William R. Garnett PharmD
Professor of Pharmacy and Neurology, Deparment of Pharmacy, Virginia Commonwealth University, Medical College of Virginia, Richmond, Virginia
Drug interactions occur frequently with the first-generation antiepileptic drugs (AEDs), such as phenobarbital, phenytoin, carbamazepine, and valproic acid (2). Therefore, the early development of lamotrigine evaluated the effects of other AEDs on lamotrigine, and these effects were included in the initial study designs. Although it has been demonstrated that other AEDs can increase and decrease the metabolism of lamotrigine, lamotrigine is not metabolized by the cytochrome P450 pathways and is less likely to interfere with the metabolism of other drugs. The low protein binding of lamotrigine also makes it less likely to interact with other drugs. More recently, some pharmacodynamic interactions between lamotrigine and other AEDs that may result in toxicity or synergy have been identified. Therefore, it is important to review the drug interaction profile of lamotrigine.
PHARMACOKINETICS OF LAMOTRIGINE
A review of the clinical pharmacokinetics of lamotrigine is useful in evaluating the potential for and possible mechanisms of the pharmacokinetic drug interactions associated with lamotrigine. Studies in normal volunteers established that lamotrigine reaches a peak concentration (Cmax) in 1 to 3 hours after an oral dose. An absolute bioavailability of 97.6% was determined by comparing the area under the concentration-time curve (AUC) after intravenous and oral doses. Thus, lamotrigine is rapidly and completely absorbed. Lamotrigine demonstrates linear pharmacokinetics, with AUC and Cmax increasing proportionally with dose. This indicates that there is no significant autoinduction or saturable metabolism. The half-life of lamotrigine is approximately 22 hours in drug-naive individuals and the clearance is approximately 42 mL/min. Although there is significant intersubject variability in clearance, there is little intrasubject variability (41). Approximately 70% of a dose of lamotrigine is eliminated by the phase II metabolic pathway (i.e., by glucuronide conjugation) in the liver (21). There is no first-pass metabolism, which means that alterations in liver blood flow do not change metabolism. The metabolism of lamotrigine may be moderately reduced in people with Gilbert's syndrome (21). A population pharmacokinetics study of patients aged 14 to 74 years receiving lamotrigine demonstrated that the apparent oral clearance was not significantly influenced by body weight, age, sex, oral contraceptives, and dose (29). Whereas a mild autoinduction effect (17.3%) and a lower clearance in Asians versus whites (28.7%) have been reported, these effects are not thought to be clinically significant and do not warrant dosage adjustments (29). Renal failure does not alter the pharmacokinetics of lamotrigine (50). Lamotrigine is only approximately 55% bound to plasma proteins, indicating a low potential for displacement with highly (>90 %) protein-bound drugs. The volume of distribution (VD) is approximately 1.1 L/kg (21). The weight-normalized clearance and VD are higher in children. Therefore, children will likely require higher weight-normalized doses at the same dosing frequency to achieve comparable serum concentrations (13) (Table 36.1).
PHARMACOKINETIC EFFECTS OF OTHER DRUGS ON LAMOTRIGINE
Effects of Antiepileptic Drugs on Lamotrigine
Several studies have shown that the clearance of lamotrigine is enhanced by coadministration with AEDs that can induce liver enzymes and reduced by drugs that inhibit liver enzymes (Table 36.2). Armijo et al. used bivariate and multivariate methods to analyze retrospectively the influence of patient age and the use of concomitant AEDs on the lamotrigine concentration-to-dose (C/D) ratio in samples from
164 patients (3). The lamotrigine C/D ratio increased with age in patients receiving lamotrigine alone, but decreased with age in patients receiving lamotrigine with enzyme inducers. The lamotrigine C/D ratio increased in patients taking valproic acid. The lamotrigine C/D ratio was 10 times lower in patients receiving lamotrigine and inducers than in those receiving lamotrigine and valproic acid.
