Becky McCall

September 18, 2017

LISBON, PORTUGAL — Continuous glucose monitoring (CGM) of blood sugar levels during pregnancy improves birth outcomes compared with traditional finger-prick tests in women with type 1 diabetes, according to results of the first trial to show benefit of CGM on health outcomes beyond glucose control.

Jointly presenting results of the randomized controlled CONCEPTT trial were the two study leads: Denice Feig, MD, from University of Toronto and Sinai Health System, Ontario, and Helen Murphy, from the University of East Anglia, United Kingdom, here at the European Association for the Study of Diabetes (EASD) 2017 Annual Meeting.

Results of the study were published simultaneously in the Lancet and prompted the researchers to call for all pregnant women with type 1 diabetes to be offered CGM to help improve outcomes for newborns.

"Use of CGM in pregnant mothers with type 1 diabetes is associated with improved neonatal health outcomes attributed to lower exposure to hyperglycemia," said Dr Murphy.

"Pregnant women using CGM had lower HbA1c — the primary outcome — but, more important, better day-to-day glucose levels as measured by CGM, resulting in 100 minutes more time per day in the recommended target range [3.5–7.8mmol/L]," she highlighted.

She added that the "CGM-related effects were comparable among [insulin]-pump and multiple-daily-injection (MDI) users.

However, in a second cohort of women with type 1 diabetes — those planning pregnancy — "no consistent benefits of CGM were seen."

Regarding neonatal outcomes — a secondary outcome among the pregnant cohort — the risk of large-for-gestational-age babies was "lowered by approximately half" among those using CGM, said Dr Murphy. Neonates of CGM mothers were also less likely to need intensive care of 24 hours or more, had 1 day shorter hospital stay, and were less likely to need intravenous dextrose for neonatal hypoglycemia, all of which have "potential for cost savings," she noted.

Commenting on the data in a press release, Daghni Rajasingam, MD, consultant obstetrician at Guys and St Thomas' NHS Foundation Trust, London, and spokesperson for the Royal College of Obstetricians and Gynaecologists, London, United Kingdom, said it is important to appreciate that previous study results of CGM in pregnancy have been conflicting.

While "this latest study does not show…significant benefits for women planning a pregnancy," the improved health outcomes for the infant include "being born a normal size, fewer admissions to neonatal intensive care [of >24 hours duration], and less time spent in the hospital."

Nevertheless, "More research will be needed…for example, by assessing the real-time device in a larger sample size" of both pregnant women and those planning pregnancy, she said.

CGM vs Finger-Prick Testing

Use of a CGM device that provides 288 glucose recordings per day allows women to rapidly respond to changes in blood glucose levels; by comparison, the so-called finger-prick method of glucose monitoring is carried out four to eight times a day, on average.

Studies in adults with type 1 diabetes have shown reduction in HbA1c and exposure time to hypoglycemia when using CGM, "but very few data exist in pregnant women," said Dr Feig, explaining the motivation for the study. With this in mind, the aim of CONCEPTT was to assess the effectiveness of CGM on glycemic control in women with type 1 diabetes who were pregnant or planning pregnancy.

The researchers ran two trials in parallel for pregnant participants (n = 215) and for participants planning pregnancy (n = 110). In both trials, participants were randomly assigned to either CGM in addition to capillary glucose monitoring or capillary glucose monitoring alone. The primary outcome was change in HbA1c from randomization to 34 weeks' gestation in pregnant women and to 24 weeks or conception in women planning pregnancy.

Results presented here focused on the group who were already pregnant.

The multinational, open-label study was carried out across 31 hospitals in Canada, England, Ireland, Italy, Scotland, Spain, and the United States. Women managed their diabetes with insulin pumps (n = 108) or MDI (n = 107), but had suboptimal control (HbA1c 6.5–10% in pregnancy; they had to be in their first trimester of a singleton pregnancy).

Half were randomly allocated to use the CGM device, and half to use the traditional finger-stick blood glucose monitoring method. The CGM was worn for approximately 24 weeks (from 10–12 weeks until the end of their pregnancy).

Women were taught how to use CGM, how to change their insulin, and how to take a capillary glucose (finger-prick) test. Study visits were every 4 weeks (weeks 12 to 36). Women in the control group did at least seven finger-stick tests a day.

CGM Led to More Time in Target HbA1c Range

Dr Murphy presented the results of the intention-to-treat analysis. The primary outcome was change in HbA1c from randomization to 34 weeks' gestation in pregnant women, with the sample size calculation intended to detect a between-group difference of 0.5%,

The researchers found a small difference in HbA1c in pregnant women using CGM, mean difference −0.19%, vs those in the control group using finger-prick tests (= .0207), but this meant the primary end point wasn't technically met.

CGM users spent more time in the target HbA1c range (68% vs 61% of controls, = .0034); and less time in hyperglycemia (>7.8 mmol/L) (27% vs 32%, = .0279).

But severe hypoglycemia episodes were comparable between the two groups (18 among CGM users and 21 in controls) as was time spent in hypoglycemia (3% vs 4%, = .10).

