Gestation-induced changes in lamotrigine pharmacokinetics: A monotherapy study

Methods.

All women attending two epilepsy clinics in The Netherlands who became pregnant between 1997 and 2003 and were using LTG monotherapy were included in the study. At least two pregnancy LTG serum levels had to be available that were sampled <1 hour apart. Eleven pregnancies (A1-H) were studied retrospectively, and the 12th (Q) was studied prospectively. Data on seizure types, seizure frequencies, and side effects were extracted from seizure diaries and medical files and compared with the 9 months before pregnancy. At each visit, blood was sampled for LTG serum levels. To compensate for adjustments in LTG dosing, a level-to-dose ratio was calculated using the formula: LTG level (μg/mL) × 100/prescribed LTG dose (mg).

For frequent LTG level monitoring in Patient Q, a novel capillary blood sampling method was used.6 The patient is instructed to sample blood from a fingertip by means of an automatic lancet and apply it on standardized filter paper. Ten antiepileptic drugs, including LTG, can be analyzed with high-performance liquid chromatography/gas chromatographic-mass spectrometric assay in an extract from a punch-out of the blood spot. Patient Q sampled morning through levels, biweekly during pregnancy, and daily in the weeks after delivery.

Results.

Twelve pregnancies from nine women were studied. Patient A had three and Patient C two pregnancies included. Mean age was 30.7 years (range, 26 to 35 years). Two patients had juvenile absence epilepsy, and eight had symptomatic or cryptogenic localization-related epilepsy. One woman (Patient A) had a migration disorder (unilateral nodular periventricular heterotopia) as probable etiology of the epilepsy.

In the 9 months before pregnancy, four women had been completely seizure free, and eight women had had minor seizures (complex partial, absences). None of the women had tonic-clonic seizures in the period before pregnancy.

Seizure aggravation was observed in 9 of 12 pregnancies, being an increase in seizure frequency with >100% in four pregnancies, and seizures recurred in three patients who had been seizure free before pregnancy; tonic-clonic seizures recurred in one patient with partial seizures, and one patient with juvenile absence epilepsy had her first tonic-clonic seizure ever. Seizures aggravated between weeks 12 and 28 in eight pregnancies and in week 40 in pregnancy A2. During seven pregnancies, LTG dose was increased to regain seizure control.

Pregnancy A1 ended in fetal death during the 5th month, with postmortem examination revealing an infectious cause. The other pregnancies and deliveries were uneventful.

After delivery, three of the seven women whose LTG dose had been adjusted during pregnancy had dizziness, diplopia, or ataxia. These side effects started between 3 and 10 days after delivery. LTG serum levels were high normal (12 to 14 mg/L) and resolved after decreasing the LTG dose.

For consecutive 10-week pregnancy periods, the mean LTG level-to-dose ratios (presented as percentage of baseline ± SD) were 82 ± 14%, 51 ± 14%, 40 ± 8%, and 48 ± 10%. Postpartum level-to-dose ratio was 97 ± 15% of baseline (figure 1). The prospectively studied Patient Q remained seizure free throughout pregnancy, and on her request LTG dosing was not changed despite a major decrease in LTG levels (figure 2).

• Download figure
• Open in new tab
• Download powerpoint

Figure 1. Lamotrigine level-to dose ratios expressed in percentage of baseline in 11 retrospective pregnancies. Level to-dose ratios decreased gradually during pregnancy to a minimum of 47% of baseline. After delivery, ratios increased again to 97% of baseline.

• Download figure
• Open in new tab
• Download powerpoint

Figure 2. Frequent blood sampling by the 12th patient (Patient Q) demonstrated a gradual decrease of lamotrigine (LTG) levels from the baseline of 5.5 mg/L to a minimum of 1.2 mg/L in the last month of gestation. After delivery, frequent sampling showed a return to baseline within 3 weeks. LTG dosing was adjusted during pregnancy despite a major reduction in LTG plasma levels.

After delivery, LTG levels in four cord plasma samples were 89.8% of maternal plasma. The mean milk/maternal plasma ratio was 0.54 in three samples, in accordance with a reported ratio of 0.61.7 One baby of a lactating mother had an LTG serum level of 1.7 mg/L 10 days after delivery, being 30.4% of mother’s plasma level at that moment, confirming literature data.8

Discussion.

It is known that pregnancy influences the pharmacokinetics of almost all anticonvulsant agents. A previous study reported a 65% increase in LTG clearance, but all but two patients in that study were using concomitant medication.5 The interaction between concomitant drugs and LTG kinetics makes polytherapy studies difficult to interpret.

The current study describes a case series of women attending the clinics of a tertial referral epilepsy center, a rather severe selection of epilepsy patients. Even with this limitation, a seizure aggravation in 75% of women with epilepsy treated with LTG monotherapy is exceptionally high.1,2⇓ Probably this seizure aggravation is associated with a gradual decrease of LTG level-to-dose ratios to a minimum of 40% of baseline.

After delivery, the induced LTG metabolism reverts quickly to baseline.5,7⇓ In the current study, the LTG level-to-dose ratio returned to 97% of baseline. Toxic side effects in three patients became apparent 3 to 10 days after delivery. Frequent LTG level monitoring by Patient Q during the weeks after delivery suggests a return of LTG kinetics to baseline within 3 weeks after delivery.

The frequent through level sampling by Patient Q also indicates pulsatile variations in LTG levels during pregnancy and after delivery. This effect has not been observed before and may be related to the maternal physical condition with fluctuations of hormonal and other endogenous homeostatic mechanisms. Variability in blood sampling times reported by the patient was <1 hour. Duplicate samples differed <10% in LTG concentrations; therefore, within-assay variability is an unlikely cause for the fluctuation but cannot be ruled out completely. Such a pulsatile course of LTG levels has to be examined in other patients and in other physical conditions.

The variability in the pharmacokinetic profile of LTG in relation to oral contraception and pregnancy complicates the treatment of women in this period of life. It has to be taken into account that LTG clearance decreases when oral contraceptives are withdrawn before pregnancy, resulting in potentially high LTG serum levels around conception.4 High anticonvulsant levels in this period should be avoided, although preliminary data suggest that LTG has limited teratogenic potential.9 This study confirms that during pregnancy LTG clearance may increase to the point that seizures may aggravate, and adjustment of LTG dose may be necessary. After delivery, the dose should be decreased again soon to avoid toxic side effects. During the lactating period, it is important to know that LTG is excreted in considerable amounts in breast milk.10 Combined with slow elimination in infants, this may result in therapeutic LTG serum levels in the newborn.8

Frequent LTG level monitoring during the period of withdrawal of oral contraceptives, during pregnancy, and after delivery is necessary to optimize treatment in women taking LTG monotherapy in this period of unstable kinetics. In lactating mothers, extra attention is needed to avoid toxic side effects in the newborn.