• Home
  • Pamphlets and Brochures
  • Donate now
  • Frequently Asked Questions
  • Please read

Our Helplines

Motherisk Helpline
1-877-439-2744 (Toll-free)
416-813-6780 (Toronto and GTA)

Alcohol and Substance

Dextromethorphan: Extrapolation of findings from reproductive studies in animals to humans

Adrienne Einarson, RN; Gideon Koren, MD, FRCPC

October, 1999



One of my patients, who is now 8 weeks pregnant, just read in the newspaper that dextromethorphan (DM), an antitussive found in a variety of cough medicines, caused birth defects in chicken embryos. The author of the study stated that even one dose could be dangerous and that he would never allow his wife to use this drug if she were pregnant. My patient was understandably very concerned because last week she was suffering from a nasty cough and had been advised by her pharmacist to use a cough mixture containing DM, which she subsequently took for several days.


You may reassure your patient that she did not put her baby at risk by using this substance. Dextromethorphan has been on the market for many years and has never been implicated as a human teratogen. Furthermore, chick embryos are not a good model for predicting teratogenic potential in humans and, consequently, were abandoned as such more than 30 years ago.

Animal studies can, but do not always, predict whether a drug will be teratogenic in humans. The main role of animal studies is to help researchers understand the mechanisms of teratogenicity. Unfortunately, animal studies were poor predictors in the case of thalidomide; the drug was tested on rats and mice, but did not originally produce birth defects.1 On the other hand, some drugs have been found teratogenic in animals and not in humans.2,3

Today, when new drugs are screened for teratogenicity, three different animal models are required for testing. Quite frequently, when certain drugs are tested on different animal species, birth defects occur, as happened in the DM study.4 Interspecies differences regarding the teratogenicity of drugs can result from differing pharmacokinetic processes that determine the crucial concentration-time relationships in an embryo.

Protein binding in the mother is also an important determinant of placental transfer because only free concentrates in maternal plasma can reach equilibration with the embryo during organogenesis. This parameter differs greatly across species (eg, valproic acid, which has five times higher free concentrations in mice and hamsters than in monkeys and humans). Laboratory animals usually eliminate drugs much faster than humans; the drastic drug fluctuations that occur during teratogenicity studies of animals, therefore, do not happen to the same degree in humans.5 Animals are also given far higher doses of drugs than humans would ever receive (eg, as in the clarithromycin study3).

Extrapolating data from these studies to humans and subsequently publishing supposed effects in the mass media, without a clear understanding of the differences, can cause women and their health professionals unnecessary anxiety. This kind of information sells newspapers but does nothing to inform the public of the real risks or safety of drugs during pregnancy.

Over the years, many studies in the literature have selectively reported research results, which ultimately find their way into the mass media.6,7 Papers also document the bias against publication of reassuring studies.8-10 The public should be wary of gaining medical knowledge through newspaper and other media reports.

The only way to ascertain the ultimate risk or safety of drugs during pregnancy is to do human studies. Prospective, controlled, epidemiologic studies of the pregnancy outcomes of women who consumed a particular drug during pregnancy is one example. Even this type of research is not ideal because, for obvious ethical reasons, women cannot be enrolled in randomized, controlled studies, and because the limited size of the studies means they do not have the statistical power to detect increased risk of rare malformations. Another example is case-control studies, where drug use among mothers whose babies have a specific malformation is compared with that of mothers whose babies do not. This method has a much higher sensitivity, but is more prone to bias.

Due to the unjustified level of anxiety stirred up by inappropriate reporting of the chick-embryo study, we decided to carry out a prospective controlled study at The Motherisk Program to provide additional evidence-based information on human exposures. This information will be added to that of two previous studies on the safety of using DM during pregnancy. The Collaborative Perinatal Project monitored 50 282 mother-child pairs, 300 of whom took DM in the first trimester and had no increase in birth defects above the 1% to 3% baseline rate.11 A surveillance study of 59 women who had used DM in the first trimester documented one malformation.12 Upon completion, this study will provide further evidence to guide women and their health professionals in making informed decisions about use of DM during pregnancy.13


We thank W.K. Buckley Limited, McNeil Consumer Products Company, Novartis Consumer Health Canada Inc, Warner Lambert Consumer Healthcare, and Whitehall-Robins Inc for their collaborative support of our study.


  1. Newman CGH. Clinical aspects of thalidomide embryopathyóa continuing preoccupation. Teratology 1985;32:133-44.
  2. Koren G, Pastuszak A, Ito S. Drugs in pregnancy. N Engl J Med 1998;338(16):1128-37.
  3. Einarson A, Phillips E, Mawji F, DíAlimonte D, Schick B, Addis A. A prospective multicentre study of clarithromycin in pregnancy. Am J Perinatol 1998;15(9):523-5.
  4. Andalaro V, Monaghan D, Rosenquist T. Dextromethorphan and other n-methyl-d-aspartate receptor antagonists are teratogenic in the avian embryo model. Pediatr Res 1998;43:1-7.
  5. Nau H. Species differences in pharmacokinetics and drug teratogenesis. Environ Health Perspect 1986;70:113-29.
  6. Cleophas RC, Cleophas TJ. Is selective reporting of clinical research unethical as well as unscientific? Int J Clin Pharmacol Ther 1999;37(1):1-7.
  7. Newcombe RG. Towards a reduction in publication bias. BMJ (Clin Res Ed) 1987;295(6599):656-9.
  8. Proudfoot AD, Proudfoot J. Medical reporting in the lay press. Med J Aust 1981;10(1):8-9.
  9. Koren G, Klein N. Bias against negative studies in newspaper reports of medical research. JAMA 1991;266(13):1824-6.
  10. Kleijnen, Knipschild P. Review articles and publication bias. Arzneimittelforschung 1992;42(5):587-91.
  11. Heinonen OP, Slone D, Shapiro S. Birth defects and drugs in pregnancy. Littleton, Md: Publishing Sciences Group; 1997.
  12. Aselton P, Jick H, Milunsky A, Hunter J, Stergachis A. First trimester drug use and congenital disorders. Obstet Gynecol 1985;65:451-5.
  13. Einarson A, Lyszkiewicz D, Koren G. The safety of dextromethorphan in pregnancy: a prospective controlled study, Colorado 1999 [abstract]. Teratology 1999;59(6):377.

© Canadian Family Physician 1999;45:2053-7.
Valid XHTML 1.0 Transitional [Valid RSS]

Motherisk is a proud affiliate of MotherToBaby, an information service of the non-profit Organization of Teratology Information Specialists (OTIS) across North America.

* - "MOTHERISK - Treating the mother - Protecting the unborn" is an official mark of The Hospital for Sick Children. All rights reserved.

The information on this website is not intended as a substitute for the advice and care of your doctor or other health-care provider. Always consult your doctor if you have any questions about exposures during pregnancy and before you take any medications.

Copyright © 1999-2018 The Hospital for Sick Children (SickKids). All rights reserved.

The Hospital for Sick Children (SickKids) is a health-care, teaching and research centre dedicated exclusively to children; affiliated with the University of Toronto. For general inquires please call: 416-813-1500.

  |  Contact SickKids  |  Terms of Use