Gestational diabetes mellitus (GDM) is by far the most common medical co-morbidity in pregnancy, with rates of 8–30 per cent reported across Australia. It’s easy to forget just how much of a ‘work in progress’ GDM is and feel overwhelmed by the huge numbers of affected women. Since the 1950s, when Pedersen proposed that maternal hyperglycaemia led to fetal hyperglycaemia and an exaggerated fetal response to insulin, the rationale for screening and treating GDM has centred on reducing the risk of fetal macrosomia and its obstetric complications. Recent data has highlighted the importance of GDM as a predictor of future cardiometabolic risks, both for affected women and their offspring1, thus providing new impetus for its identification and treatment.

The last ten years have been marked by controversy over the diagnostic criteria for GDM and subsequent treatment targets, with concerns of creeping over-medicalisation and an unnecessarily interventional approach for otherwise low-risk women. The revised International Association of Diabetes and Pregnancy Study Groups (IADPSG) diagnostic criteria (Table 1) have now been adopted by most centres. This is based on an odds ratio of 1.75 for adverse outcomes, compared to the rates seen at mean glucose levels of the HAPO3 cohort. It is a more evidence-based set of criteria than previously used. These levels are highly sensitive for detecting women at risk of adverse obstetric outcomes, but also detect a significant number of low-risk women, with substantial workforce implications.

Table 1. Diagnostic and interventional targets for GDM
BGL in mmol/L IADPSG2 diagnostic criteria 

75g OGTT

‘old’ ADIPS criteria 

75g OGTT

 ADIPS2 intervention targets  ACHOIS4 intervention targets
Fasting ≥5.1 ≥5.4 ≥5.0 ≤5.5
One hour ≥10.0 ≤7.4
Two hours ≥8.5 ≥8.0 ≤6.7 ≤7.0

The management of GDM is resource intensive, requiring a multidisciplinary approach. Optimising exercise and diet, with self-monitoring of blood glucose levels, is the mainstay of treatment, with the addition of pharmacological agents (metformin and/or insulin) required to optimise glycaemic control in roughly 50 per cent of affected women.

Exactly how low to go remains contentious. To date, there is a paucity of data available to substantiate the implementation of the current low targets outlined in the Australasian Diabetes in Pregnancy Society (ADIPS) position statement (Table 1). For many centres, the additional rate of intervention required is prohibitive5. We currently adopt a pragmatic approach, continuing to use the Australian Carbohydrate Intolerance Study in Pregnant Women (ACHOIS) targets, although admittedly, with an unofficial tendency to push lower once treatment has been commenced, until more data becomes available.

Beyond glycaemia

The benefit of screening and subsequently treating GDM extends beyond reducing pregnancy-associated complications. Recent data from the UK6 and Sweden7 support that a diagnosis of GDM entails a marked increase in risk of future cardiovascular disease and maternal type 2 diabetes, with rates up to 50 per cent at 20 years. While it is worth noting that in these cohorts, the diagnostic criteria for GDM was far less stringent than it is today (fasting BSL>6.0mmol/L), there is also a significant rise in population rates of obesity and type 2 diabetes. A diagnosis of GDM is a window of opportunity for women to receive education in key lifestyle interventions required to optimise metabolic health.

Children of pregnancies affected by GDM are more likely to be overweight or obese, show greater central adiposity, high blood pressure, insulin resistance and impaired glucose tolerance. These children are at increased risk of developing type 2 diabetes and cardiovascular disease in later life.1 Maternal dietary re-education and motivational counselling has been shown to reduce rates of GDM, limit weight gain and reduce hypertensive complications. This has potential to benefit subsequent pregnancies and future health in women, and by association their families, who are at a significantly higher risk of obesity, hypertension, and metabolic syndrome. Anecdotally, many women feel better when given the additional impetus to prioritise healthy diet and activity, thus there can be a silver lining to the diagnosis.

Is metformin a magic bullet?

Where lifestyle intervention is not enough, insulin remains the first-line treatment for GDM in Australia.8 However, ten years following the publication of the Metformin in Gestational Diabetes (MiG)9 study, documenting that metformin is an effective and safe treatment option for women with GDM, metformin is an increasingly used option that may have additional benefits. The long-term effects on offspring exposed to metformin are being carefully monitored, with published outcomes at two years reassuring.10 There is a particular role for metformin in women who are obese, reluctant to take insulin, or already on large doses of insulin. Mealtime metformin is markedly more effective than would be expected from its pharmacokinetic profile, generally well tolerated and, as seen in MiG, glycaemic targets are achieved much faster than with up-titration of insulin.

Metformin has been shown to have benefits other than the primary indication of insulin sensitisation and reduction of hyperglycaemia in the non-pregnant population. A reduction in overall mortality and bowel cancer risk make it a very compelling first-line agent in type 2 diabetes. It improves fertility in women with PCOS (sometimes unintentionally) and is associated with reduced miscarriage rates when used as part of fertility therapy. This is thought to relate to alterations in anti-angiogenic factors, which may also be associated with a reduction in risk of pre-eclampsia. While this has not yet been directly studied, a very recent meta-analysis comparing insulin to metformin for lowering glucose levels  in pregnancy,11 did show reductions in gestational weight gain and pre-eclampsia in the metformin group. While a formal randomised controlled trial is needed to confirm this, metformin is showing promise as a very useful treatment option in GDM.

Delivery planning

Timing and mode of delivery in women affected by GDM has seen a significant shift in the goalposts in recent years. Emerging data from paediatric literature emphasises the importance of fetal neurological development right up to term,12 with a move towards delivery at, or after 39 weeks, seen as best practice. This has led to a more individualised decision-making process, including closer monitoring of estimated fetal weight with ultrasound and enabling a less interventional approach to timing of delivery in women with GDM. At the same time, there has been more widespread use of antenatal corticosteroids right up to 39 weeks for planned caesarean delivery without labour.13 This necessitates additional planning for glycaemic management of both pre-existing diabetes and GDM. While the benefit of late-term steroids for fetal lung maturation has not specifically been shown in women with diabetes, they are recognised as a high-risk group. Our practice is to give steroids as per usual obstetric indications and up-titrate insulin during that period, using the strategy outlined in Table 2 as a starting point.14

Table 2. Algorithm for % insulin adjustment during antenatal steroid loading. Adapted from Mathieson.14
Day Prandial insulin (short-acting analogues eg: novorapid) Nocte insulin protaphane
D1 – first dose steroids 0% +25%
D2 – second dose steroids +40% +40%
D3* +40% +40%
D4 +20% +20%
D5 +10-20% +10-20%

Can also be applied to women on mixed insulin regimens (mixtard, novomix, Humalog mix)

*Many women will be delivered on Day 3. Insulin can still be safely ceased in women with GDM, as the absolute glycaemic derangement is generally mild.


There is no doubt that GDM has adverse effects on pregnancy outcomes and implications for the long-term wellbeing of mother and infant. While screening for and managing GDM can be mandated on the basis of improved obstetric and neonatal outcomes, it is likely that the long-term cardiometabolic benefits to both mother and child is where the real value is to be found.