And women in the CGM group had significantly more unscheduled contact than the control group (< .0001), such as by email or phone, "for example, to say they'd run out of supply or [to report] calibration problems, with some issues being more serious," reported Dr Murphy.

And in terms of frustration, 80.6% in the CGM arm compared with 12.5% in the finger-prick group encountered problems (eg, sensor not inserted properly).

Around 70% of participants in the already-pregnant group and 77% in the planning-pregnancy group used CGM for more than 75% of the time.

Writing in an accompanying editorial in the Lancet, Satish K Garg, MD, and Sarit Polsky, MD, from the Barbara Davis Center for Diabetes, University of Colorado Denver, Aurora, note that, importantly, the CGM group had more visits with the provider, "which might have contributed to the small improvement in HbA1c," and that "the biggest drawback" of the data are the lack of information on frequency of self-monitoring of blood glucose and how it relates to improvements in HbA1c in the control group.

They highlight the fact that "the primary end point of a between-group difference in HbA1c of 0.5% was not achieved in this trial. [And] it was discouraging to see no improvements in severe hypoglycemia in this high-risk population."

Neonatal Adverse Outcomes Halved with CGM

Dr Feig presented the secondary obstetric and neonatal outcomes data.

"There were very few serious adverse pregnancy outcomes and no significant differences between groups," she said. Five women in CGM vs four in controls had a miscarriage; no women in the CGM group had a stillbirth compared with one in the control arm. Similarly there were no terminations in the CGM group and one among controls. There were two cases of a congenital anomaly in the CGM group compared with three in the control arm.

There was also no significant difference between groups in terms of overall hypertensive disorders or in caesarean-section rates. "From 16 to 34 weeks there was a trend toward greater weight gain in the controls [9.7 kg, vs 8.9 kg in CGM group, respectively]. Both groups delivered around 37.4 weeks," noted Dr Feig.

The key findings relating to neonatal outcomes were: large-for-gestational-age babies (>90th centile) were born in 53% of CGM women vs 69% in control women, with an odds ratio (OR) of 0.51 (= .0210).

Neonatal hypoglycemia that required intravenous treatment occurred in 15% of babies born to CGM women vs 28% of those born to the control group (OR, 0.45; = .025); and 27% of babies born to those in the CGM group required neonatal intensive care of over 24 hours vs 43% of control-group babies (OR, 0.48; = .0157).  

"The odds ratios of having the main neonatal events were reduced by approximately 50% between groups," summarized Dr Feig.

"Of note, one needed to treat only six women with CGM to prevent one episode of large-for-gestational-age, eight women to prevent one event of neonatal hypoglycemia, and six to prevent an NICU admission of over 24 hours."

Would I Use CGM for My Pregnant Patients With T1D Tomorrow?

One session attendant, Sally Marshall, MD, of Newcastle-upon-Tyne Hospitals NHS Foundation Trust, United Kingdom, pointed out that this is a "great study" but "looking at the neonatal hypoglycemia end point, we need to see much more data on the control of glucose during labor and delivery."

For example, "did women use CGM at this time?" she wondered, and were the methods of glucose control used during labor and delivery similar across centers, or was there any imbalance? "This could have a real impact on outcomes."

Providing formal comment on the CONCEPTT results overall in Lisbon, Elisabeth Reinhardt Mathiesen, MD, chief physician for managing pregnant women with diabetes at the Rigshospitalet University Hospital in Copenhagen, Denmark, congratulated the researchers on a "well-performed study," thanking them for providing evidence-based data to consider CGM "in selected pregnant women with type 1 diabetes" and on being "brave enough to perform a randomized controlled trial in pregnancy."

She stressed that her main issue is the affordability of CGM: "The cost in around 20 pregnant women is approximately the same as the salary of one nurse."

Nevertheless, she speculated on whether she would offer CGM to all pregnant women in her clinic tomorrow: "Many women will ask for it, and this study supports its use, and my gut feeling is that CGM is the future."

Dr Murphy, Dr Feig, and study coinvestigators report grants from the Juvenile Diabetes Research Foundation during the conduct of the study. Dr Murphy sits on the Medtronic European Scientific Advisory Board. Dr Feig has no additional relevant financial relationships. Disclosures for the coauthors are listed in the paper. Dr Garg has received advisory board consulting fees from Medtronic, Roche, Merck, Lexicon, Novo Nordisk, Sanofi, Mannkind, Johnson & Johnson, T1D Exchange (JAEB Center), and Eli Lilly and has received grant funding from Eli Lilly, Novo Nordisk, Merck, Lexicon, Mannkind, Medtronic, Dario, T1D Exchange, DexCom, and Sanofi; these consulting fees and research grants are for insulin treatment, diabetes device use, and other therapies for type 1 diabetes and are all paid to the University of Colorado, Denver. Dr Polsky has received research funding from DexCom for diabetes device use in patients with diabetes paid to the University of Colorado, Denver. Dr Mathiesen has no relevant financial relationships.

Lancet. Published online September 15, 20107. Abstract, Editorial



Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.