- Email: [email protected]
He mice
American Journal of Obstetrics and Gynecology (2004) 190, 259e63
www.elsevier.com/locate/ajog
Chronic prenatal exposure to carbamazepine and perinatal outcomes of C3H/He mice H. Dix Christensen, PhD,a,b William F. Rayburn, MD,a,b,d,* K. Michael Parker, PhD,c Christina L. Gonzalez, BS,a,b Karen P. Gold, MDa,b Departments of Obstetrics and Gynecology,a Pharmacology/Toxicology,b and Pathology,c University of Oklahoma Health Sciences Center, Oklahoma City, Okla, and the Department of Obstetrics and Gynecology, University of New Mexico Health Sciences Center, Albuquerque, NMd Received December 20, 2002; revised June 27, 2003; accepted July 12, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Carbamazepine Reproduction Perinatal outcome Mice
Objectives: This study was undertaken to determine a daily therapeutic dose of carbamazepine and to measure its effect on reproductive performance and perinatal outcomes of mice. Study design: Adult C3H/He mice were given carbamazepine in rodent chow in either a 0.25% or a 1.0% mixture. Comparisons between doses included nongravid weight change, plasma drug steady-state concentrations, and response to a maximal electroshock seizure test. The strain was then fed either the preferred dose of carbamazepine or a placebo 1 week before starting to mate and throughout gestation to compare reproductive performance and offspring early development. Results: Mice who ate the 0.25% carbamazepine mixture displayed no evoked seizure activity and, in contrast to the 1.0% mixture, did not lose weight. This daily dose of 542G35 mg/kg produced a trough steady-state plasma concentration that was consistent with a protective threshold in humans. Differences from placebo controls were not statistically significant for the number of cycles necessary to conceive or for the duration of gestation. The litter size, survival rates, birth weights, weight gain, and onset of eye openings and teeth eruptions of the pups were not statistically significant between the two groups. Conclusion: Long-term prenatal exposure to a subtoxic yet therapeutic dose of carbamazepine did not impair reproductive performance or early growth and development of exposed mice offspring. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Carbamazepine is an anticonvulsant ingested daily by reproductive-age women for the treatment of psychomotor or grand mal seizures. The daily dose needs to be sufficient to maintain therapeutic plasma levels. Although Supported by the John W. Records Perinatal Research Fund. * Reprint requests: William F. Rayburn, MD, Department of Obstetrics and Gynecology, 2211 Lomas Blvd NE ACC 4, Albuquerque, NM 87131. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0002-9378(03)00899-8
fertility rates may be lower among patients with seizure disorders, most are able to conceive without difficulty.1 Similar to other anticonvulsant medications, carbamazepine should be continued during pregnancy to prevent either seizure-induced hypoxia or status epilepticus.1 Carbamazepine is transported rapidly across the placenta, achieving equilibrium in the developing fetal brain.2 Seizure disorders are associated with a 2- to 3-fold increase in congenital malformations and in fetal weight reduction.3,4 Each of the commonly used anticonvulsant
260 medications (phenytoin, phenobarbital, carbamazepine, valproate) has been implicated as being teratogenic, but there is no clear consensus concerning which drug is least harmful. Data provided by Matalon et al5 from prospective studies involving 1255 cases of carbamazepine exposure revealed an increased rate of mainly neural tube defects, cardiovascular and urinary tract anomalies, and cleft palate. A 1% risk of spina bifida has been demonstrated in infants of mothers taking carbamazepine.6,7 Furthermore, a spectrum of anomalies similar to the fetal hydantoin syndrome has been identified consisting of growth delays, cranial facial abnormalities (including clefting), and limb anomalies (including distal phalanges).8 Government regulatory agencies and multiple researchers have found a qualitative correlation between human and rodent developmental effects when the prenatal dose of a centrally acting drug is based on equivalent plasma concentrations.9 Adverse effects in gravid rodents have involved toxic oral dosages, usually at least 1.5 to 4 times the maximum human dose (based on body surface area, milligrams per meters squared).10 Objectives of this study in mice were 2-fold: (1) to determine a therapeutic yet subtoxic dose of oral carbamazepine given daily and (2) to evaluate effects from chronic prenatal exposure of carbamazepine on reproductive performances and on perinatal outcomes of exposed offspring. A placebo group served as a control in this randomized trial.
Material and methods This proposal was approved by our Animal Review Committee. The C3H/He mouse (Jackson Laboratory, Bar Harbor, Me) was chosen to increase the potential of observing subtle impairments. This inbred strain is more of a ‘‘high-risk’’ pregnant animal than an outbred strain, although C3H/He mice produce an adequate litter size of five to eight pups.11 Nongravid adult mice, received at 7 weeks of age, were allowed 1 week of adjustment in their new setting. The mice were housed at an American Association for the Accreditation of Laboratory Animal Caree approved (AALAC) facility. The isolated animal room had a controlled temperature (22(C), humidity (50%G15%), and 12-hour light/12-hour dark schedule (lights on at 6 AM). Each mouse was maintained on an adequate diet and allowed free access to food and water.
