This story describes the results of a study that suggests that bariatric surgery is associated with both positive and negative outcomes in pregnancy. The coverage overall was adequate, but missed several opportunities to make the results more useful and meaningful. Strong points include interview comments from an independent expert who helps put the findings in perspective. The major weakness was the story’s failure to quantify harms or call appropriate attention to the potentially higher fetal/newborn death rate in the women who had surgery.
It is unclear how weight loss surgery affects the risk of pregnancy complications for women and their babies. The current study addresses both groups, though there is more information about outcomes among babies than among the pregnant mothers. The development of diabetes was decreased among pregnant women who had prior bariatric surgery, but the authors do not report on whether more serious pregnancy-related complications developed (e.g. pre-eclampsia, heart problems) or post-pregnancy outcomes (such as persistent diabetes). For many of these outcomes, a larger sample size would be needed. For the babies, the outcomes are more mixed and border on frightening. While there were fewer large babies, there were smaller babies, shorter gestation periods, a trend to more pre-term births, and a 1% increase (1.7% vs 0.7%) in deaths (stillbirths+neonatal). While the difference in death rates was not statistically significant, this finding provides cause for concern. These results don’t tell us what we should suggest for obese women of childbearing age who may be interested in a future pregnancy, but they sure do point to the need for further study.
The story could have addressed the cost of bariatric surgery in several ways. There is the cost of the surgery itself, and then there is the cost of pregnancy/delivery among obese women who have had bariatric surgery or not. The latter issue would seem most relevant to the story at hand, but might also be harder to measure. At least mentioning either issue, with an attempt at quantification, would have earned the story a satisfactory rating here.
The article generally does a good job of describing the benefits. In one instance, it mentions the absolute difference in outcomes between the intervention and control group for gestational diabetes. (“Only 2 percent of women who had weight-loss surgery developed gestational diabetes, compared to 7 percent of the other group, the researchers said.”). That’s very useful. But we’re concerned that the story never explains why readers should be worried about a “larger than normal” baby. It sounds, at most, mildly worrisome, and some readers might even hope for a baby that’s “larger than normal.” But a baby that’s significantly larger than normal can increase risks for a variety of delivery complications such as difficult labor and need for cesarean delivery, vaginal tearing, and bleeding after delivery. Although we wish the story had explained this, we think its overall treatment of benefits merits a satisfactory rating.
The story uses relative ratios to express the possible negative effects of bariatric surgery (e.g. “the weight-loss surgery group was twice as likely to give birth to babies considered small for their gestational age, and their pregnancies were also of slightly shorter duration.”), which misses critical information needed to understand the clinical or real world significance of these results. What was the overall percentage of babies considered small in each group, and was this a small or a large number? How worried should people be about this? Another concern is that the story didn’t emphasize or quantify the more important harmful outcomes. There was a higher rate of stillbirths and neonatal deaths in the surgery group (1.7% in the postsurgery group and 0.7% in the control group) which, while not quite statistically significant, provides cause for concern. We think readers would be most interested in this — more deaths among babies of women who had bariatric surgery. But the story mentions merely “a slight bump in the rate of stillbirths.”
The article describes many details of the study including the sample size, which is good. It also includes feedback from an expert who provides a restraining comment about the quality of the evidence. “I don’t think the evidence from this study is enough to say now that you should absolutely get this surgery so you have a better pregnancy outcome.” While that’s also good, we were looking for the story to specifically acknowledge some limitations of this observational study, which it never did. The original NEJM study had an extensive discussion of these limitations that the story could have borrowed from. Example: “…the observational design of the study makes it impossible to determine cause and effect. There may be residual confounding, because women who undergo surgery may have differed from women in the control cohort with respect to other factors not accounted for in the analyses.”
Obesity is highly prevalent, so it would be difficult to cross the line into disease mongering here. However, we’d note that the 179,000 bariatric surgeries mentioned in the story overstates the number performed in women of childbearing age in the U.S. In addition, the registry from which these data were collected had information on more than 600,000 births, from which the authors could find only 670 women who met their criteria.
The article quotes the independent source who wrote the editorial that came with the research article. This is a strong point of the story’s coverage.
The story did not mention alternative weight loss techniques (e.g. counseling on diet and exercise, medication, etc) for women prior to surgery, or the potential benefits for obese women of minimizing weight gain during pregnancy.
The article provides nice context for the scope and nature of bariatric surgery in the U.S.: “Nearly 179,000 obese people underwent weight-loss surgery in the United States in 2013, according to the American Society for Metabolic and Bariatric Surgery. While various techniques may be used, the surgery restricts the amount of food the stomach can hold and/or reduces the intestines’ absorption of calories and nutrients from food.”
The story quotes the study author about previous research in this area. “The positive effects of bariatric surgery on health outcomes — such as diabetes and cardiovascular disease — are reasonably well-studied, but less is known about the effects on pregnancy and [post-delivery] outcomes.” We’ll call that statement barely good enough for a satisfactory rating. The NEJM paper refers to a number of studies and reviews of studies examining this issue. It would have been nice to describe how the current study confirms or differs from those studies and why.
The article goes beyond any press release.