In many parts of the world, rates of cesarean delivery are too high, and growing. In the UK, for instance, about one-quarter of babies are born by cesarean. In the US, the rate is one-third, and in Brazil, it is one-half. The World Health Organization recommends that no more than 15% of deliveries be by cesarean.
The reasons behind these variations and growing numbers are complex, and beyond my scope here. Whatever the reason, more and more babies are entering the world surgically. We need to understand the potential consequences.
The high C-section rate is an emerging global health issue
Cesarean can be a medical necessity, or even an emergency. Decision-making can be fraught, with doctors and soon-to-be parents discussing risks and benefits during labor. Let’s put that aside.
Instead, let’s talk about the kind of information that clinicians and parents want to weigh in cooler moments, when cesarean is neither a medical necessity nor an emergency.
There has been much media attention of late to cesarean delivery on maternal request. It appears that this is rare in the US. One source estimates that 2.5% of births are cesareans requested by mothers. But it appears much more common in other parts of the world. For example, in southeast China, 20% of births were recorded as cesareans on maternal request in 2006. In some middle-income countries, skyrocketing elective cesareans have become a pressing public health matter.
In the US, there’s another setting that is more relevant. Many women who have had previous cesareans are able to go on to have vaginal births, from a medical perspective. But fewer than 10% of births to women who have had prior cesareans are vaginal deliveries. Repeat cesareans are a health concern worldwide, as more women receive a cesarean with their first birth.
The bottom line is that both of these settings – maternal request in low-risk pregnancy, and prior cesarean – offer a clear opportunity for a cool, deliberative weighing of risks.
Weighing risks and benefits for the child
In the short term – the hours and days surrounding birth – different modes of delivery bring different risks. For instance, parents might want to know that babies born by cesarean are more likely to need a brief stay in the newborn intensive care unit, while children born by vaginal delivery are more likely to have serious bruises under the scalp, requiring a short course of light therapy.
But when it comes to potential long-term health risks from C-section delivery, there is less information available. To date, discussions about delivery risks have tended to focus on long-term health problems with vaginal delivery. These include the very small risk of brain damage, injury to the the nerves of the arm (from shoulder dystocia), and other conditions.
This is changing, with growing evidence that delivery by cesarean may come with an increased chance of obesity, asthma and diabetes during childhood. In a piece that appears this week in the British Medical Journal, I discuss and evaluate this evidence, along with my colleague Jianmeng Liu of Peking University.
Evidence of long-term child health risks
The gold standard for evidence in medical science is the study where two treatments are compared in groups of volunteer patients whose treatment is determined by a coin flip. This is called a randomized study. Reviewing the published literature, we found six randomized studies comparing cesarean and vaginal delivery. All were conducted in scenarios where the optimal delivery choice was unclear (for example, delivery of twins and babies coming feet first, or “breech”).
These randomized studies mostly focused on the benefits and risks for mothers’ health. Only one of the studies followed children beyond the newborn period. The Term Breech Trial found that children in the cesarean group were in worse general health than those in the vaginal delivery group at two years of age. While the researchers didn’t go into great detail about the kinds of health problems that the children had, this finding was striking. Unfortunately, the children were not followed up later in life.
We also reviewed the summary evidence from over 50 nonrandomized studies that compared the health of groups of children delivered by the two methods. Those generally find a correlation between cesarean delivery and increased likelihood of childhood obesity, asthma and diabetes.
Using the summary data and extrapolating the findings to the US population, my coauthor and I estimated a childhood obesity rate of 15.8% for children delivered vaginally versus a 19.4% rate for children delivered by cesarean, a childhood asthma rate among children delivered vaginally of 7.9% versus a 9.5% rate for children delivered by cesarean, and a childhood type 1 diabetes rate of 1.79 cases per 1,000 children versus a rate of 2.13 per 1,000 for children delivered by cesarean. Again, these are just estimates, helping to translate the statistics into accessible numbers.
But importantly, correlation isn’t causation: women who have cesareans may be less healthy, and so their children might also be less healthy, regardless of how they are delivered. As it turns out, some correlational studies suffer from this limitation more than others. I believe that we can still get useful information about risks by focusing on the strongest of these correlational, nonrandomized studies.
As always, better data and further research are needed. I’ll have more to say about that later.
Why might cesarean be linked to long term health risks?
Stepping back, why might delivery matter? One theory says that it has to do with intestinal bacteria, which are important in food uptake and fighting infections. During vaginal birth, babies swallow maternal vaginal bacteria, and those bacteria are early colonizers of the babies’ intestines. Cesarean-born babies miss this exposure. It is possible that the resulting early differences in resident gut bacteria result in differences in health, later on.
Another theory focuses on the healthy, positive stress of labor and delivery, and the ways that stress “programs” a baby’s genes. According to this theory, the key programmers are levels of hormones such as oxytocin, cortisol and adrenalin. These give rise to so-called epigenetic changes that in turn determine the risk of disease later in life.
What do the experts say?
In medicine, expert advice is often delivered in the form of written clinical guidelines that summarize the evidence for clinicians and make recommendations for treatment. Recently, two influential groups – one in the US, and the other in the UK – issued guidelines for Cesarean Delivery on Maternal Request (CDMR). These guidance documents were pieces of advice to inform decisions in this very specific nonemergency situation.
After reviewing evidence, both groups concluded that vaginal delivery should be recommended for healthy women with low-risk pregnancies. In that group, requests for cesarean should be honored, after a women receives counseling about resources that are available, including pain control. Women requesting cesarean should also understand the risks and benefits of their choice. Strikingly, neither of the two documents mentioned the relatively new evidence on long-term risks to child health, such as obesity, asthma and diabetes.
Time to talk it over
It’s time for that evidence to enter the wider conversation. A good way to start would be to review and critically assess the evidence in updated guidelines. This would educate doctors and midwives, allowing them to present fuller information to their patients.
Make no mistake: the evidence linking cesarean to worse child health outcomes is far from airtight. We look forward to getting better evidence in future clinical trials, or cleaner correlational studies.
Again: cesarean is sometimes medically indicated, and is sometimes even an emergency. But in the US, and in many nations around the world, the high cesarean rate isn’t just a question of medical need. Patients, midwives and doctors are making choices, and those choices should be as informed as possible.
The evidence isn’t perfect. But then, it rarely is.
Jan Blustein received funding from NYU CTSA grant UL1TR000038 from the National Center for the Advancement of Translational Science (NCATS), NIH.