teratology

Antidepressants in pregnancy

This morning BBC News are running with this rather terrifying looking story about the dangers of antidepressants in pregnancy. This is an area that I deal with pretty commonly, so I thought you may be interested in my assessment of the situation. 

First thing to note: things can go wrong in even a normal, healthy pregnancy. There is always a risk of malformations or miscarriage, and unfortunately these things can happen for reasons that we dont understand. The risks are usually low, but are increased by things like increased age, obesity, illnesses etc.

One of those illnesses can be uncontrolled depression. "But how can feeling a bit sad harm an unborn baby?" I hear you ask. Well firstly, depression can be a very serious illness which should be taken seriously. It may even be terminal. Pregnancy is a time of massive changes, and as a consequence is a high risk time where someone's mental health can destabilise. If you have depression, you may not be looking after yourself properly: you might not be eating well, you might be avoiding exercise etc. In the worst, most tragic cases, suicide attempts might happen. We don't have enough data to put figures on how much of an increase in risk this all adds up to, but we do know that it can increase risks in a pregnancy if not sufficiently controlled.

Of course, this doesn't even take into account the more nebulous risks to both the child and mother- how will having a depressed mum impact psychologically on the child, how will the bond be affected, and what are the long term effects of this?  

So what of the SSRIs, the most commonly used type of antidepressants in pregnancy? Looking at the risk of cardiac malformations,, the BBC article claims that:

"Currently, prescription guidelines for doctors only warn specifically against taking the SSRI, paroxetine, in early pregnancy."

 It used to be the case that we were aware of the possibility of a cardiac malformation risk with paroxetine. Up until, oh, about 2010, when a large review was published whichsuggested that the increase in risk, if it exists, may be a class effect. The UK Teratology Information Service's Guidance was changed accordingly to be more practical, to remove a heirachy of one particular SSRI, and to make the drug of choice that which is the best for the individual patient (please note that UKTIS are a service for healthcare professionals only, and pateints should not ring them directly). The fact that NICE guidelines haven't yet been updated probably says more about NICE's workload and update schedule than any evil big pharma cover up. 

As an aside, you will notice that there are a lot of words in this post which suggest uncertainty. That is because there is a lot of uncertainty in teratology: because we cant do large robust trials on pregnant women because of ethical concerns, we have to scrape together what we can and make the best of it. There are few certainties in this area.

Strange then, that the BBC are quoting a Prof Pilling from NICE:

"He says the risk of any baby being born with a heart defect is around two in 100; but the evidence suggests if the mother took an SSRI in early pregnancy that risk increases to around four in 100."

I'd love to know where these figures came from. The current status of data on the risks of SSRIs is pregnancy is as follows:

  • There is lots of data, which has had various statistical analysis methods applied to it. 

  • Some of this data suggests no increase in risk

  • Some of it suggests a small increase in risk.

So, with some data saying there isn't an increase and with some saying there is, it is virtually impossible to say for certain if there is an increase. The only thing we can say for certain at this point is that we can't say anything for certain. But given that we have lots of data, and SSRIs are commonly taken in pregnancy, I think we can say that if there is a large increase in risk, we would have known about it by now. So any increase in risk, if it is there, will be low.

Of course the BBC are reporting the relative risk, which sounds more impressive: a doubled risk sounds much more sensational than a small absolute risk. But I'm not even sure where this figure has come from, given the conflicting state of the evidence at the moment. Needless to say, research is oretty much constantly ongoing. 

All of this is a very long winded way of saying: we dont know at the moment. But the fact that we don't know, in the face of how commonly used these drugs are in pregnancy, could be seen as reassuring.

As with all things in healthcare, this is a balance. A balance between the risks of uncontrolled depression and destabilising a mother's mental health during pregnancy, compared with the -as yet unknown but likely to be small- risks of SSRI antidepressants. Of course some women with minor depression might be taking antidepressants unnecessarily, but in cases where it is required, we need to look at the bigger picture. Just focusing on a drug's teratogenic potential is not enough: we need to consider the teratogenic potential of the illness itself, and the impact on everyone's lives that might happen if treatment is withheld. 


The bottom line is, if women are thinking of becoming pregnant or are already pregnant whilst taking an SSRI, and they are worried, they shouldn't stop it of their own accord, but should make an appointment with their GP to have a discussion about their concerns. 

