, E.K., Kindling and the forces that oppose it. In: B.K. Doane and K.E., Livingston (Eds.), The Limb& System: Functional Organization and Clinical Disorders, Raven Press, New York, 1986, pp. 95-121. 8 Goddard, G.V., McIntyre, D.C. and Leach, C.K., A permanent change in brain fun~ion resulting from daily electrical stimulation, Exp. NeuroZ., 25 (1969) 295-330. 9 Lothman, E.W. and Collins, R.C., Kainic acid induced limbit seizures: metabolic, behavioral, electroencephalographic, and neuropathological correlates, Bruin Rex, 218 (1981) 299-318. 10 Lothman, E.W., Perlin, J.B. and Salerno, R.A., Response properties of rapidly recurring hippocampal seizures in rats, Epilepsy Res., 2 (1988) 356-360.
Supported in part by grants NS21671 and NS07199. We express our gratitude to Rose Powell for her assistance in preparation of the manuscript.