Dosing determination The preferred daily oral dose was determined by finding the least dose to be effective while not causing adverse effects. Carbamazepine (Tegretol, Novartis, Summit, NJ, 100-mg tablet) was mixed with 11% fat, powdered mouse chow (5015 PMI Feeds Inc, St Louis, Mo).10 Folic acid and vitamin K were not supplemented to the chow. The mixed chow was reconstituted with water
Christensen et al to form cylindric food bars with use of an open 10-mL syringe barrel. In a preliminary project, C3H/He mice were fed food bars for 2 weeks that contained 0.125%, 0.25%, 0.5%, 1.0%, or 2.0% of carbamazepine (eg, 12.5 mg, 25 mg, 50 mg, 100 mg, or 200 mg of carbamazepine in 10 g of chow.) Seizures were then evoked with electrodes placed on both corneas. A 60-Hz alternating current was delivered for 0.2 seconds with an apparatus similar to that designed by Woodbury and Davenport.12 A seizure in the restrained mouse was elicited with a current that was five to seven times that necessary to evoke minimal electroshock threshold seizures (ie, 50 mA in mice). A lack of seizure activity (no hind limb extension) was found using the 0.25% concentration or higher. The 2.0% concentration was so high that it caused excessive weight loss and concentrated the urine to signify dehydration. For this reason, the 0.25% and the 1.0% concentrations of carbamazepine were chosen for comparison of plasma concentrations and weight change. Another method of assessing efficacy included measuring plasma concentrations of the drug and epoxide metabolite. After 5 weeks on the mixed chow, 10 nongravid adult mice in either carbamazepine treatment group underwent cervical dislocation at midafternoon when food consumption, and presumably drug levels, was the lowest. Up to 1 mL of whole blood was gathered by cardiac puncture and placed in a heparinized syringe. Drug and metabolite concentrations were measured by using high-pressure liquid chromatography. This trough concentration was compared with the plasma value reported in mice to provide a seizure protection threshold.13 A means for evaluating adverse drug effect was weight loss. Food consumption and body weight were recorded in 10 nongravid mice 1 week before mating and after 2 weeks while pregnant. Food consumption was estimated by determining differences between weights of the food added and removed.
Perinatal outcomes After a clinically relevant dose of carbamazepine was selected, a treatment group and a placebo control group were compared for perinatal outcomes. A minimum of 14 nongravid female mice were exposed to food bars containing either the chosen dose of carbamazepine or a placebo (no drug). Randomization to treatment was computer generated. Being housed individually, each mouse was allowed 1 week for adjustment to the specified food bars before breeding. Each female was mated with an experienced male. The presence of a copulatory plug (gestational day 0) terminated mating during a potential 8-day period (2 cycles). After conception, the females were given nestlets throughout weaning.
261
Christensen et al Table I Steady-state plasma concentrations of carbamazepine and its epoxide metabolite in nongravid adult C3H/He mice Carbamazepine dose Carbamazepine (mg/L) Epoxide (mg/L) Total plasma concentration (mg/L) Carbamazepine/epoxide ratio
0.25%
1.0%
1.5G0.2 3.5G0.1 5.0G0.2 0.4G0.1
3.0G0.3 9.0G1.4 12.0G1.7 0.3G0.1
MeanGSEM. 0.25% = 25 mg in 10 g of chow. 1.0% = 100 mg in 10 g of chow. Total plasma concentration is the sum of carbamazepine and epoxide levels.
Beginning on gestational day 18, the dams were checked each morning for the presence of litters. After birth, dams in both treatment groups were given placebo-containing food bars. Maternal behaviors were observed by the same field biologist (C. L. G.). The number of pups was counted for each litter at birth and on postnatal day 4. The presence of any external anomalies was sought. At least eight litters were anticipated in each treatment group with a litter size of at least four viable offspring. This inbred strain of pups was not weighed at birth because of their fragility. Instead, they were weighed on postnatal days 4, 7, 11, 17, and 21 (weaning). The day on which both eyes were opened and the percentage of animals with upper and lower incisor eruptions on postnatal day 10 were recorded. Statistical comparisons were conducted with a Student t test or the Fisher exact method. The data were reported either as a mean percentage or as an SE. Because a litter is the unit of measurement, individual pup data were combined and averaged to determine a litter means and SE. A P!.05 was considered to be statistically significant.
