Irish diabetes in pregnancy program takes control of gestational diabetes
Click Here to Manage Email Alerts
Key takeaways:
- The Atlantic Diabetes in Pregnancy program has improved pregnancy and neonatal outcomes since 2005 with more research ongoing.
- The program led to better screening, more collaboration and a model care document.
ORLANDO — Since 2005, an Irish diabetes in pregnancy program has steadily improved pregnancy and neonatal outcomes through extensive research and development of services for care before, during and after pregnancy.
Fidelma Dunne, MD, PhD, FRCP, FRCPI, professor at the School of Medicine at National University of Ireland and a consultant endocrinologist at Galway University Hospitals group, told attendees about the program’s history and success during the Norbert Freinkel Award Lecture at American Diabetes Association Scientific Sessions.
In 2005, Dunne took over the diabetes in pregnancy service at National University of Galway with plans to extend it in all of Ireland. At the time, Dunne said, there was poor knowledge of diabetes in pregnancy, no combined clinics, no electronic data collection and no key care indicators. In addition, there was no information on gestational diabetes prevalence of outcomes; a wide variety of gestational diabetes screening practices, protocols and guidelines; and no information on pregnancy outcomes for women with type 1 or type 2 diabetes. To improve the landscape, the Atlantic Diabetes in Pregnancy (DIP) group was created.
“We wanted to establish a clinical and research network, develop an electronic and patient collection tool, and we looked at a company called Hicom in the U.K., who had designed the Diamond collection tool. We wrote the prepregnancy, pregnancy and post-pregnancy fields for Diamond and started looking at collecting data in real world,” Dunne said during a session at American Diabetes Association Scientific Sessions. “We wanted to define the maternal and neonatal outcomes for ongoing pregestational diabetes, the prevalence of gestational diabetes and its outcomes and a reliable prevalence for the ongoing glucose abnormalities and people post-gestational diabetes.”
Prepregnancy care
To improve diabetes in pregnancy outcomes, the first step was to understand the current landscape, according to Dunne. To accomplish this, a study was conducted at five antenatal centers in Ireland.
In this study, researchers observed twice the likelihood of congenital malformation, three times the likelihood of perinatal mortality and five times the likelihood of stillbirth for women with pregestational diabetes. In addition, 50% of infants of mothers with pregestational diabetes were born large for gestational age. Of the mothers, 28% had prepregnancy care, 43% recieved folic acid and 50% had very poor glucose levels and control at the start of pregnancy.
Atlantic DIP developed a prepregnancy care leaflet, educated patients on what to expect during pregnancy and created clinician guidelines for diabetes in pregnancy. However, the main problem was that women were not receiving prepregnancy care, according to Dunne. This led to the rollout of Atlantic DIP’s prepregnancy care program.
A single intervention study evaluated the program for pregnancy and neonatal outcomes and demonstrated lower HbA1c for women who attended vs. did not attend prepregnancy care (P < .001). The prepregnancy care group was more likely not to smoke, to take folic acid, to not require teratogenic medications, and to have a lower BMI and better glycemic control at the start of pregnancy vs. non-attenders. Researchers observed significant reductions in serious adverse neonatal outcomes, congenital malformations and neonatal ICU (NICU) admissions for women who attended vs. did not attend prepregnancy care.
Two studies demonstrated that Atlantic DIP efforts doubled the rate of acceptance and uptake of prepregnancy care, maintained folic acid at 5 mg and improved glycemic control in the first, second and third trimesters. These results translated to significant reductions in stillbirth, perinatal and congenital malformation rates.
Gestational diabetes, glucose abnormalities
At the time, gestational diabetes prevalence and outcomes were unknown, according to Dunne. To establish gestational diabetes prevalence, researchers conducted a study with women from all five antenatal centers who were invited to participate in screening with a 75 g glucose tolerance test at 24 to 28 weeks’ gestation.
Overall, 75% of women consented to screening, but 30% did not attend. Gestational diabetes prevalence was 12.4% in the screened population. Those with gestational diabetes had higher likelihood of hypertension, preeclampsia and polyhydramnios and lower likelihood of normal delivery. For neonatal outcomes, researchers observed a higher likelihood of preterm birth with about 25% of infants born large for gestational age and 25% requiring NICU care.
