Many women who come to our clinic with questions regarding the use of medications during pregnancy suffer from some type of anxiety disorder.  Most of those women benefit from treatment with an SSRI or SNRI antidepressant; however, there are a significant number of women who either do not experience complete symptom resolution with or cannot tolerate these medications and, thus, also receive treatment with a benzodiazepine.  The benzodiazepines are a class of anti-anxiety medications which includes clonazepam (Klonopin), lorazepam (Ativan), and alprazolam (Xamax). They are commonly used to treat anxiety in women, yet there is less data on the reproductive safety of the benzodiazepines, as compared to other medications.  

In a recent study, Sheehy and colleagues report on the risk of spontaneous abortion (or miscarriage) in women treated during the first trimester with benzodiazepines.  In this nested case-control study, which is part of the Quebec Pregnancy Cohort, researchers analyzed data from all pregnancies covered by the Quebec Prescription Drug Insurance Plan between January 1, 1998, and December 31, 2015. Spontaneous abortion was defined as a pregnancy loss between the beginning of the sixth and the end of the19th week of gestation. Benzodiazepine exposure was defined as one or more filled prescriptions between the first day of the last menstrual period and the index date (the date of the spontaneous abortion).

Of the 442,066 pregnancies included in the Quebec Pregnancy Cohort, 27,149 (6.1%) ended in a spontaneous abortion.  Among pregnancies ending in spontaneous abortion, 375 (1.4%) were among women taking benzodiazepines in early pregnancy, as compared to 788 (0.6%) in a group of  134,305 matched control pregnancies which did not end in miscarriage (crude OR, 2.39; 95% CI, 2.10-2.73). After adjusting for potential confounders, including maternal mood and anxiety disorders before pregnancy, benzodiazepine exposure in early pregnancy was associated with an increased risk of spontaneous abortion (adjusted OR, 1.85; 95% CI, 1.61-2.12). The risk was similar among pregnancies exposed to short-acting (284 exposed cases; adjusted OR, 1.81; 95% CI, 1.55-2.12) and long-acting (98 exposed cases; adjusted OR, 1.73; 95% CI, 1.31-2.28) benzodiazepines.

While this study may raise concerns, it should be noted that the prevalence of miscarriage in women who used benzodiazepines, although higher than in non-users, was in the normal range.  In the Quebec cohort as a whole, 6.1% of women had a spontaneous abortion. Based on the findings of this report, the estimated risk of spontaneous abortion was about 12% (adjusted OR, 1.85).The risk of miscarriage in the general population is somewhere around 15% to 20%.

One of the things that makes these studies so difficult to analyze is that there are many factors which may impact rates of spontaneous abortion, and it is difficult, if not impossible, to take them into consideration.  The women who had miscarriages were more likely than those who did not miscarry to be older, to have histories of substance abuse, to have histories of mood or anxiety disorders, and to have more hospitalizations and medical visits.  Women who miscarried were less likely to use folic acid. While it is possible to control for some of these confounding factors, others are more difficult to assess and to account for.

Impact of underlying illness on risk for miscarriage –  While the researchers controlled for a diagnosis of anxiety or mood disorder prior to pregnancy as a potential confounding variable, they were not able to control for anxiety symptoms during pregnancy.  We cannot conclude that the women who choose to use benzodiazepines during pregnancy are the same as those who do not.  It is likely that the women who used benzodiazepines during pregnancy may have had more severe anxiety symptoms than the women who chose not to use benzodiazepines.  And this is a pertinent point, as a number of studies have shown that anxiety symptoms during pregnancy may increase risk for miscarriage, so it would be very difficult using this study design to distinguish between the effects of exposure to medication versus exposure to the underlying illness.  

Exposure to medication – We have often commented that in these studies relying on large databases, it is not possible to confirm that, while the patient may have filled a prescription, that she is in fact taking the medication in question. In fact, Lupattelli and colleagues noted that about half of all women are adherent to treatment with psychotropic medications.  In addition, prescriptions for benzodiazepines are typically written to be used “as needed”, as opposed to daily or on a regular basis.  Using this study design, it is impossible to assess how frequently women were actually taking the medication.

Impact of withdrawal on risk – Using this type of study design, it is impossible to distinguish between women who use benzodiazepines consistently to manage anxiety symptoms from women who discover they are pregnant and abruptly discontinue the medication.   If a women were to abruptly discontinue a benzodiazepine, she might experience withdrawal symptoms, including fluctuations in heart rate and blood pressure and, less commonly, seizure. Withdrawal symptoms may complicate the picture and may be a potential confounding factor in determining the risks associated with exposure to medication.  


The Clinical Bottom Line

While we must consider the risks of various medications, we must also consider the risks of untreated illness in the mother.  Untreated anxiety in the mother has been associated with worse outcomes, including shorter gestation, lower birth weight, higher risk of complications, and increased vulnerability to postpartum depression and anxiety.  Thus, simply avoiding or discontinuing the medication may not be the best or safest option.

When we consult with women who are taking benzodiazepines, we ask the following questions:

  • Why is this patient taking the medication? Anxiety symptoms? Insomnia? Phobia?
  • How is the medication taken? On a daily basis or as needed?
  • Is it possible to gradually taper the benzodiazepine?  
  • If symptoms recur, are non-pharmacologic treatments, such as cognitive-behavioral therapy, effective in this setting?
  • If non-pharmacologic options are not successful, could treatment with an SSRI or an SNRI alone be an option?

In a perfect world, this discussion should take places long before conception.  Tapering benzodiazepines slowly, over a period of many months, is more likely to be successful than a quick taper, especially in women who have taken benzodiazepines consistently over a long period of time.  Time is also essential when it comes to finding alternative treatments, whether it is a behavioral approach or a different medication.

Unfortunately, in many cases, it may not be possible to completely eliminate treatment with benzodiazepines. While this study suggests an association between benzodiazepine exposure and increased risk of spontaneous abortion, the absolute risk does not exceed what we would normally expect in the general population.  Further study is needed to better understand the risks associated with benzodiazepines and to better estimate the contribution of other factors, particularly severe of symptoms during pregnancy.


Ruta Nonacs, MD PhD

Sheehy O, Zhao JP, Bérard A.  Association Between Incident Exposure to Benzodiazepines in Early Pregnancy and Risk of Spontaneous Abortion.  JAMA Psychiatry. 2019 May 15.


Related Posts