Daily consumption of the food bars containing either 0.25% carbamazepine or placebo was examined in nongravid and gravid mice. Food consumption was not different between the two treatment groups, regardless whether the mouse was gravid (Table II). The body weight increased during pregnancy in both groups. The percentage of food consumed in proportion to body weight in the carbamazepine group was not proportionally different from the placebo group. The daily dose of carbamazepine was 542G35 mg/kg at 1 week premating and was 407G13 mg/kg after 2 weeks while gravid. Most mice conceived in the first estrus cycle and all within two cycles, regardless of therapy. Reproductive performances of the carbamazepine-treated mice were not different from the placebo group (Table III). The percentages of deliveries per mating and per copulatory plug were similar between the groups. The duration of gestation was usually 19 or 20 days in each group. The number of live pups per litter (usually 6) was not different between the carbamazepine and placebo groups. The percentage of male pups (51.9%) was indistinguishable between groups. There were no stillborn pups in the carbamazepine group. No open neural tube defects, midline facial clefting, or extra digits were observed. Two runts (birth weight 2 SD below mean) were found in each of the carbamazepine and placebo groups. Transport/retrieve behaviors of the dams were indistinguishable between the two groups. Pup weights at postnatal days 4, 11, and 21 were not statistically different between carbamazepine- and placebo-exposed offspring. By postnatal day 10, nearly all the bottom teeth had erupted, whereas the top teeth were erupting with equal frequency between the groups. Both eyes opened gradually by postnatal day 15 in the carbamazepine and placebo groups.
Comment Results After 2 weeks of food-drug consumption in nongravid mice, the weight increased (1.7%G1.1%) in the 0.25% carbamazepine group (n = 10) and in the placebo group (2.4%G1.7%) but decreased (ÿ24.4%G1.4%) in the 1.0% carbamazepine group (n = 10). Steady-state plasma concentrations of carbamazepine and its epoxide metabolite are shown in Table I after 5 weeks. Compared with the 0.25% carbamazepine group, those fed the 1.0% carbamazepine food bars had a 2-fold higher concentration of the drug and a nearly 3-fold higher concentration of its metabolite. Carbamazepine/epoxide ratios were not statistically different, however, between the two dose groups. The trough steady-state plasma concentrations (5.0G0.2 mg/L in 0.25% mixture; 12.0G1.7 mg/L in 1.0% mixture) were consistent with the usual therapeutic range in humans.
About 0.52% of pregnant women have a preexisting seizure disorder, and approximately 19,000 infants are born each year in the United States to these women.3 The current investigation was undertaken to find a therapeutic dose of carbamazepine in mice to evaluate its effect on reproductive performances and on perinatal outcomes. Carbamazepine is a first-line drug in the treatment of most forms of epilepsy and also for trigeminal neuralgia. Limitations of several studies about carbamazepine during human pregnancy include a lack of reporting successful cases, no control population, and small series of patients. Safety during pregnancy about carbamazepine is less well known than with phenytoin and phenobarbital. In the current investigation that used a common inbred mouse strain, we were able to establish a therapeutic dose of carbamazepine. The 0.25% concentration in
262
Christensen et al
Table II Food consumption and estimated daily dose of mice treated either with 0.25% carbamazepine or with a placebo Daily food consumption Body weight Nongravid 1 wk premating Placebo Carbamazepine Gravid during second week Placebo Carbamazepine
g
Body weight (%)
21.6G0.9 5.3G0.4 25.3G3.0 21.6G0.9 5.4G0.2 25.1G1.4 31.6G1.2 5.1G0.2 16.2G0.4 30.9G0.8 5.0G0.1 16.3G0.5
MeanGSEM. No significant dierences in food consumption were found between the mice fed 0.25% carbamazepine or a placebo.
the food bars produced a complete response to electroshock stimulation while not reducing food consumption or weight. Our data also revealed trough steady-state plasma concentrations of carbamazepine and its primary epoxide to be within the reported seizure protection threshold in mice and in adult humans (4-12 mg/L).14 This finding is especially important because detrimental effects from in utero exposure to carbamazepine may result from higher levels of epoxide intermediaries.8 The current study reports that a daily dose (542G35 mg/kg) of this centrally acting drug did not impair ovulation or delay conception. Conception rates with this mouse strain are comparable to that reported at a standard national laboratory.15 Screening for newborn development revealed no additional risks with carbamazepine exposure in utero. Reproductive toxicity (stillbirth, runting, preterm delivery) was not apparent with this ‘‘highrisk’’ inbred mouse strain exposed to a daily dose of carbamazepine before and throughout gestation. Results from this placebo-control trial offset worrisome information from pharmaceutic labeling using megadoses in rodents. The manufacturer reports that carbamazepine ‘‘potentially causes reproductive and fetal harm.’’10 We observed that carbamazepine monotherapy did not impair maternal food consumption and weight gain during pregnancy. This clinically relevant dose of carbamazepine was not toxic to pregnancies. A few reports in humans suggest that carbamazepine monotherapy is associated with a lower birth weight in term infants compared with controls.4,16 There was no difference between groups in gestational age at delivery or in the rate of spontaneous abortion. The reduced birth weight (136 g) with prenatal exposure to carbamazepine may be clinically insignificant. In our investigation, the fetal loss rate, gestational age at delivery, birth weight, and proportion of growth restricted mice were not different between the carbamazepine and placebo groups. Our investigation was intended to identify a clinically relevant dose for identifying subtle neurobehavioral changes and possible development shifts. This daily dose
Table III Reproductive performances and physical development of offspring exposed prenatally either to 0.25% carbamazepine or to a placebo Pregnant mice Delivery/mating (%) Delivery/plugs (%) Delivery/first cycle (%) Gestational length (d) Runts Live pups/litter Birth PND 4 Male pups (%) Pup weights (g) PND 4 PND 11 PND 21 Teeth eruption (PND 10) % top % bottom Both eyes open (%) PND 12 PND 13 PND 14
Carbamazepine
Placebo
8 57.1 72.7 75 19.8G0.2 2
8 53.3 72.7 75 19.8G0.3 2
6.2G0.3 6.2G0.3 51.8G8.7
5.7G0.4 5.4G0.5 63.6G9.6
2.6G0.1 7.2G0.2 10.6G0.2
2.5G0.1 7.2G0.2 10.7G0.2
62.5G15.7 97.5G2.5
61.5G15.2 100
18.7G8.0 73.7G8.4 91.6G3.2
12.7G9.6 69.0G7.4 93.7G4.8
MeanGSEM. No significant differences were found between groups exposed to carbamazepine or placebo. PND, Postnatal day.