TABLE 36.1. CLINICAL PHARMACOKINETICS OF LAMOTRIGINE
Bottiger et al. reviewed 149 lamotrigine samples from 104 adult patients to determine the effects of drug interactions on lamotrigine (10). In 20 patients on monotherapy, the C/D ratio was 65 (range, 50 to 84) nmol/L/mg. In 37 patients with concomitant carbamazepine treatment, the C/D ratio was 31 nmol/L/mg, which was less than half that of patients on monotherapy, and in 14 patients on phenytoin, it was even lower (17 nmol/L/mg). In a few patients taking phenobarbital with lamotrigine, the C/D ratio was slightly less than with monotherapy. In the 13 patients taking valproic acid concurrently with lamotrigine, the C/D ratio was significantly increased to 251 nmol/L/mg. In patients taking valproic acid with either carbamazepine or phenytoin and lamotrigine, the C/D ratio was slightly above that of monotherapy for lamotrigine.
TABLE 36.2. PHARMACOKINETIC EFFECTS OF OTHER DRUGS ON LAMOTRIGINE
Grasela et al. used NONMEN, a population pharmacokinetics computer program, to pool and analyze the plasma concentrations of lamotrigine obtained from the three adult clinical trials conducted in the United States (26). A total of 2,407 lamotrigine plasma concentrations from 527 patients were analyzed to determine the effect of body size, age, sex, race, and use of concomitant AEDs. The population mean apparent oral clearance of lamotrigine in adult patients receiving one concomitant enzyme-inducing AED and not valproic acid was estimated to be 1 mL/min/kg. The clearance of lamotrigine was increased by 13% in patients receiving more than one concomitant enzyme-inducing AED. Lamotrigine did not affect the clearance of other AEDs.
Battino et al. evaluated the effects of age and concomitant treatment on plasma lamotrigine C/D ratios in 482 samples from 106 chronically treated patients with epilepsy (5). A linear C/D relationship was observed in individual patients, but there was no correlation between the administered lamotrigine dose and plasma concentration in the cumulative analysis. Concurrent AED therapy affected the lamotrigine C/D ratio, which was significantly higher in the patients receiving valproic acid and significantly lower in those treated with enzyme-inducing AEDs. The lamotrigine C/D ratios significantly increased with increasing plasma valproic acid concentrations and significantly decreased with increasing phenytoin concentrations. The effect of enzyme-inducing AEDs increased with the number of concomitant drugs.
In a prospective study by Bartoli et al., there was more enzyme induction than inhibition in children 3 to 6 years of age than in older children (4). These investigators evaluated 45 patients 3 to 38 years of age who received lamotrigine as adjunctive therapy for uncontrolled seizures. In the
patients receiving enzyme-inducing AEDs, the lamotrigine concentrations normalized to a 1 mg/kg daily dosage were lower in children aged 3 to 6 years than in older children, adolescents, and adults. Although valproic acid increased the concentrations of lamotrigine, the age effect was less evident. In any age group, the dose-normalized lamotrigine concentrations were approximately fivefold higher in patients comedicated with valproic acid than in those comedicated with enzyme inducers.
May et al. investigated the influence of carbamazepine, phenytoin, phenobarbital, valproic acid, and combinations of these drugs on the serum concentration of lamotrigine in 588 blood samples taken from 302 patients (35). The lamotrigine serum concentration in relation to lamotrigine dose/body weight [level-to-dose ratio (LDR), in µg/mL/mg/kg] was calculated and compared for different drug combinations. The results showed that comedication had a significant effect on the lamotrigine serum concentrations. The mean LDRs for lamotrigine were 0.32 (lamotrigine + phenytoin), <0.52 (lamotrigine + phenobarbital), ≈0.57 (lamotrigine + carbamazepine), <0.98 (lamotrigine mono), ≈0.99 (lamotrigine + valproic acid + phenytoin), <1.67 (lamotrigine + valproic acid + carbamazepine), ≈1.80 (lamotrigine + valproic acid + phenobarbital), and <3.57 (lamotrigine + valproic acid). The mean lamotrigine concentrations in patients on comedication with valproic acid were approximately two times higher than in patients on lamotrigine monotherapy or on comedication without valproic acid.
