Transplacental Passage of Oxcarbazepine and Its Metabolites In Vivo


Address correspondence and reprint requests to Dr. P. Myllynen at Department of Pharmacology and Toxicology, University of Oulu, P.O. Box 5000, FIN-90014 Oulu, Finland. E-mail:


Summary:  Purpose: The purpose of this study was to investigate human fetal exposure to oxcarbazepine (OCBZ) in vivo.

Methods: Transplacental passage and placental tissue concentrations of OCBZ and its metabolites were determined. Maternal venous blood, cord blood, and placental tissue samples from 12 mothers using OCBZ during pregnancy alone or in combination with other antiepileptic drugs were collected. Samples were analyzed with high-performance liquid chromatography.

Results: Maternal venous concentrations of OCBZ and its major metabolites were at same range as cord blood concentrations (OCBZ in maternal serum, 0.19 ± 0.16 μg/ml, and in cord serum, 0.21 ± 0.19 μg/ml; 10-hydroxy-10,11-dihydrocarbamazepine (10-OH-CBZ) in maternal serum, 5.69 ± 2.49 μg/ml, and in cord serum, 5.23 ± 1.44 μg/ml; 10,11-trans-dihydroxy-10,11-dihydrocarbamazepine (10,11-D) in maternal serum, 0.29 ± 0.22 μg/ml, and in cord serum, 0.28 ± 0.14 μg/ml). OCBZ (0.17 ± 0.16 μg/g placental tissue), 10-OH-CBZ (3.49 ± 1.34 μg/g placental tissue) and 10,11-D (0.25 ± 0.11 μg/g placental tissue) were detected in the placental tissue. The amount of OCBZ detected from placental tissue was 0.01% of the daily dose.

Conclusions: OCBZ, like other antiepileptic drugss, is transferred significantly through the placenta in humans.

Epilepsy is one of the many chronic diseases that must be treated during pregnancy. The potential risks caused by antiepileptic drugs (AEDs) for woman and fetus must be considered before pregnancy. For the development of safe and effective drug treatments, it would be essential to know more of the placental pharmacokinetics of the drugs. Animal testing gives valuable information, but it also is known that the structure and functions of the placenta are quite species-specific in mammalians (1), and because of these interspecies differences, extrapolation of data to the human in vivo situation is difficult.

Oxcarbazepine (10,11-dihydro-10-oxo-5H-dibenz(b,f)-azepine-5-carboxamide; OCBZ) is a relatively new AED. OCBZ is rapidly and almost completely metabolized to an active metabolite, 10-hydroxy-10,11-dihydro-carbamazepine (10-OH-CBZ), by a hepatic cytosolic arylketone reductase in human (2). This metabolite is largely conjugated to glucuronic acid and then excreted in urine. A very small amount of 10-OH-CBZ is further metabolized to 10,11-dihydroxy-10,11-dihydro-carbamazepine (10,11-D; 2).

Although OCBZ is used during pregnancy, very little is known about the developmental effects or transplacental passage of OCBZ. Animal studies have shown an increase in fetal structural abnormalities during OCBZ treatment (2). Of the 12 women receiving OCBZ during the first trimester of pregnancy that Friis et al. (3) described, nine had a term baby without structural abnormalities, and three had a spontaneous abortion. Transplacental passage of OCBZ has been studied in a few pairs of serum samples earlier (4,5). Bülay et al. (4) reported concentrations from one mother and neonate in their work. We reported concentrations from three mothers and cord blood (5). These results implicate that OCBZ crosses the placenta in significant amounts (4,5). Placenta also is capable of metabolizing OCBZ to some extent to its major metabolite, 10-OH-CBZ (5,6). We report here maternal and cord blood concentrations along with placental drug concentrations of OCBZ and its major metabolites in 12 mothers using OCBZ during pregnancy.


Clinical samples

Serum samples from maternal venous and cord blood (mixed arteriovenous blood) were obtained during delivery at the Department of Gynecology and Obstetrics, University Hospital of Oulu, between January 1996 and May 2000. Corresponding maternal and venous blood samples were taken at the same time during delivery. Either OCBZ alone or in combination with other AEDs was used for the treatment throughout the pregnancy (Table 1). This includes the majority of the pregnant mothers at University Hospital of Oulu receiving OCBZ therapy during that time. The Ethical Committee of the Medical Faculty of University of Oulu approved the study protocol. Informed consent was obtained from mothers.

Table 1.  Antiepileptic medications, weeks at delivery, and samples taken
Dosage of AEDsWeeks at delivery
(wks + days)
Serum samplesb
tissue samples
  • a

     Data from cases number 6 to 8 was published earlier in Pienimäki et al. (1997).

  • b

     From mother and cord mother as 11.

  • c

     Same mother as 11.

  • d

     Same mother as 2.

  • e

     Same mother as 3.

