[Automatically generated] From the JAMA Network, this is JAMA Clinical Reviews, interviews and ideas about innovations in medicine, science, and clinical practice. Welcome and thank you for joining us for this JAMA Clinical Review podcast. I'm Dr. Tracy Lieu, Associate Editor at JAMA, a pediatrician and Director of the Division of Research Kaiser Permanente Northern California. Today we'll be discussing the topic of metformin for diabetes and pregnancy. We have two valuable new randomized trials on this topic, one published on October 3 by Fidelma Dunne and colleagues on gestational diabetes, and another published today by Kim Boggess and colleagues on type 2 diabetes in pregnancy. Today JAMA also is publishing an insightful commentary titled Metformin for Diabetes in Pregnancy. Are we closer to defining its role? And I'm delighted to welcome this commentary's author Dr. Denise Feig, she's an internationally recognized expert on this topic and a professor of medicine at the University of Toronto with appointments in endocrinology, obstetrics and gynecology, and health policy. She is head of the Diabetes and Endocrine and Pregnancy Program at Mount Sinai Hospital and past chair of the Diabetes and Pregnancy Interest Group for the American Diabetes Association. Welcome, Dr. Feig. Thank you so much. Let's start with some context. For our listeners who aren't specialists in this area, can you tell us why gestational diabetes is so important? Well, firstly, gestational diabetes is fairly common, occurring in between 7 and up to 18 percent of women, depending on the definition and population. So it is fairly common and can be associated with adverse pregnancy outcomes, including preeclampsia, c section rates for mothers, macrosomia, birth injury, hypoglycemia, and ICU admission in the neonates. And it's also associated with long term risks in both mother and child. Your commentary mentions the first randomized trial of metformin in gestational diabetes that came out in 2008. Can you tell us what they found and how clinical practice has changed since then? Sure. So the MIG trial was the first and largest trial of metformin compared to insulin in 750 women whose blood sugars were not well controlled after lifestyle intervention. The MIG trial found no significant difference in a composite of neonatal outcomes, and therefore was metformin was deemed safe to use, and women preferred the use of metformin over insulin. Since the MIG trial, many countries now use metformin as first line treatment over insulin for gestational diabetes, including the UK and much of Europe. This is not the case in the U. S., where metformin is not recommended as first line treatment. So the randomized trial by Fidelma Dunne and colleagues in JAMA October 3rd, called the EMERGE trial, evaluated patients with gestational diabetes and whether starting metformin earlier had benefits compared with usual care. Can you describe what usual care was and what starting metformin earlier meant? Yes, so starting metformin earlier meant starting metformin when the diagnosis of gestational diabetes was made and before failure of lifestyle modification compared with usual care which was starting insulin once lifestyle intervention failed or blood sugars remained above target. Yeah, and then what were the trial's key findings? The primary outcome was a composite of insulin initiation or fasting glucose above 92 mg per deciliter at 32 or 38 weeks gestation. While they didn't find a significant difference in the composite, they did find that fewer women had a fasting glucose above 92 mg per deciliter or acquired insulin at 38 weeks. Mean fasting glucose was significantly lower both at 32 and 38 weeks as was hemoglobin A1c, and maternal weight gain was also significantly reduced in the early metformin group. They found lower rates of insulin initiation in the metformin group, not surprisingly, and they found a longer time to insulin initiation in those who required insulin. Importantly, there were significantly fewer LGA or large for gestational age infants in the metformin group, with no increase in small for gestational age infants. Mm hmm. So, what do these findings mean for clinicians who are treating individuals with gestational diabetes? Well, I think it tells us that using this early metformin strategy is safe, and it does show benefits which are similar to the other trials where they used metformin following lifestyle modification failure, although with the added benefit of improved glycemic control, which you didn't always see that in the other trials. So, it sounds like a strategy to consider strongly. And then, you know, kind of turning our attention to individuals with type 2 diabetes in pregnancy. You personally led a very large trial of metformin called the MiTy trial. Could you review your trial's key findings with us? Yes, of course. So, in the MiTy trial, we set out to determine if the use of metformin added