would not be expected to cause malformations.17 Furthermore, the cause of malformations is multifactorial, not just dose related. We were encouraged that external anomalies, including open neural tube defects and facial clefting, were not observed in the carbamazepineexposed offspring. No differences between the carbamazepine and placebo group were found in newborn development that included food consumption, weight gain, interaction with the mother, and onset of teeth eruptions/eye openings. Research described here will aid in developing more physiologically based dose-response models in animals. Anticonvulsant properties of carbamazepine are thought to result from reducing postsynaptic responses and by blocking posttetanic potentiation. With the use of this therapeutic dose throughout gestation, it will be possible to screen for long-term neurodevelopmental effects from carbamazepine. Interpretation of behavior patterns, social interaction, and cognitive tests of exposed offspring will provide more insight into any subtle long-term development delays.
References 1. Cramer JA, Jones EE. Reproductive function in epilepsy. Epilepsia 1991;32:519-26. 2. Bertilsson W, Tomson T. Clinical pharmacokinetics and pharmacological effects of carbamazepine and carbamazepine-10, 11epoxide: an update. Clin Pharmacol Kinet 1986;11:177-98.
Christensen et al 3. American College of Obstetricians and Gyncologists. Seizure disorders in pregnancy. Washington (DC): The Colleges; 1996. Educational bulletin No.: 231. 4. Diav-Citrinoo, Schechtman S, Arnor J, Ornoy T. Is carbamazepine teratogenic? A prospective controlled study of 210 pregnancies. Neurology 2001;57:321-4. 5. Matalon S, Schechtman S, Goldzweig G, Ornoya. The teratogenic effect of carbamazepine: a meta-analysis of 1255 exposures. Reprod Toxicol 2002;16:9-17. 6. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991;324:674-7. 7. Nulman I, Scolnik D, Chitayat D, Farkas LD, Koren G. Findings in children exposed in utero to phenytoin and carbamazepine monotherapy: independent effects of epilepsy and medications. Am J Med Genet 1997;68:18-24. 8. Jones KL, Lacro RV, Johnson KA, Adams J. Pattern of malformations in the children of women treated with carbamazepine during pregnancy. N Engl J Med 1989;320:1661-6. 9. Stanton ME. Workshop on the qualitative comparability of human and animal development neurotoxicity in humans and laboratory animals. Neurotoxicol Teratol 1990;12:261-7. 10. Tegretol. Physicians’ desk reference, 25th ed. Montvale (NJ): Medical Economics; 2002. p. 2204-7.
263 11. Rayburn WF, Christensen HD. The mouse as a model to investigate neurobehavioral effects from prenatal exposure to prescription medications. J Matern Fetal Invest 1997;7:3-6. 12. Woodbury LA, Davenport VD. Design and use of a new electroshock seizure apparatus, and analysis of factors altering seizure threshold and pattern. Arch Int Pharmacodyn Ther 1952; 92:97-104. 13. Bourgeois BFD, Wad N. Individual and combined antiepileptic and neurotoxic activity of carbamazepine and carbamazepine10, 11-epoxide in mice. J Pharmacol Exp Ther 1984;231: 411-5. 14. Battino D, Binelli S, Bossi L. Plasma concentrations of carbamazepine and carbamazepine 10, 11-epoxide during pregnancy and after delivery. Clin Pharmacokinet 1985;10:279-84. 15. Green MC, Witham BA. Handbook on genetically standardized JAX mice. 4th ed. Bar Harbor (ME): Jackson Laboratory; 1991. p. 35-8. 16. Bertollini R, Kallen B, Mastroiacovo P, Ribert E. Anticonvulsant drugs in monotherapy: effects on the fetus. Eur J Epidemiol 1987; 3:164-71. 17. Holmes LB, Harvey EA, Goull BA, Huntington KB, Khoshbin S, Hayes TM, et al. The teratogenicity of anticonvulsant drugs. N Engl J Med 2001;344:1132-8.
