Q&A: Can maternal lipid levels predict congenital heart disease risk?
Click Here to Manage Email Alerts
Women with high blood lipid levels during early pregnancy faced higher risks of having children with congenital heart disease, according to a study published by Acta Obstetricia et Gynecologica Scandinavica.
Although the findings are noteworthy, further research is necessary to clarify which women and infants may be most at risk, a pair of clinicians from Cedars-Sinai Medical Center told Healio.
The case-control study involved mothers whose children had congenital heart disease (CHD; n = 230) and children who did not have CHD (n = 381) at the Obstetrics and Gynecology Hospital of Fudan University in Shanghai between August 2015 and December 2018.
Fasting blood was drawn from women who had their first prenatal examination at the hospital between 8 and 14 weeks of gestation. The biochemical examination included folic acid, vitamin B12, homocysteine (Hcy), triglyceride (TG), total cholesterol (TC), HDL cholesterol (HDL-C), LDL cholesterol (LDL-C), free fatty acid, apolipoprotein-A1 (Apo-A1), apolipoprotein-B (Apo-B), fasting blood glucose and HbA1c.
Critical CHD cases, in which defects caused death or intervention was required before the age of 28 days, and serious cases, in which defects required intervention before the age of 1 year, were combined as severe CHD.
Significant CHD, in which defects persisted beyond the age of 6 months but were not classified as critical or serious, and nonsignificant cases, in which defects were not physically appreciable and did not persist after the age of 6 months, were combined as mild CHD.
According to the study, 48 (20.9%) of the infants with CHD had severe cases, and 182 (79.1%) had mild cases, with 42 severe and 33 mild cases diagnosed prenatally.
Mothers of children in the case group had significantly higher levels of TG, Apo-A1 and Apo-B than the control group (all P < .05).
The researchers concluded that a maternal lipid profile of high TG (OR = 2.46; 95% CI, 1.62-3.73; P < .01), Apo-A1 (OR = 2.73; 95% CI 1.16-6.40; P = .02) and TC/HDL-C (OR = 2.1; 95% CI, 1.07-4.13; P = .03) levels in early pregnancy was positively associated with an increased risk for CHD in offspring, as each lipid biomarker was associated with an almost twofold increased risk for CHD independent of potential confounders such as age, parity, multiple birth, HbA1c and Hcy.
Specifically, the researchers found that TG, Apo-A1 and Apo-B were significantly elevated in the mild CHD group after Bonferroni correction (all P < .0167). But they did not observe any association between the lipid profile and severe CHD, even though they expected it to have an even stronger association than in the mild CHD group.
The researchers said that because the study only involved 48 severe cases, future studies are needed to investigate the lack of an association of lipid biomarkers and CHD in this group.
Additionally, the researchers said that multiple births, gestational diabetes and assisted conception are strong predictors of CHD in offspring. Twin birth was associated with a higher risk for CHD as well, with an even higher risk with monochorionic twins.
The researchers recommended additional investigations via large prospective studies to uncover whether lipid metabolism in early pregnancy may influence pregnancy outcomes, with mechanistic studies to establish causality.
To find out more, Healio spoke with Cedars-Sinai Smidt Heart Institute’s Ruchira Garg, MD, a pediatric cardiologist and director of congenital noninvasive cardiology, and Janet Wei, MD, FACC, a cardiologist in the Cedars-Sinai Barbra Streisand Women’s Heart Center. Gard and Wei were uninvolved in the research.
Healio: How prevalent is congenital heart disease?
Garg: The heart is the most common organ to be affected by a developmental error. The incidence is about 0.8% to 1% of all pregnancies.
Unfortunately, more than 50% of congenital heart defects, even in contemporary studies, are often missed. Any conditions that increase the probability of congenital heart disease above that background risk of 1%, such as maternal diabetes or family history of congenital heart disease, are things we ought to know so we can more proactively screen these pregnancies with a fetal echocardiogram.
That’s what's interesting about this study. Over the next 5 to 10 years, we're going to see more studies that show us additional risk factors. But we have to be very careful in how we interpret these kinds of studies because it’s hard to control for other things that are going on.
Healio: How would you evaluate the lipid profiles that the study found?
Wei: The lipid profile measurements that these researchers reported in their congenital heart disease group really aren’t that high. We are in an obesity epidemic, and we are caring for many women who have metabolic syndrome — they might be overweight, they might have prehypertension or there might be some insulin resistance.
Unfortunately, pregnancy heightens that adverse metabolic profile. The pregnant woman’s body shifts from an anabolic phase to a catabolic state. Triglycerides increase, LDL-C becomes more atherogenic and their insulin sensitivity decreases.
If we look at their table of the cholesterol numbers in the first trimester of pregnancy, both the cases and controls had low HDL, but neither group had significantly elevated cholesterol levels.
