Striking Differences in Individual Anticonvulsant Response to Phenobarbital in Rats with Spontaneous Seizures after Status Epilepticus


Address correspondence and reprint requests to Dr. W. Löscher at Department of Pharmacology, Toxicology and Pharmacy, School of Veterinary Medicine, Bünteweg 17, D-30559 Hannover, Germany. E-mail:


Summary: Purpose: More than one third of patients with epilepsy have inadequate control of seizures with drug therapy, but mechanisms of intractability are largely unknown. Because of this large number of pharmacoresistant patients with epilepsy, the existing process of antiepileptic drug (AED) discovery and development must be reevaluated with a focus on preclinical models of therapy-resistant epilepsy syndromes such as mesial temporal lobe epilepsy (TLE). However, although various rodent models of TLE are available, the pharmacoresponsiveness of most models is not well known. In the present study, we used a post–status epilepticus model of TLE to examine whether rats with spontaneous recurrent seizures (SRSs) differ in their individual responses to phenobarbital (PB).

Methods: Status epilepticus was induced in Sprague–Dawley rats by prolonged electrical stimulation of the basolateral amygdala. Once the rats had developed SRSs, seizure frequency and severity were determined by continuous EEG/video recording over a 6-week period (i.e., a predrug control period of 2 weeks, followed by PB treatment for 2 weeks, and a postdrug control period of 2 weeks). PB was administered twice daily at maximal tolerated doses.

Results: Analysis of plasma drug concentrations showed that drug concentrations within the therapeutic range (10–40 μg/ml) were maintained in all rats throughout the period of treatment. In six (55%) of 11 rats, complete control of seizures was achieved, and another rat exhibited a >90% reduction of seizure frequency. These seven rats were considered responders. The remaining four (36%) rats showed either no response at all (n = 3) or only moderate reduction in seizure frequency and were therefore considered nonresponders. Plasma drug concentrations did not differ between these two groups of rats.

Conclusions: These data demonstrate that, similar to patients with epilepsy, rats with SRSs markedly differ in their individual responses to AED treatment. Pharmacoresistant rats selected by prolonged drug treatment from groups of rats with SRSs may provide a unique model to study mechanisms of pharmacoresistance and to identify novel AEDs for treating seizures of patients currently not controlled with existing therapies.

Medical intractability [i.e., absence of any response to antiepileptic drug (AED) therapy], is an unresolved problem in many patients with temporal lobe epilepsy (TLE). Mechanisms of intractability are poorly understood, thus complicating strategies aimed to develop more efficacious treatments (1,2). Animal models of TLE with spontaneous seizures not responding or with very poor response to clinically established AEDs would be useful to identify new therapeutic agents for resistance (3,4). To our knowledge, such models are not available. Rat models of TLE in which spontaneous recurrent seizures (SRSs) develop after a chemically or electrically induced status epilepticus (SE) are widely used to study the basic pathophysiology underlying development of partial epilepsy (5–7), however, the pharmacology of such models is largely unknown. Leite and Cavalheiro (8) were the first to study effects of conventional AEDs on SRSs in such a model in which SE was induced by pilocarpine. Except ethosuximide (ESM), all AEDs used [phenobarbital (PB), phenytoin (PHT), carbamazepine (CBZ), valproate (VPA)] effectively suppressed SRSs in all rats (8), which is unlike the clinical situation in TLE, in which ≤70% of patients do not respond to treatment with AEDs at maximal tolerated doses (9). However, the rats in the study of Leite and Cavalheiro (8) were observed for only 10 h/day for SRSs, and no continuous EEG or video monitoring of SRSs was used during the 2-week period of AED treatment, so that pharmacoresistant rats might have been overlooked. In a more recent study, also using the pilocarpine model of TLE, we found that the individual responses of rats to treatment with the novel AED levetiracetam (LEV) varied markedly from complete control of SRSs to no effect at all, although plasma drug levels were within the same range in all rats (10).