www.elsevier.com/locate/ajog
Chronic prenatal exposure to carbamazepine and perinatal outcomes of C3H/He mice H. Dix Christensen, PhD,a,b William F. Rayburn, MD,a,b,d,* K. Michael Parker, PhD,c Christina L. Gonzalez, BS,a,b Karen P. Gold, MDa,b Departments of Obstetrics and Gynecology,a Pharmacology/Toxicology,b and Pathology,c University of Oklahoma Health Sciences Center, Oklahoma City, Okla, and the Department of Obstetrics and Gynecology, University of New Mexico Health Sciences Center, Albuquerque, NMd Received December 20, 2002; revised June 27, 2003; accepted July 12, 2003
–––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– KEY WORDS Carbamazepine Reproduction Perinatal outcome Mice
Objectives: This study was undertaken to determine a daily therapeutic dose of carbamazepine and to measure its effect on reproductive performance and perinatal outcomes of mice. Study design: Adult C3H/He mice were given carbamazepine in rodent chow in either a 0.25% or a 1.0% mixture. Comparisons between doses included nongravid weight change, plasma drug steady-state concentrations, and response to a maximal electroshock seizure test. The strain was then fed either the preferred dose of carbamazepine or a placebo 1 week before starting to mate and throughout gestation to compare reproductive performance and offspring early development. Results: Mice who ate the 0.25% carbamazepine mixture displayed no evoked seizure activity and, in contrast to the 1.0% mixture, did not lose weight. This daily dose of 542G35 mg/kg produced a trough steady-state plasma concentration that was consistent with a protective threshold in humans. Differences from placebo controls were not statistically significant for the number of cycles necessary to conceive or for the duration of gestation. The litter size, survival rates, birth weights, weight gain, and onset of eye openings and teeth eruptions of the pups were not statistically significant between the two groups. Conclusion: Long-term prenatal exposure to a subtoxic yet therapeutic dose of carbamazepine did not impair reproductive performance or early growth and development of exposed mice offspring. Ó 2004 Elsevier Inc. All rights reserved.
––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––––– Carbamazepine is an anticonvulsant ingested daily by reproductive-age women for the treatment of psychomotor or grand mal seizures. The daily dose needs to be sufficient to maintain therapeutic plasma levels. Although Supported by the John W. Records Perinatal Research Fund. * Reprint requests: William F. Rayburn, MD, Department of Obstetrics and Gynecology, 2211 Lomas Blvd NE ACC 4, Albuquerque, NM 87131. E-mail: [email protected] 0002-9378/$ - see front matter Ó 2004 Elsevier Inc. All rights reserved. doi:10.1016/S0002-9378(03)00899-8
fertility rates may be lower among patients with seizure disorders, most are able to conceive without difficulty.1 Similar to other anticonvulsant medications, carbamazepine should be continued during pregnancy to prevent either seizure-induced hypoxia or status epilepticus.1 Carbamazepine is transported rapidly across the placenta, achieving equilibrium in the developing fetal brain.2 Seizure disorders are associated with a 2- to 3-fold increase in congenital malformations and in fetal weight reduction.3,4 Each of the commonly used anticonvulsant
260 medications (phenytoin, phenobarbital, carbamazepine, valproate) has been implicated as being teratogenic, but there is no clear consensus concerning which drug is least harmful. Data provided by Matalon et al5 from prospective studies involving 1255 cases of carbamazepine exposure revealed an increased rate of mainly neural tube defects, cardiovascular and urinary tract anomalies, and cleft palate. A 1% risk of spina bifida has been demonstrated in infants of mothers taking carbamazepine.6,7 Furthermore, a spectrum of anomalies similar to the fetal hydantoin syndrome has been identified consisting of growth delays, cranial facial abnormalities (including clefting), and limb anomalies (including distal phalanges).8 Government regulatory agencies and multiple researchers have found a qualitative correlation between human and rodent developmental effects when the prenatal dose of a centrally acting drug is based on equivalent plasma concentrations.9 Adverse effects in gravid rodents have involved toxic oral dosages, usually at least 1.5 to 4 times the maximum human dose (based on body surface area, milligrams per meters squared).10 Objectives of this study in mice were 2-fold: (1) to determine a therapeutic yet subtoxic dose of oral carbamazepine given daily and (2) to evaluate effects from chronic prenatal exposure of carbamazepine on reproductive performances and on perinatal outcomes of exposed offspring. A placebo group served as a control in this randomized trial.
Material and methods This proposal was approved by our Animal Review Committee. The C3H/He mouse (Jackson Laboratory, Bar Harbor, Me) was chosen to increase the potential of observing subtle impairments. This inbred strain is more of a ‘‘high-risk’’ pregnant animal than an outbred strain, although C3H/He mice produce an adequate litter size of five to eight pups.11 Nongravid adult mice, received at 7 weeks of age, were allowed 1 week of adjustment in their new setting. The mice were housed at an American Association for the Accreditation of Laboratory Animal Caree approved (AALAC) facility. The isolated animal room had a controlled temperature (22(C), humidity (50%G15%), and 12-hour light/12-hour dark schedule (lights on at 6 AM). Each mouse was maintained on an adequate diet and allowed free access to food and water.