Eriksson et al. assessed the interactions of lamotrigine with other AEDs in 31 children (19). The median elimination half-life in patients receiving concomitant valproic acid was 43.3 hours, in patients receiving carbamazepine and/or phenobarbital, it was 14.1 hours, and in patients receiving both valproate and carbamazepine/phenobarbital or other AEDs, it was 28.9 hours.
Valproic acid inhibits the metabolism of lamotrigine. The half-life of a single dose of lamotrigine given to four patients stabilized on valproic acid was 59 ± 26 hours. The half-life of lamotrigine was 69.6 ± 14.8 hours in 18 healthy volunteers who took valproic acid for 70 days and added 50-, 100-, and 150-mg doses of lamotrigine for 7 days in a random fashion (51). In a study of six normal volunteers who received single doses of 100 mg of lamotrigine and six doses of 200 mg of valproic acid every 8 hours beginning 1 hour before the lamotrigine, Yuen et al. reported a 21% decrease in the clearance of lamotrigine, with an increase in the half-life of lamotrigine from 37.4 to 48.3 hours and an increase in the AUC of lamotrigine from 70.9 to 91.8 µg/mL/hr (53). The reduction in elimination occurred within the first hour. Cmax, tmax, and renal elimination were not altered. The proposed mechanism for the interaction was competition between valproic acid and lamotrigine for glucuronidation. The effects detected in this study may not have been maximal because the dose of valproic acid was not at steady state (53). Anderson et al. evaluated the steady-state pharmacokinetics of lamotrigine and valproic acid with three different doses of lamotrigine in normal volunteers (1). Eighteen normal male volunteers received 500 mg orally of valproic acid twice a day throughout the study. Each subject randomly received 50, 100, and 150 mg of lamotrigine for 1 week each, with a 2-week washout period between lamotrigine treatment periods. Concomitant valproic acid markedly increased the half-life of lamotrigine and decreased lamotrigine clearance without altering the linear pharmacokinetics of lamotrigine (1). In a study using population pharmacokinetics to describe the pharmacokinetics of lamotrigine in developmentally disabled patients, the clearances of lamotrigine as monotherapy, lamotrigine plus inducers, and lamotrigine plus valproic acid were 0.69 ± 0.2, 1.60 ± 0.65, and 0.2 ± 0.05 mL/kg/min, respectively (22). The addition of the valproic acid concentration to the model used to define the clearance of lamotrigine did not significantly improve the estimates of clearance. This suggests that valproic acid inhibition of lamotrigine is maximal within the usually accepted target ranges for valproic acid (22). Kanner and Frey evaluated the clearance of lamotrigine in 28 patients with intractable epilepsy who were treated with a combination of lamotrigine and valproic acid (31). Correlations between lamotrigine clearance and the dose and steady-state concentration of valproic acid demonstrated that the degree of inhibition of lamotrigine clearance is independent of the dose and steady-state concentration of valproic acid (31). Chen also used population pharmacokinetics to establish a basis for dosage recommendations for lamotrigine in children and reported that to achieve the same concentrations, children receiving enzyme-inducing AEDs without valproic acid require higher doses of lamotrigine than those receiving valproic acid (12). Also, heavier children require higher doses (12). May et al. found that valproic acid increases the concentration of lamotrigine by 211% (34). The combination of lamotrigine with enzyme-inducing AEDs and valproic acid reduces some of the effect of valproic acid on the inhibition of lamotrigine metabolism (30). Therefore, if valproic acid is discontinued from the regimen of a patient taking lamotrigine, the concentrations of lamotrigine will decrease, necessitating a dosage increase (Tables 36.3 and 36.4).