In the present study, we examined whether SRSs developing after an electrically induced SE also respond differently to an AED among individual rats. For this purpose, we used PB, which was administered at a dosing protocol by which plasma levels within or above the “therapeutic range” could be maintained in rats. SRSs were monitored by continuous video-EEG recording. For induction of SE, a recently described rat model with prolonged electrical stimulation of the basolateral amygdala (BLA) was used (11). In this model, most rats develop SRSs after a latency of ∼4 to 6 weeks. As other post-SE models, rats with SRSs from this BLA model show a number of parallels with a common form of human drug-resistant epilepsy, mesial TLE (11).



Based on our previous experience in different rat strains and genders, female Sprague–Dawley rats were used for this study because prolonged BLA stimulation in these rats results in a high frequency of generalized convulsive SE, which induces development of SRSs in >90% of the animals (11). The rats were purchased at a body weight of 200 to 230 g (Harlan-Winkelmann Versuchstierzucht, Borchen, Germany). After arrival, the rats were kept under controlled environmental conditions (24–25°C; 50–60% humidity; 12-h light/dark cycle; light on at 5:00 a.m.) with free access to standard laboratory chow (Altromin 1324 standard diet) and tap water. Before being used in the experiments, the rats were allowed to adapt to the new conditions for ≥1 week. All experiments were done in compliance with the German Animal Welfare Act. All efforts were made to minimize pain or discomfort of the animals used.

Electrode implantation

For electrode implantation, 14 rats were anesthetized with chloral hydrate (360 mg/kg, i.p.). A Teflon-isolated bipolar stainless steel electrode was stereotactically implanted into the right anterior BLA, as described recently (11), and served as a stimulation- and recording-electrode. A screw, placed above the left parietal cortex, served as the indifferent reference electrode. Additional skull screws and dental acrylic cement anchored the electrode assembly. After surgery, the animals were allowed to recover for a period of ≥2 weeks.

Induction of a self-sustaining status epilepticus

About 2 weeks after electrode implantation, the baseline EEG (recorded from the BLA electrode) was recorded for 5 to 10 min. Thereafter, the rats were electrically stimulated via the BLA electrode for induction of a self-sustained SE (SSSE). The following stimulus parameters were chosen: stimulus duration, 25 min; and stimulus consisting of 100-ms trains of 1-ms alternating positive and negative square-wave pulses. The trains were given at a frequency of two per second, and the intratrain pulse frequency was 50 per second. Peak pulse intensity was 700 μA. For this pulsed-train stimulation, an Accupulser A310C stimulator connected with a Stimulus Isolator A365 (World Precision Instruments, Berlin, Germany) were used. All rats developed an SSSE with generalized convulsive seizures (n = 12) or focal seizures interrupted by occasional episodes of generalized convulsive seizures (n = 2). After 4 h, SSSE was interrupted by diazepam (DZP; 10 mg/kg, i.p.). Starting ∼4 weeks later, the rats were monitored by video-EEG recordings for ≤2 months until the first spontaneous seizure was detected, as described recently (11). In 12 (86%) of the 14 rats, SRSs were recorded. About 3 months after SE, these 12 rats were video-EEG monitored for 12 h/day over a period of 4 consecutive days to examine the frequency of SRSs. For the drug study described later, we included only rats that showed at least one spontaneous seizure over this period to minimize the risk of having rats without SRSs during the 2 weeks before drug treatment (see later). Based on this criterion, 11 of the rats were chosen for the experiments with PB.

Treatment with phenobarbital

Before the experiments in rats with SRSs, the pharmacokinetics of this drug in female Sprague–Dawley rats were determined to obtain a basis for a dosing protocol allowing maintenance of plasma drug concentrations in or above the therapeutic range [10–40 μg/ml; (12)]. Furthermore, our aim was to administer PB at maximal tolerated doses, so that rats were closely observed for adverse effects. PB was i.p. injected at either 20, 25, or 30 mg/kg in two rats per dose, and blood (∼0.5 ml) was sampled by retroorbital puncture (after local anesthesia with tetracaine) at 0.5, 1, 2, 6, and 12 h after injection. Based on pharmacokinetics calculated from PB analysis in plasma samples in these experiments (see Results), we calculated bolus and maintenance doses to obtain plasma concentrations of 20 to 40 μg/ml throughout 24 h, by using the pharmacokinetic data-analysis system TopFit 2.0 (13). Based on these calculations, a dosing protocol with an i.p. bolus dose of 25 mg/kg in the morning, followed 10 h later by 15 mg/kg i.p., and then twice-daily 15 mg/kg i.p. (at 8:00 a.m. and 6:00 p.m.) on subsequent days for 1 week was used in a preliminary experiment in six rats. Blood samples were taken 1 h before and 1 h after the last drug administration on day 7, resulting in average PB plasma levels of 34.1 μg/ml (range, 21.2–39.8μg/ml) and 55.5 μg/ml (range, 42.4–60.5 μg/ml), respectively. Rats exhibited marked sedation during the 1-week period of treatment that excluded the use of higher doses. This dosage protocol therefore fulfilled our aims and was used for the experiment in rats with SRS.