Dosing determination The preferred daily oral dose was determined by finding the least dose to be effective while not causing adverse effects. Carbamazepine (Tegretol, Novartis, Summit, NJ, 100-mg tablet) was mixed with 11% fat, powdered mouse chow (5015 PMI Feeds Inc, St Louis, Mo).10 Folic acid and vitamin K were not supplemented to the chow. The mixed chow was reconstituted with water
Christensen et al to form cylindric food bars with use of an open 10-mL syringe barrel. In a preliminary project, C3H/He mice were fed food bars for 2 weeks that contained 0.125%, 0.25%, 0.5%, 1.0%, or 2.0% of carbamazepine (eg, 12.5 mg, 25 mg, 50 mg, 100 mg, or 200 mg of carbamazepine in 10 g of chow.) Seizures were then evoked with electrodes placed on both corneas. A 60-Hz alternating current was delivered for 0.2 seconds with an apparatus similar to that designed by Woodbury and Davenport.12 A seizure in the restrained mouse was elicited with a current that was five to seven times that necessary to evoke minimal electroshock threshold seizures (ie, 50 mA in mice). A lack of seizure activity (no hind limb extension) was found using the 0.25% concentration or higher. The 2.0% concentration was so high that it caused excessive weight loss and concentrated the urine to signify dehydration. For this reason, the 0.25% and the 1.0% concentrations of carbamazepine were chosen for comparison of plasma concentrations and weight change. Another method of assessing efficacy included measuring plasma concentrations of the drug and epoxide metabolite. After 5 weeks on the mixed chow, 10 nongravid adult mice in either carbamazepine treatment group underwent cervical dislocation at midafternoon when food consumption, and presumably drug levels, was the lowest. Up to 1 mL of whole blood was gathered by cardiac puncture and placed in a heparinized syringe. Drug and metabolite concentrations were measured by using high-pressure liquid chromatography. This trough concentration was compared with the plasma value reported in mice to provide a seizure protection threshold.13 A means for evaluating adverse drug effect was weight loss. Food consumption and body weight were recorded in 10 nongravid mice 1 week before mating and after 2 weeks while pregnant. Food consumption was estimated by determining differences between weights of the food added and removed.
Perinatal outcomes After a clinically relevant dose of carbamazepine was selected, a treatment group and a placebo control group were compared for perinatal outcomes. A minimum of 14 nongravid female mice were exposed to food bars containing either the chosen dose of carbamazepine or a placebo (no drug). Randomization to treatment was computer generated. Being housed individually, each mouse was allowed 1 week for adjustment to the specified food bars before breeding. Each female was mated with an experienced male. The presence of a copulatory plug (gestational day 0) terminated mating during a potential 8-day period (2 cycles). After conception, the females were given nestlets throughout weaning.
261
Christensen et al Table I Steady-state plasma concentrations of carbamazepine and its epoxide metabolite in nongravid adult C3H/He mice Carbamazepine dose Carbamazepine (mg/L) Epoxide (mg/L) Total plasma concentration (mg/L) Carbamazepine/epoxide ratio
0.25%
1.0%
1.5G0.2 3.5G0.1 5.0G0.2 0.4G0.1
3.0G0.3 9.0G1.4 12.0G1.7 0.3G0.1
MeanGSEM. 0.25% = 25 mg in 10 g of chow. 1.0% = 100 mg in 10 g of chow. Total plasma concentration is the sum of carbamazepine and epoxide levels.
Beginning on gestational day 18, the dams were checked each morning for the presence of litters. After birth, dams in both treatment groups were given placebo-containing food bars. Maternal behaviors were observed by the same field biologist (C. L. G.). The number of pups was counted for each litter at birth and on postnatal day 4. The presence of any external anomalies was sought. At least eight litters were anticipated in each treatment group with a litter size of at least four viable offspring. This inbred strain of pups was not weighed at birth because of their fragility. Instead, they were weighed on postnatal days 4, 7, 11, 17, and 21 (weaning). The day on which both eyes were opened and the percentage of animals with upper and lower incisor eruptions on postnatal day 10 were recorded. Statistical comparisons were conducted with a Student t test or the Fisher exact method. The data were reported either as a mean percentage or as an SE. Because a litter is the unit of measurement, individual pup data were combined and averaged to determine a litter means and SE. A P!.05 was considered to be statistically significant.