Two studies have evaluated the effects of methsuximide on the pharmacokinetics of lamotrigine. Besag et al. evaluated lamotrigine serum levels in 16 patients taking methsuximide before starting or after stopping methsuximide (6). Methsuximide lowered the concentration of lamotrigine in every case, with a mean decrease of 53% (range, 36% to 72%) (6). May et al. retrospectively evaluated 376 blood samples from 222 patients to determine the effects of methsuximide and oxcarbazepine on lamotrigine concentrations (34). Methsuximide was found to have a strong inducing effect on the metabolism of lamotrigine, and decreased the lamotrigine concentration by approximately 70% compared with lamotrigine monotherapy. Methsuximide could
attenuate most of the inhibitory effect of valproic acid on the metabolism of lamotrigine. When given with lamotrigine alone, valproic acid increased the lamotrigine concentration by 211%, but the increase was only 8% when methsuximide was given concomitantly with valproic acid and lamotrigine (34).
TABLE 36.3. LAMOTRIGINE: ADULT DOSING
In addition to finding an effect of methsuximide on the pharmacokinetics of lamotrigine, May et al. found an enzyme-inducing effect of oxcarbazepine, which they estimated to be 29% (34). The enzyme-inducing effect of oxcarbazepine was less than that of carbamazepine, which was 54%. Oxcarbazepine reduced some of the inhibitory effect of valproic acid on the metabolism of lamotrigine. Valproic acid increased the lamotrigine concentration by 211% when given alone, but increased the concentration only by 111% when given with oxcarbazepine (34).
Gidal et al. compared the pharmacokinetics of lamotrigine in six patients concomitantly receiving felbamate with five patients receiving lamotrigine as monotherapy (23). There was no statistically significant difference in either apparent lamotrigine oral clearance (0.026 ± 0.005 L/kg/hr versus 0.0924 ± 0.01 L/kg/hr) or in mean elimination half-life (33.7 ± 7.5 hours versus 40.2 ± 15.05 hours). It was concluded that felbamate did not have a significant effect on lamotrigine pharmacokinetics (23). Glue et al. analyzed the degree of agreement between in vivo interaction studies performed in patients with epilepsy and healthy individuals and in vitro studies that identified the cytochrome P450 enzymes inhibited by felbamate (25). They found no interactions between felbamate and lamotrigine (25). Colucci et al. administered 1,200 mg of felbamate every 12 hours to normal volunteers who also took 100 mg of lamotrigine every 12 hours for 10 days in a double-blind, randomized, placebo-controlled, two-way crossover study (15). Although there was a 13% increase in Cmax and a 14% in the AUC of lamotrigine when it was given with felbamate compared with placebo, the 90% confidence intervals of the pharmacokinetic parameters were within the 80% to 125% bioequivalance limits, and it was determined that felbamate had no clinically relevant effects on the pharmacokinetics of lamotrigine (15).
TABLE 36.4. LAMOTRIGINE: PEDIATRIC DOSING
Effects of Nonantiepileptic Drugs on Lamotrigine
Rifampicin is a potent inducer of cytochrome P450 and of the uridine 5′-diphosphate-glucuronyl transferase enzyme systems, and cimetidine is an inhibitor of the cytochrome P450 enzyme system. The effects of these drugs on the pharmacokinetics of lamotrigine were evaluated by Ebert et al. in 10 normal volunteers who received a single 25-mg oral dose of lamotrigine after pretreatment with either 400 mg of cimetidine twice a day, 600 mg of rifampicin once a day, or placebo (17). Rifampicin significantly increased the clearance-over-bioavailability ratio (5.13 ± 1.05 L/hr versus 2.60 ± 0.40 L/hr) and the amount of lamotrigine excreted
as glucuronide (12.12 ± 0.94 mg versus 8.90 ± 0.77 mg) compared with placebo, whereas both the half-life (14.1 ± 1.7 hours versus 23.8 ± 2.1 hours) and the AUC (396.24 ± 60.18 µg/mL/min versus 703.99 ± 82.31 µg/mL/min) were significantly decreased compared with placebo. Cimetidine did not alter the pharmacokinetics of lamotrigine. The authors concluded that rifampicin induced the enzymes responsible for the glucuronidation of lamotrigine, but that cimetidine had negligible effects (17).