For the experiments with the effect of PB on SRSs, 11 rats with SRSs obtained from the BLA model (see earlier) were used. About 5 months after SE, these rats were used for drug testing as follows. Baseline seizure frequency was determined over a two-week period (predrug control period), and then PB was administered over a 2-week period, followed by a postdrug control period of 2 weeks. During the control periods, rats received twice daily an i.p. injection of saline (3 ml/kg). After the predrug control period, treatment with PB was started by an i.p. bolus dose of 25 mg/kg in the morning, followed 10 h later by an administration of 15 mg/kg. On the subsequent 13 days, PB was injected twice daily at 15 mg/kg. During the 6 weeks of the experiments, all saline and PB injections were done between 7:30 and 8:00 a.m. and 5:30 and 6:00 p.m., respectively. Blood was sampled 10 h after the first drug injection and 12 h after the last drug injection. In one rat with an extremely high seizure frequency and intensely aggressive behavior (see Results), it was not possible to sample blood.

In each of the 6 weeks of the experiment, seizures were continuously (24 h/day, 7 days/week) monitored by video-EEG recording (see later). For rating of seizure severity of spontaneous seizures, Racine's scale (14) was used. In addition to the five seizure stages rated by this grading system, a stage 6 was used to characterize running-bouncing seizures, which were occasionally observed before or after a generalized convulsive seizure. Based on this scale, seizures were subdivided into nonconvulsive (stages 1 and 2) and convulsive (stages 3–6). Some rats displayed an additional convulsive seizure type not covered by the Racine scale, which was characterized by jerky backward movements of the head sometimes associated with a short, clonic movement of one or both forelimbs.

Monitoring of spontaneous recurrent seizures

For continuous EEG monitoring, an eight-channel amplifier (CyberAmp 380; Axon Instruments, Inc., Foster City, CA, U.S.A.), eight one-channel bioamplifiers (ADInstruments Ltd., Hastings, East Sussex, U.K.), and two analogue–digital converters (PowerLab/800s; ADInstruments Ltd.) were used. This system allowed simultaneous recording of EEGs from ≤16 rats. The data were recorded and analyzed with the Chart4 for Windows software (ADInstruments Ltd.). The sampling rate for the EEG recording was 200 Hz. A high-pass filter for 0.1 Hz and a low-pass filter for 60 Hz were used.

Simultaneous with the EEG recording, the rats were video-monitored, by using four light-sensitive black–white cameras (CCD-Kamera-Modul; Conrad Electronic, Hannover, Germany), which allowed video-recording of up to four rats per camera. The cameras were connected to a multiplexer (TVMP-400; Monacor, Bremen, Germany) that converted the digitalized signals from the four cameras to a video recorder (Time Lapse recorder, Sanyo TLS-9024P; Monacor, Bremen, Germany). To allow videorecording of seizures during the night, the animal room was weekly illuminated by red light during the dark phase. Rats were housed in clear glass cages (one per cage) to allow optimal video observation.

For detection of spontaneous seizures, the EEG recordings were visually analyzed for characteristic ictal events. To evaluate the severity of motor seizure activity during a detected seizure in the EEG, the corresponding video-recording was viewed. In addition to seizures observed by video and EEG recordings, seizures observed during handling, other manipulations of the animals, or by direct observation of the rats in their home cages were noted. Based on individual responses of rats to treatment, they were considered either responders or nonresponders. Responders were defined by complete seizure suppression during treatment or a seizure suppression of >50% compared with seizure frequency in the control period before treatment.

