Date: June 11th, 2019

Reference: Huybrechts et al. Association of Maternal First-Trimester Ondansetron Use With Cardiac Malformations and Oral Clefts in Offspring. JAMA Dec 2018.

Guest Skeptic: Dr. Nick Papalia completed his MD at Western University. He is currently completing an Obstetrics and Gynecology residency at the University of Calgary.

Case: A 24-year-old woman G2T1P1A0L1 who presents with nausea and vomiting of pregnancy at nine weeks gestational age. She has stopped her iron pills, taken ginger, used acupressure, tried vitamin B6 with doxylamine and dimenhydrinate. She is frustrated nothing is working and wants to try something else. Her friend got little wafers that dissolved under her tongue (ondansetron). She is worried because her google search said it could cause a birth defect like a cleft lip.

Background: Many women suffer from nausea and vomiting when pregnant. These symptoms can become clinically significant in over 30% of woman. Hyperemesis gravidarum is the most common reason for hospitalization in early pregnancy and impacts a small percentage of these pregnancies.

The Society of Obstetricians and Gynecologists of Canada (SOGC) published a guideline for the management of nausea and vomiting of pregnancy in 2016 (Campbell et al 2016). They make 13 recommendations:

  1. Women experiencing nausea and vomiting of pregnancy may discontinue iron-containing prenatal vitamins during the first trimester and substitute them with folic acid or adult or children’s vitamins low in iron. (II-2A)
  2. Women should be counselled to eat whatever pregnancy-safe food appeals to them and lifestyle changes should be liberally encouraged. (III-C)
  3. Ginger may be beneficial in ameliorating the symptoms of nausea and vomiting of pregnancy. (I-A)
  4. Acupressure may help some women in the management of nausea and vomiting of pregnancy. (I-B)
  5. Mindfulness-based cognitive therapy as an adjunct to pyridoxine therapy may be beneficial. (I-B)
  6. Pyridoxine monotherapy or doxylamine/pyridoxine combination therapy is recommended as first line in treating nausea and vom- iting of pregnancy due to their efficacy and safety. (I-A)
  7. Women with high risk for nausea and vomiting of pregnancy may benefit from preemptive doxylamine/pyridoxine treatment at the onset of pregnancy. (I-A)
  8. H1 receptor antagonists should be considered in the management of acute or chronic episodes of nausea and vomiting of pregnancy. (I-A)
  9. Metoclopramide can be safely used as an adjuvant therapy for the management of nausea and vomiting of pregnancy. (II-2B)
  10. Phenothiazines are safe and effective as an adjunctive therapy for severe nausea and vomiting of pregnancy. (I-A)
  11. Despite potential safety concerns of ondansetron use in pregnancy, ondansetron can be used as an adjunctive therapy for the management of severe nausea and vomiting of pregnancy when other antiemetic combinations have failed. (II-1C)
  12. Corticosteroids should be avoided during the first trimester because of possible increased risk of oral clefting and should be restricted to refractory cases. (I-B)
  13. When nausea and vomiting of pregnancy is refractory to initial pharmacotherapy, investigation of other potential causes should be undertaken. (III-A)

The primary literature used to support the acupressure recommendation is very weak. A review by Roscoe and Matteson 2002 showed conflicting results from seven methodologically flawed trials. The conclusion was that acupressure might (might not) be beneficial.

The American College of Obstetricians and Gynecologists (AGOC) has published a practice Bulletin (January 2018) on nausea and vomiting of pregnancy. It starts with non-pharmacologic options. Pharmacologic options include Vitamin B6 alone or in combination with doxylamine. The next step is adding dimenhydrinate or prochlorperazine or promethazine. The algorithm then dichotomizes into no dehydration or dehydration with persistent symptoms. It is this step when ondansetron is added as a possible treatment.

There is conflicting evidence on the fetal safety of ondansetron. An observational study concluded that ondansetron taken during pregnancy was not associated an important increased risk of fetal harm (Pasternak et al NEJM 2013).

Clinical Question: Is ondansetron exposure in pregnancy associated with congenital anomalies (primarily cardiac and oral clefts)?

Reference: Huybrechts et al. Association of Maternal First-Trimester Ondansetron Use With Cardiac Malformations and Oral Clefts in Offspring. JAMA Dec 2018.

  • Population: Pregnant women ages 12-55 years of age on Medicaid from three months prior to conception to one month postpartum. Infants were required to have Medicaid coverage for the first three months of life.
  • Exposure: Pregnant women who filled at least one prescription for ondansetron duringthe first three months (12 weeks) of pregnancy.
    • Excluded: Women who filled a prescription during the three months before the start of their pregnancy.
  • Comparison: Pregnant women who filled a prescription for pyridoxine, promethazine, metoclopramide, or any of these alternative treatments.
  • Outcome:
    • Primary Outcomes: Presence of cardiac malformations and oral clefts diagnosed within 90 days after delivery.
    • Secondary Outcomes: Subgroups of cardiac malformations and oral clefts were evaluated along with congenital malformations overall.