 1Oxcarbazepine, 600 mg inline image+ 2TermSerum
 2Oxcarbazepine, 300 mg 1inline image+ 239 + 6Serum
 3Oxcarbazepine, 300 mg 2 + 3
Clonazepam, 2 mg 1 + 1
Vigabatrin, 500 mg 2 + 3
41 + 3Serum
 4cOxcarbazepine, 300 mg 1inline image+ 1inline image39 + 3Serum
 5dOxcarbazepine, 600 mg 1 + 1TermSerum
 6aOxcarbazepine, 600 mg 1 × 340 + 2Serum
 7aOxcarbazepine, 600 mg 1inline image× 239 + 2Serum
 8aOxcarbazepine, 300 mg 1 × 235 + 4Serum
 9Oxcarbazepine, 300 mg 2 × 240 + 3Serum, tissue
10Oxcarbazepine, 600 mg inline image+inline image40 + 1Serum, tissue
11Oxcarbazepine, 300 mg 1inline image+ 1inline image41 + 3Serum, tissue
12Oxcarbazepine, 300 mg 1 + 1inline image39 + 0Serum, tissue
13Oxcarbazepine, 300 mg 1inline image+ 240 + 5Tissue
14eLamotrigine, 100 mg 1 × 2
Clonazepam, 2 mg 1 × 1
Oxcarbazepine, 300 mg × 3
40 + 6Tissue
15Oxcarbazepine, 600 mg 2 + 1inline image39 + 5Tissue

Samples were centrifuged and serum stored at –20°C until analyzed. Pieces of placental tissue were collected. Placental tissue samples were placed in liquid nitrogen immediately after birth. The tissue samples were stored at –70°C until analyzed. Concentrations of OCBZ and its metabolites were determined with high-performance liquid chromatography (HPLC).


The 10,11-trans-dihydroxy-10,11-dihydro-carbamazepine (10,11-D), OCBZ, and 10-hydroxy-10,11-dihydro-carbamazepine (10-OH-CBZ) were gifts from Novartis (Basel, Switzerland). HPLC-grade acetonitrile and methanol were purchased from Merck (Darmstadt, Germany).

Analysis of oxcarbazepine

The analysis was done as described by Pienimäki et al. (7). The HPLC system consisted of a Merck-Hitachi L-6200 gradient pump and L-4250 UV-VIS Detector set at 212 nm. The data were processed with a personal computer using Merck Hitachi D-6000 HPLC software. Chromatographic separation was achieved at room temperature by using a mobile phase consisting of acetonitrile (20%) and buffer (80%, 20 mM KH2PO4 with 0.05% triethylamine) at a flow rate of 1.0 ml/min in conjunction with a LiChrospher 100 RP-18 (4 × 4 mm i.d., 5 μm) precolumn and Superspher 60 RP-select B (125 × 4 mm i.d., 4 μm) column from Merck.

The 500-μl serum samples were extracted twice with 5 ml methyl-tert-butylether. The evaporated samples were dissolved in 80 μl methanol:H2O (5:2, vol/vol). A 20-μl aliquot was injected into HPLC. Placental tissue samples were homogenized in 0.1 M phosphate buffer, pH 7.4 (1 g tissue, 4 ml buffer) before extraction, and 3 ml of homogenate was extracted twice with 5 ml of methyl-tert-butylether. All the samples were analyzed in duplicate. The total amount of OCBZ in the placental tissue was calculated by using the concentration detected in the placental tissue and the weight of the placenta.

The previously reported limit of detection of this HPLC method is 12 ± 5 ng/ml for OCBZ, 8 ± 3 ng/ml for 10-OH-CBZ, and 8 ± 3 ng/ml for 10,11-D. The limit of quantification is 55 ± 21 ng/ml for OCBZ, 24 ± 15 ng/ml for 10-OH-CBZ, and 24 ± 15 ng/ml for 10,11-D (7). The day-to-day variation in the quantification was 8% with eight analyses of a clinical sample over the period of 24 days (7).

Statistical tests

Statistical analysis [one-way analysis of variance (ANOVA)] was performed by using Microcal Origin version 4.1 (Microcal Software Inc., Northampton, MA, U.S.A.). The p values <0.05 were taken as significant.


Clinical data with the AEDs used by the mothers during pregnancy are shown in Table 1. Daily doses of OCBZ varied between 750 and 2,100 mg, the average being 1,137.5 mg. The average age of the mothers was 26.7 years (range, 19–38 years). No major malformations of newborns were detected at birth.

Umbilical cord concentrations of OCBZ and its major metabolites were at the same range as maternal venous concentrations (Fig. 1, Table 2). The OCBZ levels detected in this study were low, but the method was sensitive enough for reliably reproducible results. The OCBZ concentrations were clearly above detection limit except in three samples in which OCBZ was not detected at all (Table 2). In one sample (pregnancy number 5), the OCBZ levels were detectable but below the quantification limit.

Figure 1.

The average oxcarbazepine (OCBZ), 10-OH-CBZ, and 10,11-D concentrations in maternal blood (μg/ml, open columns), cord blood (μg/ml, light gray columns), and placental tissue (μg/g tissue, solid columns). 10-OH-CBZ was significantly higher in maternal and cord blood than in the placental tissue (*p < 0.05). Standard deviations are set as error bars.