In addition to monitoring of seizures, rats were observed for adverse effects of PB by using both video recordings and direct observation during the period of treatment. The most obvious adverse effect was pronounced sedation, illustrated by reduced motility and reduced reactions to handling.

Drug analysis in plasma

PB levels in plasma were determined by high-performance liquid chromatography (HPLC) with ultraviolet detection, as described recently (15).


PB (as sodium salt) was purchased from Serva (Heidelberg, Germany). Solutions were prepared in 0.9% saline at an administration volume of 3 ml/kg.


The significance of differences between the predrug, drug, and postdrug recordings within one group of rats was calculated by analysis of variance (ANOVA), followed by post hoc testing for individual differences. Tests for paired replicates were used, the choice of test depending on whether data were normally distributed. In case of nonparametric data, the Friedman test followed by the Wilcoxon signed rank test for paired replicates was used. Significance of differences between different groups of rats was calculated by either Student's t test or the Wilcoxon test, depending on whether data were normally distributed. All tests used were two-sided; a value of p < 0.05 was considered significant.


Pharmacokinetics of phenobarbital in Sprague–Dawley rats

PB was rapidly and dose-dependently absorbed after i.p. administration in female Sprague–Dawley rats (Fig. 1). Elimination half-lives determined from the decay of plasma levels ranged between 12.7 and 20.4 h (mean ± SEM, 16.9 ± 1.43 h) without any apparent differences between doses. From these data, twice-daily application of PB with a bolus dose of 25 mg/kg and maintenance doses of 15 mg/kg, were calculated for maintenance of therapeutic plasma levels (10–40 μg/ml), which was confirmed in a preliminary experiment with this dosing protocol (see Methods).

Figure 1.

Plasma concentrations of phenobarbital in female Sprague–Dawley rats after i.p. injection of single doses, ranging from 20 to 30 mg/kg. Each curve is from one individual rat.

Effect of phenobarbital on spontaneous seizures

Average plasma concentrations determined 10 h after the bolus administration of PB (25 mg/kg) in rats with SRSs were 25.1 μg/ml (range, 22.2–28.1 μg/ml). Twelve hours after the last PB injection of the treatment period, average plasma levels of 27.8 μg/ml (range, 14.0–37.3 μg/ml) were determined. Thus all rats had plasma levels within the therapeutic range (10–40 μg/ml) without significant differences between onset and end of the period of treatment.

Average seizure frequency in the 2 weeks before onset of treatment with PB was 34.3 ± 20.1 (Fig. 2A). This average seizure frequency tended to be increased by treatment with PB when all 11 rats with SRSs were used for calculation (Fig. 2A). After termination of treatment, average seizure frequency (44.8 ± 26.8 seizures per 2 weeks) was comparable to that determined before onset of treatment (Figure 2A), indicating that no temporal effect on seizure frequency was present in the absence of drug treatment.

Figure 2.

Effect of phenobarbital (PB) on spontaneous recurrent seizures (SRSs) in rats. SRSs were recorded over a period of 2 weeks before onset of PB treatment (predrug control), followed by drug treatment for 2 weeks, and then a 2-week postdrug control period. All data are shown as mean ± SEM. A–C: Average number of seizures recorded over the three 2-week periods. D: Average plasma concentration of PB from the blood samples taken at the end of the treatment period. A: Average seizure data from the 11 rats of this experiment are given. B: Respective data from the seven responders. C: Data from the four nonresponders of this group (see text for definitions). In the responder group, PB significantly suppressed SRSs compared with the pre- and postdrug periods (*p < 005).

However, by averaging data from all 11 rats, as shown in Figure 2A, striking differences in individual response to PB were masked. Therefore the individual data of all 11 rats are shown in Figures 3 to 5. Furthermore, the types of seizures (convulsive vs. nonconvulsive) are shown for each rat. Six rats were completely protected from seizures during PB treatment (Figs. 3 and 4). These rats were considered responders. Furthermore, one rat (DS 117) exhibited only one seizure (at the first day of treatment) during the 2 weeks of treatment with PB (Fig. 4), corresponding to 94% reduction of seizure frequency compared with predrug control. This rat therefore also was considered a responder. Average data from these seven responders are shown in Figure 2B, demonstrating a significant anticonvulsant effect of PB in this group, both compared with predrug (p = 0.0007) and postdrug (p = 0.0437) seizure frequency. Postdrug seizure frequency in this group was lower than predrug seizure frequency (p = 0.0263; Fig. 2B), which could be related to a carryover effect after PB treatment in the responder group. However, at least some of the rats of this group exhibited SRSs shortly after termination of treatment with PB (e.g., rat DS 108 in Figure 3 and DS 119 in Figure 4).