Authors’ Conclusions: “Among offspring of mothers enrolled in Medicaid, first-trimester exposure to ondansetron was not associated with cardiac malformations or congenital malformations overall after accounting for measured confounders but was associated with a small increased risk of oral clefts.”

Quality Checklist for Observational Study:

  1. Did the study address a clearly focused issue? Yes
  2. Did the authors use an appropriate method to answer their question? Yes
  3. Was the cohort recruited in an acceptable way? Yes
  4. Was the exposure accurately measured to minimize bias? Unsure
  5. Was the outcome accurately measured to minimize bias? Yes
  6. Have the authors identified all-important confounding factors? No
  7. Was the follow up of subjects complete enough? Yes
  8. How precise are the estimate of risk? Fairly tight for cardiac, less tight for oral clefts and very tight for any congenital abnormality.
  9. Do you believe the results? Unsure
  10. Can the results be applied to the local population? Unsure
  11. Do the results of this study fit with other available evidence? Yes

Key Results: The data set included 1.8 million pregnancies from 1.5 million women. The mean age was 24 years and 5% were potentially exposed to ondansetron in the first trimester.

 No increased risk cardiac malformation but a slight statistical increase in risk of oral clefts.

  • Primary Outcomes:
    • Cardiac Malformation: Adjusted RR 0.99 (95% CI, 0.93 to 1.06)
    • Oral Clefts: Adjusted RR 1.24 (95% CI, 1.03 to 1.48)
    • Overall Malformations: Adjusted RR 1.01 (95% CI, 0.98 to 1.05)

1) Observational Study: We can only conclude an association and that some other confounders and co-variates could have been responsible for any difference observed. They did try to control for these issues with adjusted analyses and propensity scores.

2) Primary Outcomes: There can be only one, primary outcome (Highlander). Unsure why they had three (cardiac, oral cleft and overall). Why not pick one and then have others as secondary outcomes? It makes me skeptical. I could not see where this was a pre-registered trial. Could it be that they originally set it up to demonstrate no increased risk of oral cleft and then when the data demonstrated a slight increase, they changed their primary outcome to all malformations?

3) Exposed Women: They had different baseline characteristics than those women not exposed to ondansetron. Exposed women were more likely to smoke, have psychiatric diagnosis, neurologic condition, be white, and fill a prescription for other nausea and vomiting medication, psychotropics, steroids, and suspected teratogens.

4) Conflicts of Interest: There were multiple financial conflicts of interest declared by the authors. This does not make the data wrong, but it does make us more skeptical of their interpretations of the results.

5) External Validity: This dataset captured 1.8 million pregnancies but that was only 50% of all the pregnancies in the USA. They were Medicaid patients who may be different than those with other insurance. It also may not be representative of Canada and other countries with different healthcare systems.

Comment on Authors’ Conclusion Compared to SGEM Conclusion: We agree with the authors’ that there was a lack of association any congenital malformations or with cardiac malformations. They did observe a small increased risk of oral clefts associated with ondansetron exposure.

SGEM Bottom Line: There appears to be no overall increased risk of fetal congenital malformations in women exposed to ondansetron during their first trimester. There may be a small increased risk of oral clefts.

Case Resolution: Your patient requested if any other medications are available that are safer and you advised her that metaclopromide is recommended to be utilized first. She responds well to an oral dose of 5mg so you discharge her with a prescription and advise her to return if her nausea and vomiting does not improve and she is unable to tolerate oral intake or is losing weight.

Dr. Nick Papalia

Clinical Application: This study did show an association between oral clefts and the ondansetron exposed population in the 6 week to 12 week window. There is biologic plausibility, but I can’t get past the fact that if oral clefts are more likely with no increased risk of ANY anomaly, is something then decreased? This doesn’t make sense and would suggest these results occurred due to random chance or perhaps p-hacking in exploratory analyses.

Practically, nausea and vomiting in pregnancy is most prominent from 8-12 weeks gestational age and generally improves in the second trimester. Ondansetron may have a slightly increased risk of an oral cleft but if uncontrolled with first or second line treatments, it is reasonable to prescribe a short course of as needed or scheduled ondansetron. I advise my patients of the limited literature associating the oral clefts with ondansetron but also that unmanaged nausea, vomiting can lead to weight loss, electrolyte disturbances, and increased cortisol release which may have its own adverse effect on the pregnancy.

What Do I Tell My Patient? Nausea and vomiting in pregnancy is common and can be difficult to treat. You need to be able to eat to grow a healthy baby. There are a few other options besides using ondansetron. We can try those first

Keener Kontest: Last weeks’ winner was Dr. Yuki Mori a PGY6 fellow at the US naval Hospital Yokosuka. They knew it was 1954   when Lieutenant Colonel Carl Hughes of the US military published an article describing the use of REBOA (Hughes CW Surgery 1954).

Listen to the SGEM podcast on iTunes to hear this weeks’ question. If you know the answer send an email to with “keener” in the subject line. The first correct answer will receive a cool skeptical prize.

Remember to be skeptical of anything you learn, even if you heard it on the Skeptics’ Guide to Emergency Medicine.