Prescribing in pregnancy - technical
- Identifying teratogenic drugs
- Effects of drugs on the fetus
- Effects of pregnancy on drugs
Only prescribe if it is essential to do so—the guiding principle when prescribing in pregnancy, or in a woman who may become pregnant, is to use no drug at all if possible. If it is necessary to prescribe, then use the smallest number of drugs in the lowest possible doses, balancing the benefits of treatment for the mother against what (if anything) is known about its risks to the developing baby. Prepregnancy evaluation is invaluable for chronic conditions.
Stage of pregnancy—the period of organogenesis, which extends for about 2 months following the last period, is the critical time for major structural abnormalities, but growth and development can be affected by some drugs in the remainder of pregnancy.
Handling of drugs by the body—the distribution and elimination of some drugs is affected to a clinically significant extent by pregnancy.
Breastfeeding—many commonly used drugs, including some that are unsafe in pregnancy, can safely be used by mothers who are breastfeeding.
Prescribing in pregnancy is essentially about balancing risks. The damage that a drug may cause to the fetus must be weighed against the harm that may befall the mother and her unborn child if a disease goes unchecked.
While knowledge in most therapeutic areas has grown rapidly in recent decades, information on the use of drugs in pregnancy has developed sporadically, with case reports being more usual than large, prospective clinical trials. The reasons are not surprising and largely relate to concern about teratogenesis.
Thalidomide is a name inescapably associated with prescribing in pregnancy. Drug-induced fetal abnormality did not begin with thalidomide: there is an Old Testament exhortation to have ‘no strong drink, neither eat any unclean thing’ during pregnancy.
However, the scale of the thalidomide tragedy brought to the general public for the first time the realization that drugs could harm the developing baby. Thalidomide was marketed in Germany in 1956 and subsequently in other countries as a sedative and hypnotic which had the particular attraction of being safe in overdose. Indeed, the drug was considered so safe that in some countries it was available without prescription.
Then between 1960 and 1961 Germany experienced what amounted to an epidemic of phocomelia, a birth defect involving absence of the long bones with hands and feet being attached directly to the trunk. What had previously been an extremely rare condition (no cases had been reported in the 10 years to 1959) was being seen almost commonly.
Various causes—viral, radioactivity, food preservatives—were considered as culprits, until one doctor retrospectively questioned his patients and found that 20% had taken thalidomide in early pregnancy. On repeat questioning, asking specifically about the drug, 50% admitted taking thalidomide, many having not mentioned it before since the drug was so obviously innocent. In fact, around 80% of women who took thalidomide in the first trimester had a deformed baby. More than 10 000 such babies had been born before the drug was removed from the market.
The thalidomide experience had far-reaching ramifications. Drug regulation as we know it stems largely from this disaster. Doctors and their patients recognized that there is no such thing as a safe drug. In addition, the pharmaceutical industry has largely avoided obtaining systematic information on drug use in pregnancy.
The reasons are obvious and understandable, but for the prescribing doctor the statement that ‘the safety of this drug in pregnancy has not been established’ is not helpful when faced with a woman who is, or may become, pregnant.
Identifying teratogenic drugs
Information on drug-induced fetal abnormality comes from case reports, case studies, and epidemiological studies. Case reports are a two-edged sword. Describing a single association between a drug and a fetal abnormality can be very useful in first identifying a real problem: warfarin was first linked to teratogenesis in this way. However, the problem with case reports is that they may be showing nothing more than a chance association, because fetal abnormalities occur in around 2% of pregnancies, and caution must be exercised in their interpretation. This is well demonstrated by the Debendox saga.
Most cases of morning sickness do not require treatment. However, some do, and the drug for which most information is available was withdrawn from the market in 1983 in view of mounting public concern about its safety. This drug was a mixture of doxylamine succinate and pyridoxine hydrochloride and was marketed as Debendox or Bendectin. Despite having been used by over 30 million pregnant women over a quarter of a century, and notwithstanding carefully designed clinical trials suggesting that the drug was not teratogenic, individual case reports linking the use of the drug to fetal abnormality were given considerable publicity and led to its withdrawal.
In view of the extremely high number of exposures, many chance associations between drug use and fetal abnormality were inevitable. This episode illustrated that in an emotional area such as the use of drugs during pregnancy, well-chosen and carefully presented anecdotes can be more powerful than a substantial body of scientific data carefully accumulated over many years.
Case studies are more secure in that they describe several patients where the same drug and malformation were linked: phenytoin and the retinoids were found to be teratogenic in this way. Epidemiological studies are of two major types: cohort studies, which prospectively study exposed and unexposed groups, and case–control studies, which retrospectively compare the pregnancies of abnormal and normal offspring.
So far as teratogenesis is concerned, case–control studies are the norm because of the size and expense of cohort studies. The relationship between use of diethylstilbestrol in the first trimester and vaginal adenocarcinoma in teenage offspring was found in a case–control study.