Fetal measurements for guiding the medical management of women with diabetes in pregnancy to improve outcomes for mother and baby

What is the issue?

Diabetes during pregnancy affects between one and three in ten pregnant women (10% to 30%), and is associated with an increased risk of adverse outcomes. The baby may grow to a large size, leading to injury of the baby during birth, or the need for caesarean delivery. It is known that carefully controlling the mother's blood glucose levels during pregnancy reduces the risk of these and other adverse outcomes. All women with diabetes during pregnancy have their glucose levels closely monitored, and are treated using dietary changes, exercise or medications such as insulin or oral medication, even when their babies are not showing signs of being affected.

The most common sign of being affected by high glucose and insulin levels is a baby who is larger than expected, and so at risk of macrosomia (weighing more than 4000 g at birth).

Why is this important?

Only 14% to 22% of women with diabetes during pregnancy have macrosomic babies (babies weighing more than 4000 g at birth). At present, all women with diabetes during pregnancy, even those whose babies are not showing signs of being larger, are closely monitored. This costs time and money for both the women and the health services. If we are able to prove that targeting intensive monitoring and treatment to those women whose unborn babies are affected by overgrowth does not increase the risk of adverse outcomes, we may be able to save time, resources and anxiety.

What evidence did we find?

We searched for evidence on the 29th of January 2019 and identified three small randomised controlled trials (involving a total of 524 women) for inclusion in our review. The overall trial quality was low to moderate. The trials did not report the majority of outcomes of interest in this review, including outcomes relating to cost or use of resources.

Compared with monitoring the mothers' blood glucose levels alone, the addition of ultrasound may make little or no difference to the risk of having a caesarean birth (2 trials, 428 women, low-certainty evidence). Very low-certainty evidence means that we are unclear about the results relating to the risks of the mother having blood pressure disorders during pregnancy (2 trials, 325 women). The included trials did not report on the important maternal outcomes of low blood glucose, or development of type 2 diabetes.

Using ultrasound in addition to monitoring the mother's blood glucose levels may make little or no difference to the risk of the newborn baby having low blood glucose levels (3 trials, 524 women, low-certainty evidence). Very low-certainty evidence means that we are unclear about results relating to the risks of: having a baby that is large for gestational age (3 trials, 524 women); the baby's shoulders becoming entrapped in the birth canal (1 trial, 96 women); death or illness in the newborn baby (1 trial, 96 women); or the baby dying during pregnancy or birth (1 trial, 96 women).

What does this mean?

This review was based on limited evidence from three trials (involving 524 women). The trials did not report some important outcomes of interest to this review, and the majority of our secondary outcomes were also unreported. The certainty of the available evidence ranged from low to very low. There was insufficient evidence to evaluate the use of ultrasound (in addition to maternal blood glucose concentration values) to assist in guiding the medical management of GDM, and the effect on important short- and long-term outcomes for the mother or her baby, or the associated costs.

Large, randomised trials are needed. Such trials could consider important short- and long-term outcomes (as listed in this review) for the mother, her baby, and resource use.

Authors' conclusions: 

This review is based on evidence from three trials involving 524 women. The trials did not report some important outcomes of interest to this review, and the majority of our secondary outcomes were also unreported. The available evidence ranged from low- to very low-certainty, with downgrading decisions based on limitations in study design, imprecision and inconsistency.

There is insufficient evidence to evaluate the use of fetal biometry (in addition to maternal blood glucose concentration values) to assist in guiding the medical management of GDM, on either maternal or perinatal health outcomes, or the associated costs.

More research is required, ideally larger randomised studies which report the maternal and infant short- and long-term outcomes listed in this review, as well as those outcomes relating to financial and resource implications.

Read the full abstract...
Background: 

Gestational diabetes mellitus (GDM) is a common medical condition that complicates pregnancy and causes adverse maternal and fetal outcomes. At present, most treatment strategies focus on normalisation of maternal blood glucose values with use of diet, lifestyle modification, exercise, oral anti-hyperglycaemics and insulin. This has been shown to reduce the incidence of adverse outcomes, such as birth trauma and macrosomia. However, this involves intensive monitoring and treatment of all women with GDM. We propose that using medical imaging to identify pregnancies displaying signs of being affected by GDM could help to target management, allowing low-risk women to be spared excessive intervention, and facilitating better resource allocation.

Objectives: 

We wanted to address the following question: in women with gestational diabetes, does the use of fetal imaging plus maternal blood glucose concentration to indicate the need for medical management compared with glucose concentration alone reduce the risk of adverse perinatal outcomes?

Search strategy: 

We searched Cochrane Pregnancy and Childbirth’s Trials Register (29 January 2019), ClinicalTrials.gov, the World Health Organization International Clinical Trials Registry Platform (ICTRP) (both on 29 January 2019), and reference lists of retrieved studies.

Selection criteria: 

Randomised controlled trials, including those published in abstract form only. Studies using a cluster-randomised design and quasi-randomised controlled trials were both eligible for inclusion, but we didn't identify any. Cross-over trials were not eligible for inclusion in our review.

We included women carrying singleton pregnancies who were diagnosed with GDM, as defined by the trials' authors. The intervention of interest was the use of fetal biometry on imaging methods in addition to maternal glycaemic values for indicating the use of medical therapy for GDM. The control group was the use of maternal glycaemic values alone for indicating the use of such therapy.

Data collection and analysis: 

Two review authors independently assessed trials for inclusion and assessed risk of bias. Two review authors extracted data and checked them for accuracy.

Main results: 

Three randomised controlled trials met the inclusion criteria for our systematic review - the studies randomised a total of 524 women.

We assessed the three included studies as being at a low to moderate risk of bias; the nature of the intervention made it difficult to achieve blinding of participants and personnel and none of the trial reports contained information about methods of allocation concealment (and were therefore assessed as being at an unclear risk of selection bias).

In all studies, the intervention was the use of fetal biometry on ultrasound to identify fetuses displaying signs of fetal macrosomia, and the use of this information to indicate the use of medical anti-hyperglycaemic treatments. Those pregnancies were subject to more stringent blood glucose targets than those without signs of fetal macrosomia.

Maternal outcomes

The use of fetal biometry in addition to maternal blood glucose concentration (compared with maternal blood glucose concentration alone) may make little or no difference to the incidence of caesarean delivery (risk ratio (RR) 0.81, 95% confidence interval (CI) 0.59 to 1.10; 2 trials, 428 women; low-certainty evidence). We are unclear about the results for hypertensive disorders of pregnancy (RR 0.80, 95% CI 0.34 to 1.89; 2 trials, 325 women) due to very low-certainty evidence. The included trials did not report on development of type 2 diabetes in the mother or maternal hypoglycaemia.

Fetal and neonatal outcomes

The use of fetal biometry may make little or no difference to the incidence of neonatal hypoglycaemia (RR 0.90, 95% CI 0.57 to 1.42; 3 trials, 524 women; low-certainty evidence). Very low-certainty evidence means that we are unclear about the results for large-for-gestational age (RR 0.81, 95% CI 0.38 to 1.74; 3 trials, 524 women); shoulder dystocia (RR 0.33, 95% CI 0.01 to 7.98; 1 trial, 96 women); a composite measure of perinatal morbidity or mortality (RR 1.00, 95% CI 0.21 to 4.71; 1 study, 96 women); or perinatal mortality (RR 0.33, 95% CI 0.01 to 7.98; 1 trial, 96 women).