Because both lamotrigine and acetaminophen are metabolized by hepatic glucuronidation, the potential for an interaction exists. A study was done in eight normal volunteers who took a single 300-mg dose of lamotrigine before and after taking acetaminophen 900 mg three times a day for 11 days. There was a 15% increase in the clearance of lamotrigine, a 15% decrease in the half-life, and a 20% decrease in the AUC after acetaminophen treatment (16). This suggests that acetaminophen may stimulate the metabolism of lamotrigine. However, the differences are small and within the intersubject differences.
Kaufman and Gerner reported two patients who had sertraline added to their lamotrigine regimen (32). In the first case, a 25-mg dose of sertraline that did not produce detectable blood levels of sertraline or its metabolite resulted in a doubling of the concentration of lamotrigine, with symptoms of side effects. In the second case, a 25-mg reduction in the dose of sertraline resulted in a 50% reduction in the concentration of lamotrigine despite a 33% increase in the lamotrigine dosage. The authors hypothesize that sertraline inhibits the glucuronidation of lamotrigine (32).
PHARMACOKINETIC EFFECTS OF LAMOTRIGINE ON OTHER DRUGS
Effect of Lamotrigine on Other Antiepileptic Drugs
The potential for lamotrigine to induce mixed-function oxygenase enzymes was evaluated in normal volunteers (39). In a randomized, double-blind, parallel-group study, nine normal volunteers received 100 mg of lamotrigine twice a day for 14 days and nine subjects received placebo. The clearance of antipyrine, the 48-hour urinary excretion of 6-β-hydroxycortisol, and the plasma γ-glutamyl transferase concentrations were assessed before treatment, within 24 hours of the last dose, and 14 days later. There were no significant changes in the clearance of antipyrine or 6-β-hydroxycortisol, and no significant changes in the plasma γ-glutamyl transferase concentrations. The authors concluded that it is unlikely that lamotrigine induces mixed-function oxygenase enzymes to an extent that would result in drug interactions (39). Lamotrigine is only approximately 56% protein bound, and it is unlikely to alter the binding of other highly protein-bound drugs (e.g., phenytoin, valproic acid) or to be affected by them. This was confirmed in an in vitro study using equilibrium dialysis (36). The binding of lamotrigine was constant over the concentration of 1 to 3 µg/mL and was unaffected by concentrations of phenytoin, phenobarbital, or valproic that were in the target range (36).
TABLE 36.5. PLASMA ANTIEPILEPTIC DRUG CONCENTRATIONS POOLED ACROSS FOUR STUDIES
Data collected during the clinical trials of lamotrigine showed no changes in the concentrations of carbamazepine, phenytoin, valproic acid, primidone, phenobarbital, or clobazam (Table 36.5). In the NONMEN analysis of the 2,407 samples from the 527 patients in the three adult trials of lamotrigine in the United States, it was documented that lamotrigine had no effect on the plasma levels of phenytoin or carbamazepine (26). In the previously mentioned study by Eriksson et al. (19), there were no clinically important changes in the plasma levels of carbamazepine, valproic acid, ethosuximide, or phenobarbital after the addition of lamotrigine. There was a reduction in the plasma concentration of clonazepam when lamotrigine was added (19).