Figure 3.

Effect of phenobarbital (PB) on spontaneous recurrent seizures (SRSs) in rats. SRSs were recorded over a period of 2 weeks before onset of PB treatment (predrug control), followed by drug treatment for 2 weeks, and then a 2-week postdrug control period. The figure shows the occurrence of seizures over the 6 experimental weeks for four individual rats. PB completely suppressed seizures during treatment, so that these rats were considered responders. Solid bars, Convulsive seizures; open bars, nonconvulsive seizures.

Figure 4.

Effect of phenobarbital (PB) on spontaneous recurrent seizures (SRSs) in rats. SRSs were recorded over a period of 2 weeks before onset of PB treatment (predrug control), followed by drug treatment for 2 weeks, and then a 2-week postdrug control period. The figure shows the occurrence of seizures over the 6 experimental weeks for three individual rats. PB completely suppressed seizures in two rats and by >50% in one rat, so that these rats were considered responders. Solid bars, Convulsive seizures. Open bars, Nonconvulsive seizures.

Figure 5.

Effect of phenobarbital (PB) on spontaneous recurrent seizures (SRSs) in rats. SRSs were recorded over a period of 2 weeks before onset of PB treatment (predrug control), followed by drug treatment for 2 weeks, and then a 2-week postdrug control period. The figure shows the occurrence of seizures over the 6 experimental weeks for four individual rats. PB did not suppress SRSs in any rat, so that these rats were considered nonresponders. Solid bars, Convulsive seizures. Open bars, nonconvulsive seizures.

In three other rats (DS 107, DS 110, DS 118), PB did not exert any anticonvulsant effect (Fig. 5), so that these rats were considered nonresponders. Furthermore, rat DS 113 was considered a nonresponder because seizure frequency during treatment was not reduced by >50% compared with predrug control (Fig. 5). The average seizure frequency of these four nonresponders is shown in Fig. 2C. Interestingly, in one of these rats (DS 107), the types of seizures changed during treatment from convulsive to nonconvulsive (Fig. 5), whereas most seizures recorded in the other rats during the control or treatment periods were generalized convulsive (stage 4–6) seizures.

As shown in Figure 2, responders and nonresponders differed in their average seizure frequency during the control periods, which was considerably higher in nonresponders. This, however, was predominantly due to one rat (DS 118), which exhibited an extremely high seizure frequency during all periods of the experiment (Fig. 5). This rat also differed in its behavior from all other rats, in that it was extremely aggressive (toward persons handling the animal) and hyperexcitable, which was not observed in the other rats of this experiment.

The different individual responses to PB were not due to pharmacokinetic differences, as illustrated by the average plasma levels of PB in responders and nonresponders shown in Figure 2D. Although a much wider interindividual variation existed in plasma PB levels at the end compared with the beginning of the treatment phase, this was not correlated with lack of seizure control in individual rats. For instance, the nonresponder DS 107 (Fig. 5) had a plasma level of 32.7 μg/ml at the end compared with 24.1 μg/ml at the beginning of the treatment phase. Furthermore, all rats received PB at maximal tolerated doses, as indicated by the marked sedation that was observed in all rats during treatment.

The spontaneous seizures were not equally distributed over 24 h per day but occurred significantly more frequently during the light phase. Thus the average number of seizures recorded over the whole experiment was 92 ± 55 during the day versus 62 ± 39 during the night (p = 0.0098), indicating that ∼60% of all seizures occurred during the day. As reported recently (11), a wide variation in baseline seizure frequency was found between the rats (two to 236 seizures over a 2-week period). A similar variation in seizure frequency has previously been found in a post-SE model of TLE in which SE was induced by sustained electrical stimulation of the lateral nucleus of the amygdala (16). Interindividual differences in seizure frequency were associated with the extent of morphologic brain alterations, in that rats with frequent seizures showed more intense mossy fiber sprouting in the hippocampus than did rats with rare seizures (16). Whether a similar association also is seen in rats of the present post-SE model of TLE remains to be investigated.