“We knew from this that we had a problem in our region with this condition, and that we really needed to address screening in a good way to attract as many people as possible into the program,” Dunne said. “We were interested then in the 30% of women who didn’t attend for a screening, and we were able to geocode these women by their addresses. For every 10 km the ladies were required to travel to a screening site, screening was reduced by 2%. So, if you lived 100 km from your screening hospital, you were 20% less likely to come for screening.”
In a randomized trial, women were randomly assigned to general practice (n = 391) or hospital screening (n = 390). Fifty-two percent of women in the general practice group attended screening at that location while 32.5% attended hospital screening and 14.8% had no screening. Conversely, 89% of women in the hospital group attended screening at a hospital, 2% were screened at a general practice and 8% had no screening.
A cost-effectiveness analysis found that universal screening would be cost-effective, and hospital screening was the most cost-effective.
Dunne and colleagues applied risk factor-based screening guidelines at the time to the universally screened population. In this, 681 women were diagnosed with gestational diabetes compared with 521 women when applying national guidelines.
“This told me that the risk factor-based screening had problems and maybe we should look at expanding our screening to a more universal system,” Dunne said.
Atlantic DIP then offered antenatal and postnatal screening to universally screened women with and without gestational diabetes. At the time, ongoing glucose abnormalities were observed for 15.6% of women with gestational diabetes and 4.8% of women without, according to Dunne.
“This taught us that we needed to follow up these women effectively, and we put in place an administrator for the region and could maintain postpartum follow-up rates for screening of 75%,” Dunne said.
At 5 years postpartum, ongoing glucose abnormalities were observed for 25.9% of the same cohort with gestational diabetes and 3.6% of women without.
A clinical study found that 90% of women with abnormal glucose tolerance could be identified using cutoffs of 5.7% for HbA1c and 3.6 mmol/L for fasting glucose. This would reduce the need for glucose tolerance testing by 66%, Dunne noted.
In a study of screened Aboriginal women, glucokinase monogenic diabetes (GCK-MODY) was present in 0.1%, increased to 0.9% in women with gestational diabetes and went up to 2.56% for women with fasting glucose greater than 5.5 mmol/L. Researchers also observed higher fasting glucose and lower BMI for women with GCK-MODY and gestational diabetes vs. women with just gestational diabetes.
Based on these data, Atlantic DIP offers GCK-MODY testing to women with a BMI less than 25 kg/m2 and a fasting glucose greater than 5.5 mmol/L.
Gestational weight gain
A lot of focus was around weight gain in pregnancy, according to Dunne, which was poor for Atlantic DIP. A previous study demonstrated that 57% of women had excessive weight gain compared with National Academy of Medicine categories.
In addition, women with excessive gestational weight gain had higher odds of gestational hypertension (adjusted OR = 1.72; 95% CI, 1.04-2.85; P = .04), macrosomia (aOR = 2.17; 95% CI, 1.32-3.55; P < .01) and large for gestational age infants (aOR = 2.01; 95% CI, 1.24-3.25; P = .01). Odds of gestational hypertension (aOR = 1.62; 95% CI, 0.79-3.32; P = .01), macrosomia (aOR = 5.63; 95% CI, 2.16-14.69; P < .001) and large for gestational age infants (aOR = 2.8; 95% CI, 1.23-6.38; P = .01) were higher for women with gestational weight gain treated with insulin
“That got us to think about the treatments that we were offering ladies with gestational diabetes, and took us back again to our dataset, our collection tool, where we had all of this information,” Dunne said.
A retrospective cohort study demonstrated a 21% reduction in poor neonatal outcomes for women with gestational diabetes who received a diet and exercise intervention. In another study with 752 women with gestational diabetes treated with insulin, large and small for gestational age infants, birth weight, delivery time and more were comparable and not significantly different vs. women without gestational diabetes. However, researchers observed higher NICU admissions, neonatal hypoglycemia and cesarean birth for BMI categories ranging from less than 25 kg/m2 through more than 30 kg/m2 for women treated with insulin vs. women not treated with insulin (P < .01 for all).