There have been anecdotal reports of an increase in the concentration of carbamazepine or carbamazepine-10,11-epoxide in patients receiving lamotrigine. Warner et al. reported that the concentration of carbamazepine increased from 28 ± 6 µmol/L before lamotrigine therapy to 32 ± 10 µmol/L after lamotrigine (48). The carbamazepine epoxide concentrations also increased from 5.6 ± 2.5 µmol/L to 7.9 ± 3.5 µmol/L. However, there was a wide range of epoxide
concentrations (48). Graves et al. reported three case histories in which there was an increase in the epoxide metabolite without a change in the parent carbamazepine concentration (27). Potter and Donnelly reported that phenytoin and lamotrigine caused a relative increase in the carbamazepine-10,11-epoxide concentration and a significant decrease in the ratio of carbamazepine to the epoxide (from 5 to 3) (40). If valproic acid also was present, the concentration of parent and metabolite increased significantly, causing potential toxicity (40). However, more controlled trials have failed to demonstrate an effect of lamotrigine or its metabolite (44). In one study, 11 patients taking carbamazepine took lamotrigine or placebo in a controlled crossover study. The carbamazepine concentrations were 8.3 µg/mL during the lamotrigine period and 8.7 µg/mL during the placebo period. The carbamazepine epoxide concentrations were 2.0 µg/mL during the lamotrigine period and 2.1 µg/mL during the placebo period. None of the differences was statistically significant (45). In another placebo-controlled study, Stolarek et al. evaluated the plasma concentrations of carbamazepine and epoxide in 22 patients during the placebo period and during the period when lamotrigine was added to the patients' regimen (47). They found no significant changes in the concentrations of either carbamazepine or its metabolite (47). More recently, Besag et al. evaluated the concentrations of carbamazepine and its active epoxide metabolite in 47 patients with escalating doses of lamotrigine. There was no significant change in the serum concentrations of either the parent drug or its metabolite (8). Eriksson and Boreus also failed to find a change in the concentration of carbamazepine or carbamazepine-10,11-epoxide (18). They studied 11 children and 3 adolescents who had been treated for more than 1 year with carbamazepine when lamotrigine was titrated as concomitant therapy. Lamotrigine had no effect on mean carbamazepine concentrations, but lamotrigine decreased the carbamazepine-10,11-epoxide concentrations slightly (6.4 ± 2.6 µmol/L to 4.9 ± 2.4 µmol/L) (18). Pisani et al. compared the pharmacokinetics of a single dose of carbamazepine-10,11-epoxide in 10 patients on chronic monotherapy with lamotrigine and in 10 normal volunteers (38). The pharmacokinetic parameters of the epoxide in the patients were similar to those in the normal volunteers, indicating that lamotrigine had no effect on the disposition of carbamazepine-10,11-epoxide (38). Gidal et al. evaluated the apparent oral clearance of carbamazepine and the steady-state ratio of carbamazepine epoxide to carbamazepine in nine patients with epilepsy before and after the initiation of adjunctive treatment with lamotrigine (24). The apparent oral clearance of carbamazepine did not change after the initiation of lamotrigine (5.58 ± 1.60 L/hr versus 5.81 ± 1.74 L/hr), and the ratio of carbamazepine epoxide to carbamazepine did not change (0.241 ± 0.082 versus 0.232 ± 0.082) (24). Thus, there is no clear evidence that lamotrigine affects the concentrations of carbamazepine or its metabolite. The data from controlled trials support the notion that there is no significant pharmacokinetic interaction between lamotrigine and carbamazepine.
The study by Anderson et al. was a bidirectional study of the interaction between valproic acid and lamotrigine (1). In addition to finding that valproic acid significantly increased the half-life and decreased the clearance of lamotrigine, it also was reported that lamotrigine caused a 25% decrease in the steady-state valproic acid plasma concentration. The oral clearance of valproic acid was increased from 7.2 ± 1.1 mL/hr/kg before lamotrigine treatment to 9.0 ± 2.0 mL/hr/kg on day 28 of lamotrigine therapy. There was no change in the formation of the metabolite of valproic acid believed to be hepatotoxic (1).