The present experiments demonstrate that the individual response to an AED of rats with SRSs markedly vary, although the drug is given at maximal tolerated doses with plasma levels at the upper border of therapeutic range. The data also show that continuous video-EEG recording for 24 h/day, 7 days/week is an absolute requirement for correct identification of responders and nonresponders to AED treatment because of large interindividual variation in seizure occurrence and seizure frequency across experimental weeks. This may explain, at least in part, why Leite and Cavalheiro (8) did not detect any marked interindividual variation in AED response in male Wistar rats with SRSs, because in their experiments EEG recording was done only for 2 h, 3 days per week. Instead of continuous video recording, Leite and Cavalheiro (8) used direct animal observation by three persons from 8:00 a.m. to 6:00 p.m per day, 5 days per week, and rats with low seizure frequency were excluded from pharmacologic studies. PB was given at 20 mg/kg, s.c., twice daily in eight rats with SRSs for 2 weeks, resulting in complete remission of seizures in four rats and a marked suppression of seizure frequency in the remaining animals (8). Similar effects were obtained with CBZ (40 mg/kg, t.i.d.), PHT (50 mg/kg, b.i.d.), and VPA (200 mg/kg, t.i.d.), whereas a lower dose of VPA (150 mg/kg, t.i.d.) failed to reduce seizure frequency in most rats (8). Apart from the effort used for monitoring of SRSs, several other differences were noted in the present study, including the rat strain and gender and the method used to produce SRSs. It is not known yet whether rats from different strains and gender differ in the response of their SRSs to AEDs. Furthermore, it is not known whether the method to produce SRSs (e.g., pilocarpine vs. electrically induced SE) affects the pharmacologic sensitivity of SRSs developing with a latency of ∼4 weeks after the SE. The fact that we recently described AED responders and nonresponders for SRSs developing after pilocarpine-induced SE in Wistar rats (10) argues against factors such as strain or method to produce SRSs but rather indicates that the effort used for monitoring SRSs is a major factor in correctly identifying the individual response of rats to treatment. Furthermore, the AED doses used are certainly important. The high doses of AEDs administered in the study of Leite and Cavalheiro (8) induced marked adverse effects, such as muscle relaxation. In this regard, it is important to consider that pharmacoresistance in patients with epilepsy is usually defined as no clinically relevant improvement of seizures at maximal tolerated doses of standard AEDs (1). This definition of drug resistance also should be used in animal models, because an anticonvulsant effect of a drug at a toxic dose is without relevance in terms of prediction of clinical efficacy.

Using the maximal electroshock seizure (MES) and pentylenetetrazole (PTZ) seizure tests in mice and rats, we previously showed that the use of doses leading to plasma concentrations similar to the therapeutic plasma concentration range known from patients with epilepsy results in clinically predictive data, in that the drugs are effective in rodents in the same plasma concentration range as in humans, although doses (in milligrams per kilogram) needed to achieve these drug levels in rodents are generally much higher than respective doses in humans because of more rapid elimination of AEDs by rodents (17,18). However, in contrast to the previous (10) and present studies on rats with SRSs, drug testing in the MES and PTZ tests does not allow delineation of interindividual differences in drug response. Furthermore, the pharmacology of elicited and spontaneous seizures may differ (19). Thus a model with SRSs provides pharmacologic data on AED efficacy that cannot be obtained with simple seizure models.

Recently, pharmacoresistance in animal models of epilepsy was defined as persistent seizure activity that does not respond, or responds poorly, to monotherapy at tolerable doses with at least two current AEDs (4). The finding of the present study that 36% of the rats with SRSs did not respond adequately to treatment with maximal tolerable doses of PB is intriguing but does not yet fulfill the aforementioned criteria of pharmacoresistance. To fulfill these criteria, rats that do not respond to PB should be tested with at least one other current AED. However, in patients with newly diagnosed epilepsy not adequately responding to AED monotherapy, only 14% became seizure free during treatment with a second or third drug, indicating that inadequate response to initial treatment is a predictor of intractable epilepsy (20). It will be important to study whether the same is true for post-SE models of TLE such as the model used in the present experiments.