Table 2.  Maternal venous, cord blood and placental tissue concentrations of OCBZ and its metabolites
Cord blood
Cord blood
Placental tissue
Cord blood
Placental tissue
  • Differences between maternal and cord blood concentrations were not statistically significant for either metabolite. 10-OH-CBZ concentrations were statistically significantly lower in the placental tissue than in maternal venous or cord blood.

  • a

     Data from cases number 6 to 8 were published earlier in Pienimäki et al. (1997).

  • b

     OCBZ was detected below quantitation limit.

  • ND, not done; OCBZ, oxcarbazepine.

Average ± SD0.19 ± 0.160.21 ± 0.190.17 ± 0.165.69 ± 2.495.23 ± 1.443.49 ± 1.340.29 ± 0.220.28 ± 0.14

The differences of the mean concentrations of OCBZ, 10-OH-CBZ, and 10,11-D in the maternal serum and the cord serum were not statistically significant (p > 0.05; Table 2). The 10-OH-CBZ serum concentrations were clearly higher than concentrations of the parent drug or the other metabolite, 10,11-D. In some cases, OCBZ concentrations in cord blood were two- to fourfold higher than the maternal venous concentrations. No such differences between maternal and cord blood concentrations of 10,OH-CBZ and 10,11-D were found.

OCBZ (0.17 ± 0.16 μg/g) and 10,11-D (0.25 ± 0.11 μg/g) were detected in placental tissue (Fig. 1, Table 2). The amount of OCBZ detected from placental tissue was 0.01% of the daily dose. The amount of 10-OH-CBZ (3.49 ± 1.34 μg/g) was> 10 times higher than the amount of OCBZ in the placental tissue. The tissue concentrations of 10,11-D and 10-OH-CBZ were close to the maternal and fetal serum concentrations (Fig. 1). However, the tissue concentration of 10-OH-CBZ was statistically significantly lower than the serum concentrations (p < 0.05; Fig. 1), assuming that 1 g of tissue is equivalent to 1 ml of serum.


Placental transfer of OCBZ and its major metabolites appears to be considerable, as suggested by earlier studies both in vivo and in vitro (4,5). The cord-blood concentrations of the major metabolite of oxcarbazepine, 10-OH-CBZ, were similar to maternal concentrations. The majority of OCBZ metabolism takes place in the liver (2). We detected some placental metabolism in placental perfusion and in vitro incubations in our earlier studies (5,6). However, most of the 10-OH-CBZ detected in this in vivo study probably originates from maternal liver metabolism because of the low level of placental metabolism (5,6) and immaturity of fetal liver enzymes (8,9).

Unlike 10-OH-CBZ and 10,11-D, the concentration of OCBZ tended to be a little higher in the cord blood than in maternal plasma in some of our cases, although these differences were not statistically significant, and concentrations were still very low. Generally, maternal drug concentrations are higher than fetal or similar to fetal in the in vivo human studies. However, it is known that cord-blood concentrations of some compounds (p-hydroxy-phenobarbitone, valproic acid, diazepam, naphthoxylactic) tend to be higher than maternal concentrations (10 and references within). The mechanism for this is unclear. It is possible that higher OCBZ concentrations in cord blood in some of the cases are due to slightly slower clearance of this compound from the fetal side than from the maternal side. Fetal liver has been shown to metabolize many foreign compounds (11), but it also is known that half-life of most drugs is prolonged in human neonate (9). Fetal metabolism is generally slower than that in the adult because metabolizing (e.g., cytochrome P-450) enzyme activities are much lower in the fetal liver (8).

The concentrations of OCBZ were similar in placental tissue and both maternal and cord serum, and it seems that OCBZ does not accumulate in the placental tissue in significant amounts in vivo. Conversely, the tissue concentrations of 10-OH-CBZ were lower than its serum concentrations. It is known that lipophilic compounds tend to accumulate in the placental tissue (12). The partition coefficient between n-octanol/aqueous buffer for 10-OH-CBZ is 8.8, and it is less lipophilic than OCBZ (partition coefficient, 20.4). In contrast to this in vivo study, our earlier experiments done with the placental-perfusion method suggested significant accumulation of 10-OH-CBZ in the placental tissue. Perfused tissue is always somewhat edematous (13). It is possible that the tissue edema during the ex vivo perfusions has favored accumulation of 10-OH-CBZ in the tissue, which may explain the difference between in vivo and ex vivo results.

In conclusion, it seems that, like many other AEDs (10,14), OCBZ also is transferred significantly through the placenta. In contrast to our earlier perfusion study, there are no signs of significant accumulation of OCBZ or its metabolites in the fetal circulation or placental tissue.

Acknowledgment: We thank the nursing personnel in the Delivery Room and Maternity Outpatient Clinic of the Department of Obstetrics and Gynecology for their cooperation. We thank Ms. Kirsi Salo and Mr. Kauno Nikkilä for practical help and Novartis for providing oxcarbazepine and its metabolites. We also are grateful to Prof. Olavi Pelkonen for critical review of the manuscript.