Effect of Lamotrigine on Nonantiepileptic Drugs
Because enzyme-inducing AEDs can cause oral contraceptive failure, an interaction between lamotrigine and oral contraceptives was evaluated in 12 healthy women taking an oral contraceptive containing ethinyl estradiol and levonorgestrel. The subjects took lamotrigine 150 mg/day for 14 days. There were no significant changes in the mean plasma concentrations of ethinyl estradiol or levonorgestrel or in the urinary excretion of 6-β-hydroxycortisol (28). Thus, there is no interaction between lamotrigine and oral contraceptives.
Chen et al. studied 20 normal volunteers who took 2 g of lithium gluconate anhydrous every 12 hours for 5 days and in the morning of the sixth day with or without 100 mg of lamotrigine. Lamotrigine did not cause any significant changes in the pharmacokinetics of lithium as determined by noncompartmental methods (14).
PHARMACODYNAMIC INTERACTIONS WITH LAMOTRIGINE
Warner et al. (48) and Graves et al. (27) published anecdotal case reports of patients who experienced carbamazepine side effects when lamotrigine was added, which they attributed to an increase in the carbamazepine or epoxide concentration. However, although there is an increase in central nervous system (CNS) side effects when lamotrigine is added to the regimen of a patient taking carbamazepine, there is a lack of controlled data to support a pharmacokinetic interaction between lamotrigine and carbamazepine. It has been postulated that the increase in CNS side effects is related to an interaction at the receptor site (i.e., a pharmacodynamic drug interaction) (Table 36.6) (49). Besag et al. evaluated the effect of escalating doses of lamotrigine in 47 patients taking carbamazepine (8). Although they did not find a change in the concentration of carbamazepine or
its epoxide, they did find that nine patients demonstrated clinical signs of CNS toxicity (e.g., diplopia and dizziness). The concentrations of lamotrigine were below those normally associated with CNS side effects. In seven of the nine patients experiencing toxicity, the carbamazepine concentration was >8 mg/L on the initiation of lamotrigine. The authors concluded that CNS toxicity is likely if the carbamazepine concentration is >8 mg/L when lamotrigine is added, and that the mechanism is a result of a pharmacodynamic interaction at the receptor site. They advised clinicians to be prepared to decrease, but not stop, the dose of carbamazepine when lamotrigine is added (8). Eriksson and Boreus evaluated 11 children taking carbamazepine who received lamotrigine as add on therapy (18). Two of these children developed diplopia, two experienced agitation, and one experienced an increase in seizures. None of these children had an increase in carbamazepine or carbamazepine epoxide concentrations. The authors also concluded that the increase in side effects seen with the addition of lamotrigine to the regimen of patients taking carbamazepine is the result of a pharmacodynamic effect (18). Therefore, clinicians should monitor patients on carbamazepine carefully when lamotrigine is added, and be prepared to decrease the dose.
TABLE 36.6. PHARMACODYNAMIC DRUG INTERACTIONS WITH LAMOTRIGINE
The most serious adverse reaction associated with lamotrigine is skin rash (33) (Table 36.6). This rash typically is maculopapular and appears in the first 4 weeks of therapy. However, it may progress to Stevens-Johnson syndrome (SJS) (43) or to toxic epidermal necrolysis (TEN) (9), which has been associated with death (46). Using a case-controlled methodology, Rzany et al. determined that 73 of 352 (21%) patients with SJS/TEN were taking an AED, whereas only 28 of 1,579 (2%) of the control group reported taking an AED (43). For individual AEDs, the univariate relative risk of SJS/TEN for 8 weeks or less of use was 57 for phenobarbital, 91 for phenytoin, 120 for carbamazepine, 25 for lamotrigine, and 24 for valproic acid (43). The incidence of rash is higher in patients who initiate lamotrigine therapy at a high dose, have a rapid dose escalation, and are taking valproic acid. Conversely, the rate of skin rash is lower in patients taking enzyme-inducing agents (52). Presumably, valproic acid increases the incidence of skin rash because it significantly increases the half-life and decreases the clearance of lamotrigine (2). The combination of lamotrigine and valproic acid also may be associated with an increase in tremor (42). The association of lamotrigine-induced skin rash with the use of valproic acid is not a sine qua non for not using the drugs in combination. In fact, there are data indicating that the two drugs are synergistic when used together (11). However, the initial dose of lamotrigine should be low and the rate of titration should be slow in a patient taking valproic acid (7) (Tables 36.3 and 36.4).