In the EMERGE trial, 535 women with gestational diabetes were randomly assigned to metformin (n = 268) or placebo (n = 267) plus usual care. All women were followed until 12 weeks postpartum. The primary outcome was insulin initiation and fasting glucose combined.
Researchers observed adverse events among 24% of women in the metformin group and 4% in the placebo group. Thirteen women discontinued treatment due to adverse events. Overall, 92% of women achieved greater than 80% adherence to metformin. Researchers observed a 25% reduction in insulin initiation (P = .004) and lower fasting glucose at 32 (P = .03) and 38 weeks’ gestation (P < .001) in the metformin group.
Maternal weight gain was also lower (P = .003) and women were happier with their treatment (P = .04) in the metformin vs. placebo groups.
In the metformin group, birth weight was an average of 113 g lower with significant reductions in large for gestational age infants (P = .003) and macrosomia (P = .02).
“The trial did not show statistical superiority of early metformin vs. placebo for the combined composite primary outcome, but metformin had a positive impact on important prespecified secondary maternal and neonatal outcomes, including weight gain, glycemic control and infant size,” Dunne said.
The EMERGE Mother and Infant Follow Up Program is underway to assess outcomes at follow-up in this population.
Impact of Atlantic DIP
Now, the landscape of diabetes in pregnancy in Ireland is clearer, Dunne said. Atlantic DIP has launched a model care document, which will be used by all collaborating antenatal centers and was referred to by the minister for health for funding.
“We now have a robust retinal screening program, that happens as part of the national diabetes screening program, and these women are offered two screening sessions in their pregnancy, and if they have any worsening of the retinopathy, they get immediate attention in a retinal clinic,” Dunne said. “This is associated with almost 100% take up by the women.”
Dunne said the minister for health now allows women with prior gestational diabetes to enter the National Diabetes Prevention Program. This will help identify possible CV and type 2 diabetes, according to Dunne.
“We’ve also used all of the data that we did with the universal screening to propose to the minister, under the national screening advisory committee, that they might look at our application for consideration for universal screening,” Dunne said. “This is currently under consideration.”
Atlantic DIP has worked collaboratively with international communities to develop four core outcome sets and a patient-reported outcome set, which will allow for combined data moving forward and are used by the international community.
“For me, the process of clinical care delivery is really fundamental to a good pregnancy outcome,” Dunne said. “Methodical collection of routine data is very powerful. Teamwork is essential. One should revisit all data and ask new questions, connections and collaboration is important.”
References:
Bogdanet D, et al J Clin Endocrinol Metab. 2016;doi:10.1210/jc.2016-2911.
Egan AM, et al. J Clin Endocrinol Metab. 2014;doi:10.1210/jc.2013-2684.
Chakera AJ, et al. Diabetes Care. 2014;doi:10.2337/dc13-2248.
Egan AM, et al. J Clin Endocrinol Metab. 2016;doi:10.1210/jc.2015-4046.
Danyliv A, et al. Diabetologia. 2016;doi:10.1007/s00125-015-3824-0.
Dunne FP, et al. Diabetes Care. 2009;doi:10.2337dc09-1118.
Dunne FP, et al. JAMA. 2023;doi:10.1001/jama.2023.19869.
Noctor E, et al. Eur J Endocrinol. 2013;doi:10.1530/EJE-13-0491.
Oratile K, et al. J Clin Endocrinol Metab. 2015;doi:10.1210/jc.2015-3259.
O’Sullivan EP, et al. Ir Med J. 2012;105(5 Suppl):13-5.
Owens LA, et al. Diabetes Care. 2012;doi:10.102337/dc12-0120.
Owens LA, et al. J Clin Endocrinol Metab. 2016;doi:10.1210/jc.2015-3817.
Tierney M, et al. Diabetologia. 2015;doi:10.1007/s00125-015-3713-6.
Collapse
Dunne F, et al. Novel approaches to treatment and characterization of diabetes in pregnancy. Presented at: American Diabetes Association Scientific Sessions; June 21-24, 2024; Orlando.
Click Here to Manage Email Alerts