Daily consumption of the food bars containing either 0.25% carbamazepine or placebo was examined in nongravid and gravid mice. Food consumption was not different between the two treatment groups, regardless whether the mouse was gravid (Table II). The body weight increased during pregnancy in both groups. The percentage of food consumed in proportion to body weight in the carbamazepine group was not proportionally different from the placebo group. The daily dose of carbamazepine was 542G35 mg/kg at 1 week premating and was 407G13 mg/kg after 2 weeks while gravid. Most mice conceived in the first estrus cycle and all within two cycles, regardless of therapy. Reproductive performances of the carbamazepine-treated mice were not different from the placebo group (Table III). The percentages of deliveries per mating and per copulatory plug were similar between the groups. The duration of gestation was usually 19 or 20 days in each group. The number of live pups per litter (usually 6) was not different between the carbamazepine and placebo groups. The percentage of male pups (51.9%) was indistinguishable between groups. There were no stillborn pups in the carbamazepine group. No open neural tube defects, midline facial clefting, or extra digits were observed. Two runts (birth weight 2 SD below mean) were found in each of the carbamazepine and placebo groups. Transport/retrieve behaviors of the dams were indistinguishable between the two groups. Pup weights at postnatal days 4, 11, and 21 were not statistically different between carbamazepine- and placebo-exposed offspring. By postnatal day 10, nearly all the bottom teeth had erupted, whereas the top teeth were erupting with equal frequency between the groups. Both eyes opened gradually by postnatal day 15 in the carbamazepine and placebo groups.
Comment Results After 2 weeks of food-drug consumption in nongravid mice, the weight increased (1.7%G1.1%) in the 0.25% carbamazepine group (n = 10) and in the placebo group (2.4%G1.7%) but decreased (ÿ24.4%G1.4%) in the 1.0% carbamazepine group (n = 10). Steady-state plasma concentrations of carbamazepine and its epoxide metabolite are shown in Table I after 5 weeks. Compared with the 0.25% carbamazepine group, those fed the 1.0% carbamazepine food bars had a 2-fold higher concentration of the drug and a nearly 3-fold higher concentration of its metabolite. Carbamazepine/epoxide ratios were not statistically different, however, between the two dose groups. The trough steady-state plasma concentrations (5.0G0.2 mg/L in 0.25% mixture; 12.0G1.7 mg/L in 1.0% mixture) were consistent with the usual therapeutic range in humans.
About 0.52% of pregnant women have a preexisting seizure disorder, and approximately 19,000 infants are born each year in the United States to these women.3 The current investigation was undertaken to find a therapeutic dose of carbamazepine in mice to evaluate its effect on reproductive performances and on perinatal outcomes. Carbamazepine is a first-line drug in the treatment of most forms of epilepsy and also for trigeminal neuralgia. Limitations of several studies about carbamazepine during human pregnancy include a lack of reporting successful cases, no control population, and small series of patients. Safety during pregnancy about carbamazepine is less well known than with phenytoin and phenobarbital. In the current investigation that used a common inbred mouse strain, we were able to establish a therapeutic dose of carbamazepine. The 0.25% concentration in
262
Christensen et al
Table II Food consumption and estimated daily dose of mice treated either with 0.25% carbamazepine or with a placebo Daily food consumption Body weight Nongravid 1 wk premating Placebo Carbamazepine Gravid during second week Placebo Carbamazepine
g
Body weight (%)
21.6G0.9 5.3G0.4 25.3G3.0 21.6G0.9 5.4G0.2 25.1G1.4 31.6G1.2 5.1G0.2 16.2G0.4 30.9G0.8 5.0G0.1 16.3G0.5
MeanGSEM. No significant dierences in food consumption were found between the mice fed 0.25% carbamazepine or a placebo.
the food bars produced a complete response to electroshock stimulation while not reducing food consumption or weight. Our data also revealed trough steady-state plasma concentrations of carbamazepine and its primary epoxide to be within the reported seizure protection threshold in mice and in adult humans (4-12 mg/L).14 This finding is especially important because detrimental effects from in utero exposure to carbamazepine may result from higher levels of epoxide intermediaries.8 The current study reports that a daily dose (542G35 mg/kg) of this centrally acting drug did not impair ovulation or delay conception. Conception rates with this mouse strain are comparable to that reported at a standard national laboratory.15 Screening for newborn development revealed no additional risks with carbamazepine exposure in utero. Reproductive toxicity (stillbirth, runting, preterm delivery) was not apparent with this ‘‘highrisk’’ inbred mouse strain exposed to a daily dose of carbamazepine before and throughout gestation. Results from this placebo-control trial offset worrisome information from pharmaceutic labeling using megadoses in rodents. The manufacturer reports that carbamazepine ‘‘potentially causes reproductive and fetal harm.’’10 We observed that carbamazepine monotherapy did not impair maternal food consumption and weight gain during pregnancy. This clinically relevant dose of carbamazepine was not toxic to pregnancies. A few reports in humans suggest that carbamazepine monotherapy is associated with a lower birth weight in term infants compared with controls.4,16 There was no difference between groups in gestational age at delivery or in the rate of spontaneous abortion. The reduced birth weight (136 g) with prenatal exposure to carbamazepine may be clinically insignificant. In our investigation, the fetal loss rate, gestational age at delivery, birth weight, and proportion of growth restricted mice were not different between the carbamazepine and placebo groups. Our investigation was intended to identify a clinically relevant dose for identifying subtle neurobehavioral changes and possible development shifts. This daily dose
Table III Reproductive performances and physical development of offspring exposed prenatally either to 0.25% carbamazepine or to a placebo Pregnant mice Delivery/mating (%) Delivery/plugs (%) Delivery/first cycle (%) Gestational length (d) Runts Live pups/litter Birth PND 4 Male pups (%) Pup weights (g) PND 4 PND 11 PND 21 Teeth eruption (PND 10) % top % bottom Both eyes open (%) PND 12 PND 13 PND 14
Carbamazepine
Placebo
8 57.1 72.7 75 19.8G0.2 2
8 53.3 72.7 75 19.8G0.3 2
6.2G0.3 6.2G0.3 51.8G8.7
5.7G0.4 5.4G0.5 63.6G9.6
2.6G0.1 7.2G0.2 10.6G0.2
2.5G0.1 7.2G0.2 10.7G0.2
62.5G15.7 97.5G2.5
61.5G15.2 100
18.7G8.0 73.7G8.4 91.6G3.2
12.7G9.6 69.0G7.4 93.7G4.8
MeanGSEM. No significant differences were found between groups exposed to carbamazepine or placebo. PND, Postnatal day.