Numerous trials document that when lamotrigine is added to other AEDs there is a reduction in seizure frequency, and lamotrigine appears to be a broad-spectrum AED (33). However, there appears to be a synergistic response to the combination of lamotrigine and valproic acid. Ferrie and Panayiotopoulos reported the case of a 13-year-old patient with refractory myoclonic epilepsy who achieved seizure control only when lamotrigine was combined with valproic acid. They attributed this to a specific pharmacodynamic interaction (20). Brodie and Yuen published the results of a large, multicenter European trial of lamotrigine in 347 patients who had not achieved complete seizure control on valproic acid, carbamazepine, phenytoin, or phenobarbital as monotherapy (11). The response rate was better in patients with idiopathic tonic-clonic seizures (61%) than in patients with partial seizures (43%). The addition of lamotrigine to patients taking valproic acid resulted in a significantly better response (64%) than when added to carbamazepine (41%) or phenytoin (38%). A slow rate of dosage titration was associated with lower incidences of skin rash and patient withdrawal. The increased response for the combination of lamotrigine and valproic acid was better for both tonic-clonic and partial seizures. They stated that their data supported a therapeutic synergy between lamotrigine and valproic acid (11). Pisani et al. assessed the comparative therapeutic value of valproic acid, lamotrigine, and the combination of the two in 20 patients with complex partial seizures resistant to other AEDs (37). Patients were started on valproic acid and then given lamotrigine and then the combination for 3 months. The patients progressed to the next treatment only if they failed to respond to the previous treatment. There was a >50% reduction in seizure frequency in 3 of 20 patients given valproic acid and in 4 of 17 patients given lamotrigine. In the other 13 patients who were given the combination of lamotrigine and valproic
acid, 4 became seizure free and 4 had a 62% to 78% reduction in seizure frequency. In the patients responding to the combination, the doses and peak serum concentration of lamotrigine and valproic acid were lower than those during separate administration of either drug. These data also support a synergistic effect between lamotrigine and valproic acid (37) (Table 36.6).
Lamotrigine is not highly protein bound and is predominantly metabolized by glucuronidation pathways in the liver. Although lamotrigine has a mild stimulating effect on glucuronosyl transferase enzymes, it has no effect of cytochrome P450 isoenzymes. Therefore, the pharmacokinetics of lamotrigine support a reduced potential for drug interactions (2). Enzyme-inducing drugs (e.g., carbamazepine and phenytoin) increase the clearance and decrease the half-life of lamotrigine, whereas valproic acid significantly decreases the clearance and increases the half-life of lamotrigine. If phenytoin, carbamazepine, or valproic acid is given in combination with lamotrigine and then discontinued, the concentration of lamotrigine will increase when the inducers are removed and decrease when the inhibitor is removed. The addition of lamotrigine to patients taking valproic acid should begin with a very low dose, and the dose should be titrated very gradually. If valproic acid is discontinued in a patient taking lamotrigine, the concentration of lamotrigine will decrease and an increase in dose may be necessary. Clinicians should be prepared to reduce, but not stop, the dose of carbamazepine when lamotrigine is added to a patient's regimen because of a pharmacodynamic effect. A synergistic effect may exist with the combined use of lamotrigine and valproic acid. The drug interactions of lamotrigine can be anticipated and patients can be dosed and monitored carefully to maximize the use of lamotrigine. It is wise to monitor carefully any patient who is taking any other medication at the time lamotrigine is added, as well as patients who are taking lamotrigine in whom other drugs are added or discontinued.