In fact, both groups had mean lipid levels within the normal range that we would see in the general population. During pregnancy, total cholesterol, LDL and triglycerides increase twofold to threefold. However, this happens mostly after the first trimester and all the way until delivery. It’s very interesting that the authors found first trimester triglycerides to be associated with a higher risk of CHD.
The CHD group was slightly older, had higher rates of gestational diabetes and had higher prevalence of assisted reproductive technology use, factors that can confound their findings. The authors adjusted for age and HgbA1c, but HgbA1c has poor sensitivity for gestational diabetes, and the authors did not adjust for gestational diabetes nor reproductive technology.
Healio: Is it surprising that even though these numbers aren’t that particularly high, they still seem to be associated with these very significant risks?
Garg: I would temper our interpretation of these results. Assisted reproductive technologies such as in vitro fertilization were five times higher in the population with congenital heart disease. The mothers also had higher BMI, and they had higher incidence of gestational and pre-gestational diabetes, suggesting that the case control group may not have been well-matched.
Additionally, there was no difference between groups in the incidence of “major” congenital heart disease. The study only found a small difference among the minor cases, and we always have to consider statistical error before we start endorsing study results.
Only when more researchers take such findings and try to reproduce the results can we be more confident that the differences are real. It’s great to be the first one out there, but there are many studies that have been published that can’t be reproduced. We should first reliably prove elevated lipid levels are a risk factor for congenital heart disease (or other adverse outcomes) before we can endorse their routine use in clinical care.
Healio: Was there anything else about this study that was surprising or significant?
Garg: Earlier I noted that more than 50% of congenital heart disease is not diagnosed prenatally, and that was absolutely the case in this study. Of some 40 patients with major congenital heart disease, six were missed, with very important congenital heart defects that needed intervention, three of them within the first month of life, and the other three potentially needing care later. Most patients had minor congenital heart disease, and about 80% of those were not identified prenatally.
Again, we’re talking about a greater than 50% nondiagnosis rate for congenital heart disease. This illustrates why it’s so important that we identify risk factors, why we do these fetal echoes and why our perinatology and obstetric colleagues should have a very high suspicion if they’re not seeing a completely normal heart on their screens.
Wei: I’m glad that they asked about assisted reproductive technology. In their case group, it only represented less than a fifth of the cases, but it is something that I think we should be investigating more. We need to study the link between infertility and future cardiovascular health.
Healio: You both seem to agree that more research is needed. How would you set up the next study?
Garg: I really think the foundation of new research is going to be registry-based. Until the last couple of years, we have really relied on small, single-center studies. And we have the ability with electronic medical records, with computer processing speed and technology, to start extracting patterns from larger populations.
All these sorts of factors like hemoglobin A1C and other parameters can be fed into these algorithms. The setup is important, because then you are generally curating the data that comes in, so it’s a little better filtered and vetted.
Wei: The study didn’t describe the prevalence of other potentially significant adverse pregnancy outcomes that might affect the results, such as preeclampsia. We need more in-depth phenotyping of these patients’ reproductive health and their cardiovascular profile. We also need large national registries.
Cardiovascular disease is the leading cause of death in pregnant and postpartum women in the U.S. The American College of Cardiology Cardio-Obstetrics Working Group is working to promote the science, education and advocacy of maternal cardiovascular care. We are happy to see more studies of pregnant women to advance the field.
Healio: What are the next steps that providers can take to help ensure more positive outcomes in treating pregnant women who may have some of these profiles or risk factors?
Wei: Routine screening. OB/GYNs and other providers of pregnant women should obtain a thorough personal and family history of cardiovascular disease and CVD risk factors, and ideally assess all women’s cardiometabolic profile pre-pregnancy to be able to counsel preventive strategies.
Although screening of gestational diabetes and blood pressure is routine, we still don’t really know what to do with high cholesterol levels and pregnancy. The FDA recently said they are going to remove the contraindication for statins for pregnant patients, which may be helpful for patients at high risk for myocardial infarction and stroke, but there’s still uncertainty regarding the treatment of hypercholesterolemia in pregnancy.
Garg: It is important to consider that we’ve avoided pregnant women as a study group almost entirely in medicine — probably since the 1970s, when some notable teratogenic drugs had such bad neonatal outcomes. Until recently, almost no drug has been endorsed or approved for use during pregnancy. By moving that needle so far into that conservative direction, we’ve done pregnant women a disservice.
I think this article highlights that if we put pregnant women into trials specifically geared toward them, we’re likely to learn a lot because we’ve been behind in this area.
References:
- Boyd HA, et al. Circulation. 2017;doi:10.1161/CIRCULATIONAHA.116.024600.
- Cao L, et al. Acta Obstet Gynecol Scand. 2021;doi:10.1111/aogs.14225.