A possible bias in studies with prolonged drug administration in post-SE models of TLE is the natural history of untreated epilepsy in such models. Thus any temporal effect on seizure frequency, severity, or type over the period of the drug study would affect the outcome of the study. Ideally, one would need a parallel control group with no treatment against which the effect of drug could be compared. However, for the present model, repeated video-EEG SRSs monitoring of untreated rats over a period of up to about a year after induction of SE gave no evidence for any decrease in seizure frequency or severity of SRSs (ref. 11 and unpublished experiments), so that we consider that the natural history of untreated epilepsy was not a bias for the present experiments.

Using the amygdala-kindling model of TLE, we recently showed that rats not responding to PHT also have an inadequate response to most other AEDs (21). Experiments in such kindled rats have indicated that the individual genetic background of a rat and kindling-induced alterations determine whether a rat becomes a responder or nonresponder (21). With respect to kindling-induced alterations, overexpression of multidrug transporters, such as P-glycoprotein (Pgp) in capillary endothelial cells of the blood–brain barrier, which restrict the brain uptake of many lipophilic drugs, would be a likely explanation for the fact that kindled nonresponders show a reduced response to various AEDs with different mechanisms of action (21,22). Indeed, we recently reported that kindled nonresponders exhibit significantly increased Pgp expression in the kindled amygdala (23). Whether similar mechanisms are involved in the different response of rats to PB in the present model of TLE is not known, but recent experiments in rats with SRSs developing after an electrically induced SSSE showed that the messenger RNA (mRNA) of the gene (mdr1) encoding Pgp is overexpressed in such animals (24).

Drug trials in rats with SRSs are difficult and resource intensive, which is the most likely explanation for only few studies on AED testing in such models being available. The present study with PB shows that interesting results can be obtained by drug testing in epileptic rats, which are much closer to data from clinical evaluation of AEDs in patients with epilepsy than are data from brief drug testing in traditional seizure models, in which only elicited seizures are examined. However, the current study design has limitations and shortcomings, particularly because of the labor-intensive nature of the present model. Thus only a small number of rats were used, and a nonrandomized, crossover design was used. As discussed earlier, a parallel-group design should be evaluated in future studies.

In addition to using models with SRSs for evaluation of AED effects during prolonged treatment, these models also can be used to evaluate antiepileptogenic drug effects by starting treatment shortly after the SE (4,19). With respect to PB, it is interesting to note that this drug has previously been evaluated for antiepileptogenic or disease-modifying effects in two different post-SE models of TLE (25–27). In two studies using the kainate model of TLE, prolonged PB treatment starting shortly after SE did not prevent the development of SRSs, brain damage, behavioral alterations, and deficits in learning and memory (25,26). More recently, Prasad et al. (27) reported that PB, but not PHT, improved the long-term outcome of SE induced by sustained hippocampal stimulation in rats. This, however, may have been due to initial insult modification rather than to an antiepileptogenic effect, because PB prevented development of SRSs only when given 1 h after onset of sustained hippocampal stimulation, whereas administration after 2 or 4 h was ineffective in this regard (27). In clinical trials, PB failed to prevent epileptogenesis (28), which is in line with the findings from the kainate model (25,26). Improved understanding of the molecular mechanisms underlying epileptogenesis is likely to lead to new pharmacologic approaches for prevention of epilepsy that can be evaluated in post-SE models of TLE (19,29–32). Furthermore, as indicated by our present and previous studies (10), such models may be used to investigate mechanisms of pharmacoresistance and to evaluate new AEDs for efficacy against difficult-to-treat seizures. Devices for prolonged drug delivery, such as osmotic minipumps, and miniaturized systems for long-term recording of SRS, such as EEG monitoring via telemetric devices, may make long-term studies in post-SE models of TLE less labor dependent and increase the use of epileptic animals for evaluation of anticonvulsant and antiepileptogenic drug effects (4).


Acknowledgment:  We thank Christiane Bartling and Nicole Ernst for skillful technical assistance. The study was supported by a grant (Lo 274/9-2) from the Deutsche Forschungsgemeinschaft (Bonn, Germany).