Hxxx

UPDATE: I've been thinking about this 4 in 100 figure for cardiac malformations, and last night tried to find the reference source from it.

I've tweeted @bbcpanorama asking to know where this figure has come from, as have a few others. I've also tweeted @shelleyjofre, the journalist who has mad ethe programme, and have been met with a stony silence. This is really unfortunate, given that to be able to deal effectively with any enquiries from patients relating to this programme, I -and all the other health care professionals dealing with worried mums to be- need to be able to see and appraise the evidence for ourselves.

I have managed to find this document from the MHRA, which does mention a 4 in 100 figure. However, I sincerely hope that this isn't the source in question, given that:

  • The document refers to paroxetine alone, not the whole class of SSRIs

  • There is no date on the document, meaning we have no way of knowing how up to date these figures are.

  • the 4 in 100 figure cited refers to the risk of ALL malformations, not just cardiac ones.

  • the risk cited for cardiac malformations is 2 in 100. Half that which the BBC and Professor Pilling are quoting.

  • The background risk of all malformations cited is 3 in 100, and the background rate of cardiac malformations is 1 in 100. So yes, the relative risk is doubled, but the overal risk remains very low. 

As I say, I really do hope that this isn't the source, and that @bbpanorama or @shelleyjofre are able to provide me with the reference soon.  

Needless to say, I never heard back.

 

Teratology: an area where more critical thinking would be useful

(Teratology: Noun: The scientific study of congenital abnormalities and abnormal formations. From the Greek teras [meaning monster or marvel] and logos [meaning study])

"Hands up, who is afraid of advising someone takes a medicine in pregnancy?"

In the couple of training sessions I've delivered, this has been my opening line. And many hands go up. I regularly give advice on medicines in pregnancy in my job, and it strikes me that it is an area which could hugely benefit from more critical thinking and a reliance on evidence in everyday medical practice. I'm thinking might do a little series of blog posts on this subject. 

So why are so many people afraid of giving advice? One word: thalidomide. You'll all probably know what went wrong in the thalidomide disaster- a drug that was commonly used for morning sickness which unfortunately causes limb malformations when used at the time of pregnancy when women get morning sickness. That, coupled with the lack of post-marketing regulation ad monitoring back in those days- led to a perfect teratogenic storm, with obvious impacts for those affected, but which also shook the medical profession to its core. The good thing is that it forced a rethink in medicines regulation and triggered interest in pharmacovigilance, with schemes like pregnancy registries, teratology information services, and the MHRA's Yellow Card scheme being introduced in its wake. It's also left a lasting, deep-rooted fear of the potential of medicines to cause harm in pregnancy in pharmaceutical companies, patients and medical professionals alike. 

Why is this fear a problem? Well, primarily because some pregnant women do still get ill, and do still need medical treatment. I've lost count of the number of times I've heard a doctor say to me "my patient has (insert life threatening disease) and she's found out she's pregnant, so I've stopped all of her meds. What harm will that have caused?" And my first response is "how is your patient?", along with having to try very hard to stop myself saying "what do you think is going to be more of a teratogen, whatever the drug is, or having a patient who is dead?!" Such decisions to stop treatments are often done without consulting any evidence first, and there is a clear potential for harm to patients and their pregnancies in such situations. 

Another harm is that, in the event of something going wrong in a pregnancy, there's a tendency to way something to blame. I think it must be truly awful to have to think that because you have taken medicines, it's all your fault, especially as in many, many cases there will be no clear causality. 

I think we can pretty safely say that, because of the things that have been put in place since thalidomide, that another similar disaster won't happen again. There are drugs which can undoubtedly cause harm in pregnancies but as with all things in medicine, we have to consider a benefit vs risk balance. And we have to take into account what evidence we have access to, think about its limitations, and apply it to each individual situation. Just stopping all medicines because a patient is pregnant is not going to be the least risky option in most cases (and indeed in all those aforementioned cases that I've discussed, the drugs the patient has been on have turned out to have some pretty reassuring data sets). I even know of a dr who panicked so much when he found out his patient was pregnant that he put a note through her door telling her to have an abortion because he had prescribe her some drugs which turned out to have a pretty robust safety record in pregnancy. Imagine the emotional harm this sort of thing could cause. 

I don't want to end up writing a hugely long blog post so I'm going to end this one here as I'll cover some more aspects in future posts.