I Had a Complicated Pregnancy: Will It Happen Again?

An interview with Dr. Nathan Fox

It is natural for people to enter their first pregnancy with the hope and expectation that things will go smoothly. And often they do! But sometimes they do not. Pregnancy complications — extreme nausea, preeclampsia, prematurity, miscarriage — happen. When they do, in the aftermath, people often wonder: Will this happen again? If I had a preterm birth with my first pregnancy, am I likely to have another? If I miscarried, will it happen again? 

When I hear from women with these experiences, one overwhelming sense is that they feel they cannot get a good answer to these “will it happen again” questions. The answers are vague! Imprecise! They do not work for those of us who love data.

So I wanted to ask an expert, and I went to Dr. Nathan Fox, who is my favorite expert. Nate is a maternal fetal medicine (MFM) specialist in New York. He hosts an excellent podcast called Healthful Woman (website, direct link) all about pregnancy, women’s health, and sometimes COVID. I posed my questions, and our conversation is below. 

My summary is that he did not have all the answers, but I learned a lot about how to think about and approach the questions. 

Emily: Hi! I’m excited to talk about these issues. At a very basic level, I’d say the question is “Will this happen again?” It comes up in a lot of contexts — hyperemesis [very serious nausea and vomiting], preeclampsia, “incompetent” or “insufficient” cervix, miscarriage, stillbirth. And in this thread I’m hoping we’ll dive into a bit of detail about whether there are numerical answers in any of these contexts. But, big picture, I’d like to hear what you’d say in response to this concern.

Nate: I am super-excited as well. Thanks for having me! As a rule, for most pregnancy complications, if someone has one in their first pregnancy, there is a higher chance of having it again in the next pregnancy compared to baseline, but exactly how much higher depends on the complication, and the specifics of the complication in the first pregnancy. For example, if someone has a preterm birth in the first pregnancy, the chance of a preterm birth in the second pregnancy is 20%-50% (baseline risk is about 10%). The circumstances of the first preterm birth, and other risk factors, would help us determine if we think the risk in the next pregnancy is closer to 20% or 50%. This is something we try to do for women with a complication in the previous pregnancy: elicit the specifics of the complication to help determine if, and by how much, they are at increased risk in the next pregnancy. One notable exception is miscarriage. Unless someone has multiple miscarriages, or a miscarriage after 10 weeks, having an early miscarriage in the first pregnancy does not increase the risk of an early miscarriage in the next pregnancy.  

Emily: I’m going to come back to miscarriage, but I wanted to pause on preterm birth. There are a lot of reasons for prematurity [birth before 37 weeks]. My sense is that some of them that are obviously physical (insufficient cervix, for example) are much more likely to recur than ones that maybe have a more environmental cause (say, illness) or are unexplained. That makes sense to me structurally, but is it right?

Nate: Prematurity is one of the more complicated ones, because we usually do not know the cause, so it is harder to estimate the chance of recurrence (it is also why it is difficult to treat and prevent). Sometimes the cause is pretty clear — the first pregnancy was a triplet pregnancy, for example — but usually that is not the case. If someone goes into labor at 32 weeks and delivers, why did that happen? If it was due to an infection or inflammation in the uterus, it is not likely to recur, but if it was due to a malfunction in the complicated, and very poorly understood, timing mechanism of labor, it is highly likely to recur. However, we usually can’t differentiate between those two causes, because they present both with someone contracting and in labor. Cervical insufficiency is one of the most complicated of all the causes. If someone has an insufficient (formerly known as incompetent) cervix, they have a very high chance of recurrence, but it is very hard to make that diagnosis with precision, and more people get diagnosed with it than truly have it.  

In practice, when someone has a previous preterm birth, we try to examine the circumstances and whatever data we have from the preterm birth to narrow down the possible causes. That helps us better predict the likelihood of recurrence, as well as a course of monitoring and treatment for this pregnancy. But we usually end up with a range like 20%-50% and monitoring/treatment for several possible causes. 

Emily: So I think this gets in some ways to the heart of the question people ask me. The broad message is “The risk is elevated,” and (if they have you as a doctor!) there is likely a bit more information than that. But there is still a lot of uncertainty in the data. If I wanted to build a calculator where people could put in their risk factors and previous history and it would spit out a number … I think I’d have a hard time doing that, or the numbers it would spit out would be huge ranges.