would not be expected to cause malformations.17 Furthermore, the cause of malformations is multifactorial, not just dose related. We were encouraged that external anomalies, including open neural tube defects and facial clefting, were not observed in the carbamazepineexposed offspring. No differences between the carbamazepine and placebo group were found in newborn development that included food consumption, weight gain, interaction with the mother, and onset of teeth eruptions/eye openings. Research described here will aid in developing more physiologically based dose-response models in animals. Anticonvulsant properties of carbamazepine are thought to result from reducing postsynaptic responses and by blocking posttetanic potentiation. With the use of this therapeutic dose throughout gestation, it will be possible to screen for long-term neurodevelopmental effects from carbamazepine. Interpretation of behavior patterns, social interaction, and cognitive tests of exposed offspring will provide more insight into any subtle long-term development delays.
References 1. Cramer JA, Jones EE. Reproductive function in epilepsy. Epilepsia 1991;32:519-26. 2. Bertilsson W, Tomson T. Clinical pharmacokinetics and pharmacological effects of carbamazepine and carbamazepine-10, 11epoxide: an update. Clin Pharmacol Kinet 1986;11:177-98.
Christensen et al 3. American College of Obstetricians and Gyncologists. Seizure disorders in pregnancy. Washington (DC): The Colleges; 1996. Educational bulletin No.: 231. 4. Diav-Citrinoo, Schechtman S, Arnor J, Ornoy T. Is carbamazepine teratogenic? A prospective controlled study of 210 pregnancies. Neurology 2001;57:321-4. 5. Matalon S, Schechtman S, Goldzweig G, Ornoya. The teratogenic effect of carbamazepine: a meta-analysis of 1255 exposures. Reprod Toxicol 2002;16:9-17. 6. Rosa FW. Spina bifida in infants of women treated with carbamazepine during pregnancy. N Engl J Med 1991;324:674-7. 7. Nulman I, Scolnik D, Chitayat D, Farkas LD, Koren G. Findings in children exposed in utero to phenytoin and carbamazepine monotherapy: independent effects of epilepsy and medications. Am J Med Genet 1997;68:18-24. 8. Jones KL, Lacro RV, Johnson KA, Adams J. Pattern of malformations in the children of women treated with carbamazepine during pregnancy. N Engl J Med 1989;320:1661-6. 9. Stanton ME. Workshop on the qualitative comparability of human and animal development neurotoxicity in humans and laboratory animals. Neurotoxicol Teratol 1990;12:261-7. 10. Tegretol. Physicians’ desk reference, 25th ed. Montvale (NJ): Medical Economics; 2002. p. 2204-7.
263 11. Rayburn WF, Christensen HD. The mouse as a model to investigate neurobehavioral effects from prenatal exposure to prescription medications. J Matern Fetal Invest 1997;7:3-6. 12. Woodbury LA, Davenport VD. Design and use of a new electroshock seizure apparatus, and analysis of factors altering seizure threshold and pattern. Arch Int Pharmacodyn Ther 1952; 92:97-104. 13. Bourgeois BFD, Wad N. Individual and combined antiepileptic and neurotoxic activity of carbamazepine and carbamazepine10, 11-epoxide in mice. J Pharmacol Exp Ther 1984;231: 411-5. 14. Battino D, Binelli S, Bossi L. Plasma concentrations of carbamazepine and carbamazepine 10, 11-epoxide during pregnancy and after delivery. Clin Pharmacokinet 1985;10:279-84. 15. Green MC, Witham BA. Handbook on genetically standardized JAX mice. 4th ed. Bar Harbor (ME): Jackson Laboratory; 1991. p. 35-8. 16. Bertollini R, Kallen B, Mastroiacovo P, Ribert E. Anticonvulsant drugs in monotherapy: effects on the fetus. Eur J Epidemiol 1987; 3:164-71. 17. Holmes LB, Harvey EA, Goull BA, Huntington KB, Khoshbin S, Hayes TM, et al. The teratogenicity of anticonvulsant drugs. N Engl J Med 2001;344:1132-8.