to insulin in those with type 2 diabetes would be beneficial to their neonates. Women were randomized to receive either metformin, one gram twice a day, or placebo added to insulin. And we found no difference in the composite serious neonatal outcome, but we found several other benefits, including better glycemic control in the metformin group, reduced maternal weight gain, reduced c section, lower insulin requirements. On average, they required 45 units of insulin less per day, and the infants had lower birth weights, lower birth weight Z-scores, less extremely large infants, and less adiposity. Now, the randomized trial by Kim Boggess and colleagues that was published in JAMA Today called MOMPOD addressed a similar question as your earlier MiTy trial in pregnant persons with type 2 diabetes. What does this new trial tell us? What was similar or different from the findings of the earlier trial? Yes, MOMPOD was a very similar trial and similar to MiTy, the investigators did find a significant difference between the groups. In terms of better glycemic control in the metformin group and a significant reduction in large for gestational age. reduced birth weight, reduced birth weight Z-score. Unlike MiTy, they did not find a reduction in cesarean sections, in insulin doses, in pregnancy weight gain, and there was no difference in neonatal fat mass. Mind you, they measured pregnancy weight gain as above the IOM criteria. They didn't give us absolute weight gain. So there have been a number of studies now, and yet there's still unanswered questions out there, and your commentary highlights questions where additional research is needed. What do you see as the highest priority questions for future research in this area? In the EMERGE trial, while it answers the question, what are the benefits of metformin when initiated at the time of GDM diagnosis compared to usual care with insulin, one cannot determine whether giving metformin early is better than giving metformin only after lifestyle modifications fail, as is standard practice in several countries. It would also be interesting to know if these differences in glycemic control translate into neonatal benefits in the larger study. So more research is needed to clarify these points. Interesting. Yeah. For women with type 2, the question is why were these differences seen between MiTy and MOMPOD? Was it the difference in population, the pattern of practice regarding cesarean sections, or compliance with the medication? Some of these questions may be answered by an individual participant meta analysis, which we plan to do using both MiTy and MOMPOD data, or some may be answered by future real world data. Which will be very interesting. Well, thank you for summarizing all of these ambitious studies for us. Do we yet know whether metformin in pregnancy has long term effects on the offspring later in childhood? I think the most pressing question for both populations is the question about long term effects of metformin in in utero exposed offspring. We know that children of both populations have an increased risk of obesity and diabetes themselves. And we need to know if metformin will increase or decrease this risk, or be neutral, and if the risk is the same in all populations. To date, the evidence is conflicting and we definitely need more data that looks at children post puberty and into early adulthood to better help decision making in pregnancy. We published two year data now from MiTy Kids that showed no difference, which is reassuring. And we plan to follow the MiTy Tykes kids up to 5 to 11 years. And I believe that the EMERGE and MOMPOD trials are planning to follow their Children as well. And I think that is most important question really. Yeah, so some very long term follow up studies it sounds like. Do you have an overall take home message for our clinical listeners? What does this lead one to recommend for clinical practice? The take home message is, in the short term, the plurality of the data supports a role for metformin in women with gestational diabetes and type 2 diabetes during pregnancy. especially women with type two who require such high doses of insulin and often have very large babies. But in both populations, more research is needed to assess the long term effects. Well, Dr. Feig, thank you so much for your really thoughtful commentary for all the research you're leading to optimize health outcomes for persons with diabetes and pregnancy and their infants. Thank you so much for your questions. I appreciate it. I'm Dr. Tracy Lieu, and I've been speaking with Dr. Denise Feig about metformin for diabetes and pregnancy. You can find a link to the paper in this episode's description. To follow this and other JAMA Network podcasts, please visit us online at jamanetworkaudio. com or search for JAMA Network wherever you get your podcasts. This episode was produced by Daniel Morrow at the JAMA Network. Thanks for listening.