If you agree, I wonder why this is. Is it an inherently unknowable thing? Is there just not enough data? Are we reluctant to attach numbers to it because they can never be “for sure” or “absolutely not,” so having a specific number feels less useful?

Nate: In theory, you could create a calculator, but the data are not specific enough to differentiate why there was a preterm birth, which is probably the most important variable.  Perhaps artificial intelligence could be used for this over large data sets, but I am unaware of any existing AI programs for this question. A calculator would probably include the gestational age at the preterm delivery, other OB history, certain medical problems, the mother’s age and weight, and whether she had treatments or not. Maternal race is complicated because we know it is associated with preterm birth, but there has been a rethinking of whether it is a good thing, or potentially a bad thing, to include race in calculators, as it might end up being more self-perpetuating than helpful. Ultimately, the calculator could spit out a risk, but it would really be an estimate, and to be fair it would have to give a statistical confidence range, which would likely be too wide to be useful. I think right now, for preterm birth, the best way for people to know what their risk is in a subsequent pregnancy is to find an MFM, or an OB-GYN with a lot of experience, and have a consultation to go over your specific history and get the expert’s take on the risk of recurrence, as well as a plan for the next pregnancy. I do these types of consultations all the time. But I agree it is not very helpful to simply get “the risk is elevated” with nothing else.

Emily: Fair enough. I will hold off on my calculator plan.

In the case of miscarriage, I read what you said at the top as that a single miscarriage is not very informative because, unfortunately, it’s so common. But more than one (or more than two) would entail an evaluation. Is that right? As a side note, this further reinforces my frustration that we keep miscarriage such a secret.

Nate: Since the vast majority of miscarriages are due to an abnormal chromosomal count or arrangement (aka karyotype) in the embryo, which is almost entirely a function of “luck” and the age of the mother, for the vast majority of women, having a miscarriage does not increase the risk of having another one. Certain situations raise the suspicion of a different cause and should be evaluated, although the conclusion may still be that there is no increased risk:  miscarriage of an embryo measuring 10 weeks or more, multiple early miscarriages (generally three or more, although two or more can also raise suspicion, especially if the miscarriage is tested and had normal chromosomes), and having medical problems. In these settings, we usually do a few tests to check for other potential causes: abnormalities of the uterus, certain clotting issues, certain medical issues, and some genetic issues. In truth, those workups are most often completely normal. 

I think many women keep miscarriages a secret out of traditional superstition, but that has been changing over the years. Ultimately, it is a very personal decision whether to share the story of a miscarriage with family, friends, and Facebook, and there is no right answer.  However, the story should absolutely be shared with your doctors, as this is something we definitely need to know to help you best.

Emily: Let me ask one other specific question, and then I have a more general one. 

We’ve been talking about recurrence of experiences related to the outcome of pregnancy. The other piece of this relates to experiences during pregnancy. I’m talking about things like weight gain, nausea, physical discomfort, etc. If I had hyperemesis in one pregnancy, for example, am I likely to have it again? More generally, should people pretty much imagine their experience in multiple pregnancies will be similar? 

Nate: That’s a really good question, but hard to answer with reliable numerical data for several reasons. First, many experiences are not easily quantifiable, so it is hard to study if they recur. Second, many of them are multifactorial, so it is hard to get independent recurrence risks. With those disclaimers out of the way, I will say that often women have similar experiences in each of their pregnancies, but not so often that we can know for sure.  For example, women with severe nausea and vomiting tend to have it every pregnancy, but it is not uncommon at all for a woman with severe nausea and vomiting in one pregnancy not to have it in the next pregnancy, and we have no idea why that is. To wit, in four large studies looking at the recurrence risk of hyperemesis gravidarum (the most severe form of nausea and vomiting), the recurrence risk was 15%, 20%, 24%, and 89%, respectively. How not useful is that?! Weight gain is usually similar between pregnancies, unless her diet or exercise patterns change significantly. Aches and pains tend to get worse with age (sadly, this is true for nonpregnant men such as myself as well) and with prior childbirth, so those often get worse in subsequent pregnancies.

To summarize, although there is some correlation from one pregnancy to another in these experiences, it is not strong enough or predictable enough to rely upon in either direction. In other words, all bets are off.

