F. M. Sullivan

A comparison of the teratogenic activity of the antiepileptic drugs carbamazepine, clonazepam, ethosuximide, phenobarbital, phenytoin, and primidone in mice.

Abstract The teratogenic activity of six commonly used antiepileptic drugs, carbamazepine, clonazepam, ethosuximide, phenobarbital, phenytoin, and primidone, was studied at three dose levels (3, 9, and 18 times the human therapeutic dose) in CD1 mice. The drugs, suspended in 1% carboxymethylcellulose, were administered by gastric intubationon Days 6–16 of pregnancy, which covers the period from implantation to the end of structural development. The dams were killed 1 day before term and the fetuses were examined. Full-length cleft palates were observed in all treated groups with the exception of the clonazepam-treated group. This defect was not observed in any of the control groups. Other defects such as exencephaly, enlarged cerebral ventricles, exomphalos, open eyes, undescended testes, abnormal palatal bones other than cleft palate, and fused ribs were seen less frequently. If these results are assessed by ranking the drugs in order of teratogenicity, assessed on either a pooled fetus or a pooled litter basis, the drugs fall into three categories: Phenytoin produced the highest incidence of teratogenic effects; carbamazepine, phenobarbital, and primidone produced a lower but still significant incidence of defects; and clonazepam and ethosuximide were the least teratogenic. Nevertheless, there was at least a two-fold increase in defects in these latter two treatment groups compared with the controls.

Plasma corticosterone responses to stress following chronic oral administration of diazepam in the rat

The effect of daily, oral administration of diazepam on plasma corticosterone responses to stressors of varying intensity was investigated. In rats exposed to the mild stress of noise, diazepam, 10 mg kg−1 but not 1.0 or 0.1 mg kg−1, reduced plasma corticosterone concentrations by 30% in comparison with controls. However, in rats exposed to the more severe stressors, foot‐shock or immobilization, none of these doses of diazepam reduced plasma corticosterone responses. In unstressed rats, diazepam 10 mg kg−1 raised plasma corticosterone concentrations. It is suggested that plasma corticosterone concentrations are not a reliable indicator of the tranquillizing effect of diazepam during stress.

Studies on the teratogenic effects of different oral preparations of caffeine in mice.

Caffeine in doses up to 250 mg/kg per day in drinking water or up to 150 mg/kg per day in sustained release pellets was administered to pregnant mice. Apart from a low incidence of cleft palate, in the 50 mg/kg and 150 mg/kg caffeine pellet groups no gross abnormalities were observed which were attributable to caffeine treatment. The most important effect observed was a reduction in fetal weight. Retarded ossification particularly of the supraoccipital bones was observed in fetuses when caffeine (150 mg/kg) was administered in drinking water but not when the same dose was given as a sustained release pellet. Analysis of caffeine blood level data showed that the total exposure from the pellets was greater than from the drinking water. It would thus appear that the effect on the supraoccipital bones is an indirect one mediated through reduced food and water intake of the dams when caffeine is administered in drinking water.

The Consequences of Iron Overdose and its Treatment with Desferrioxamine in Pregnancy

A study was carried out to assess the effect on the outcome of pregnancy of iron overdose and its treatment with desferrioxamine. Sixty-eight cases were drawn from those reported to the United Kingdom National Poisons Information Centre and the Teratology Information Service at Guys Hospital, London, and follow-up was obtained in 51 of these. Two were subsequently reported not to be pregnant and there were 49 records of pregnant patients who took iron overdoses and where outcome of the pregnancy was known. Twenty-five of these patients were treated with desferrioxamine. In 48 of the 49 patients the dose of iron allegedly taken was known and in 28 (60%) was over 20 mg kg-1, sufficient to put them at risk of toxicity. In the 36 whose serum iron levels were measured, 20 patients had levels in excess of 60 μmol l-1, indicating a risk of moderate or severe toxicity. Of the 49 pregnancies, 43 resulted in live babies, two had spontaneous abortions and there were four elective terminations. Of the live babies, three were premature, two of whom had problems, and there were three other babies with abnormalities. All babies with malformations were associated with overdoses after the first trimester and so the malformations cannot be directly related to the overdose. A total of 25 patients received desferrioxamine of whom two had malformed babies, but the desferrioxamine can be excluded as a cause. There was no correlation between the serum iron levels and the birthweights. In conclusion, iron overdose in pregnancy can be fatal and antidote treatment if appropriate should not be withheld. The majority of second and third trimester iron overdoses, treated with desferrioxamine or other antidotes, will have a normal pregnancy outcome. The risk of spontaneous abortion is low but cannot be excluded.

Acute studies to investigate the mechanism of action of caffeine as a teratogen in mice.