Emily: This is actually a good segue into my last question. For some of us — me for sure, and I’m guessing a lot of my readers and your patients — there is a desire to have more concrete answers to these questions. Your hyperemesis numbers above are a useful example: the range from 15% to 89% is basically the whole range. As you note, that’s not helpful. What it means as well is that when people ask about it, they get answers that feel vague. Yet we need to make decisions based on this vague information. Maybe I have to decide whether or not to get pregnant again, or what kind of interventions to undertake. In the face of this uncertainty, how do you counsel people to make decisions? What conversations should they be having with their doctors? 

Nate: You are correct — it can be frustrating to have vague answers, both on the face of it but also for practical reasons in regard to planning. I wish it were more clear. But in my experience, most pregnancy planning (and medicine in general) ends up being like that. We can have a sense of when someone has a relatively low risk of something or a somewhat increased risk, but it is typically hard to pin that down into a workable number. So, what to do? I think it is important for people to realize that for many of the risks, with advance planning, there might be ways to mitigate it or lessen the impact of it. 

Continuing with nausea and vomiting as our example, let’s say I am meeting with someone (not currently pregnant) with a history of hyperemesis gravidarum in a prior pregnancy to discuss her risk of hyperemesis in a future pregnancy. Based on the data, she has a 15%-89% chance of recurrence. That range is quite wide, but I might be able to give an educated guess which end of the range she may be in based on how many pregnancies she had before with hyperemesis, how she responded to treatment, how long it lasted, how severe it was, whether it runs in her family, whether she has other related risk factors, etc., etc. Those are not easily input into a calculator, but since I’ve taken care of a lot of women with hyperemesis, I likely have a pretty good sense of what to ask, and what follow-up questions to ask, to determine where she falls in that range. Also, we may be able to develop a monitoring and management plan for an upcoming pregnancy that could potentially reduce her risk, or at least reduce the severity of the hyperemesis. After that meeting, she will probably have a much better sense of what to expect and what we can do about it, and then she can decide if she wants to embark on another pregnancy. Not exactly a supercomputer, but it’s pretty good.

Here’s a good analogy: Let’s say I am a trial lawyer for the defense and I am about to cross-examine the key witness for the prosecution. Would I rather write up a list of questions to ask the witness and have my law student ask them, or ask them myself? Clearly the latter, because my knowledge and experience will give me the ability to pivot, ask follow-up questions, read the witness, pick up on subtle cues from the witness and the jury, and ask additional questions. The same is true for medicine. If I wanted to predict the outcome of a future pregnancy, I’d much rather have an hour with someone to hear their story and their history and probe into the details than have several data points to input into a calculator. 

I think the takeaway message is this: A prior pregnancy has some predictive value for a future pregnancy, but it is usually not precise enough to be useful as a stand-alone predictor. For a clearer picture of what to expect and what we can do about it, I highly recommend meeting with an experienced obstetrician or maternal fetal medicine specialist prior to the next pregnancy. These pre-pregnancy consultations are invaluable to the woman herself and to the doctor. We have plenty of time to go over the history, discuss options, order tests if needed, and think about what to do, with none of the pressure of already being pregnant. I usually block out a full hour for these consultations. It is a much more pleasant experience going into pregnancy with (a) reasonable expectations and (b) a plan! 

(If you don’t have easy access to one of these specialists, you can start by listening to my podcast! Also, nowadays in the world of virtual visits, access has opened up significantly.  Don’t feel limited by geography to have a consultation — almost always, all we need to do is speak to come to a plan.) 

Emily: One actual last question. Should healthy pregnant people get a booster shot, yes or no? [Author note: Of course I had to ask a COVID question…]

Nate: I think we don’t really know yet, as those recommendations are still being developed. But if they are going to be recommended for nonpregnant people of a similar age and risk, I presume they will be recommended for pregnant women as well. The issue is that we really don’t know for sure who exactly needs a booster: everyone? everyone with certain risk factors? no one? What if someone was vaccinated and got COVID — do they need a booster? Teens? It’s just not possible to have all of the nuances worked out without more studies, which take time. They also rely on assumptions that may or may not be true, like how well do antibody levels correlate with risk of getting COVID again, or risk of severe disease? Also, if someone is vaccinated, how risky is it to get COVID again (seems to mostly not be so risky) and, if so, who exactly “needs” to be vaccinated? 

The recommendations are being made on the fly, which isn’t actually that unique in the history of medicine, but this time it is all being done on CNN and Facebook! 

So my short answer today is: Maybe probably. How’s that for a hedge?    

Emily: Perfect note to end on.  


Did You Miss It?


Share

Leave a comment