1 In Charles River CD 1 mice, a single dose of 100 mg kg-1 caffeine injected intraperitoneally on day 14 of pregnancy caused a low incidence of cleft palate in the fetuses. 2 Single oral doses of caffeine of 200 and 300 mg kg-1 but not 100 mg kg-1 on day 14, caused cleft palate in some of the fetuses, but was clearly toxic to the dams. 3 Oral doses of caffeine up to 300 mg kg-1 on day 14 of pregnancy did not reduce utero-placental blood flow, placental transfer function, or amniotic fluid volume. 4 An oral dose of 100 mg kg-1 caffeine induced a marked stimulation of adrenocortical secretion producing plasma corticosterone levels of 1248 ± 129 ?g per 100 ml by 2 h and with elevated levels persisting more than 8 h. 5 It is suggested that the elevated plasma corticosterone is the cause of the cleft palate induced in mice by caffeine. Since corticosterone is a known cleft palate inducer in mice but not in man these results do not predict a hazard from normal caffeine consumption in man.

Paracetamol Poisoning in Pregnancy: An Analysis of the Outcomes of Cases Referred to the Teratology Information Service of the National Poisons Information Service

A study was carried out to investigate the outcome of pregnancy in 115 women who had been exposed to paracetamol overdose. Follow up was obtained in 48 cases. Exposure occurred in all trimesters, and the most striking feature of this series is that the majority of the pregnancy outcomes were normal. None of the mothers died. There were 39 live born infants with no malformation, 14 of whom had been exposed in the first trimester. Four babies, exposed in the third trimester had neonatal problems, but these seem unrelated to paracetamol. There were two live born infants with gross malformations (spina bifida occulta; and cleft lip and palate). However, as the overdoses occurred at weeks 26 and 28 respectively, long after the structural development of these organs, the malformations could not have been caused by the paracetamol. There were two spontaneous abortions, both in the first trimester, which occurred two weeks after the overdose which may be related to the paracetamol. The overall conclusion is that paracetamol overdose per se is not necessarily an indication for termination of pregnancy.

Effects of stress and adrenocorticotrophin administration on plasma corticosterone levels at different stages of pregnancy in the mouse.

Stress or administration of ACTH to pregnant mice gave rise to much higher plasma corticosterone levels in the second half of pregnancy than in the first half, suggesting that there may be increased adrenal sensitivity to ACTH or decreased metabolism of corticosterone during the second half of pregnancy.

Reproductive Toxicity Tests: Retrospect and Prospect:

1 The design of the classical three segment reproductive test for new drugs is described and the relative merits of the USA/EEC and the Japanese guidelines are discussed. 2 The importance of pharmacokinetic studies in the interpretation of teratology studies and extrapolation to humans is mentioned in relation to caffeine, sodium valproate and cyclophosphamide. 3 Changing ideas on the design of multi-generation studies are reviewed. 4 Recent developments in the fields of behavioural teratology, chemicals in breast milk and transplacental carcinogenesis are described.

The Metabolism of Primidone in Non-pregnant and 14-day Pregnant Mice

1. Non-pregnant and day-14 pregnant mice treated with a single dose of primidone (5 to 150 mg/kg) by gastric intubation were bled at 1 and 4 h after dosage, and the plasma analysed for primidone, phenobarbitone and phenylethylmalondiamide (PEMA) by g.l.c. 2. Marked differences in the rate of metabolism were observed between the non-pregnant and day-14 pregnant mouse. Plasma levels of primidone and PEMA, but not phenobarbitone, were much lower in the day-14 pregnant mouse than in the non-pregnant animal.

Effect of prenatal phenytoin administration on brain tryptophan metabolism of rat offspring during the preweaning period.

Serum 5‐hydroxytryptamine (5‐HT) and 5‐hydroxyindole acetic acid (5‐HIAA) concentrations in control rat offspring increased progressively during the preweaning period reaching adult values by day 21. It has been shown that prenatal phenytoin administration (100 mg kg−1 orally, days 7–19 of pregnancy) increased serum tryptophan and brain tryptophan, 5‐HT and 5‐HIAA of rat offspring at 3 days of age but not at 4, 15 or 21 days of age. The effect of prenatal phenytoin administration on the offspring at 3 days of age was not observed when these pups were cross‐fostered to control mothers at 2 days of age suggesting that the alteration in brain tryptophan metabolism during the development of tryptaminergic neurons in rat offspring, as a result of prenatal phenytoin administration is mediated through changes in lactation or nursing ability of the mothers. It is important that such non‐specific factors are controlled when studying the effect of prenatally administered drugs on neonatal brain transmitter concentrations.