SIGN Diabetes in Pregnancy
Gestational diabetes is associated with a range of adverse outcomes in pregnancy and may also indicate the risk of long-term adverse metabolic outcomes for the mother with rates of up to 70% of women diagnosed with GDM being diagnosed with T2DM at 10 years.129 Earlier detection of abnormalities in glucose metabolism, including T2DM, impaired fasting glucose (IFG) and impaired glucose tolerance (IGT), after delivery may allow the implementation of lifestyle interventions to reverse onset of T2DM or delay and avoid progression of IFG and IGT to frank T2DM.
A small cohort study of women with previous GDM in Ireland showed rates of abnormal glucose tolerance of 20% four years after delivery if WHO criteria were applied, increasing to 56% if ADA criteria (which includes the use of HbA1c as a test for prediabetes) were applied.130 Finding a quicker more convenient and acceptable test may have benefits in improving uptake of screening. Outwith screening women with a past history of GDM, FPG and HbA1c are commonly used tests for diagnosing diabetes.
The NICE guideline on diabetes in pregnancy: management from preconception to the postnatal period identified 13 studies investigating postnatal classification of glucose tolerance in women who have had GDM.14 All studies focussed on use of FPG to diagnose diabetes postnatally, with four studies also investigating use of FPG to detect IFG and IGT postnatally. One study investigated the diagnostic accuracy of HbA1c to detect diabetes postnatally. These studies were very low quality with very serious limitations.
A further four studies comprising one systematic review and three cohort studies were identified.
Evidence reported by NICE showed that an FPG cut-off of 6.0 mmol/L appeared to provide the best balance between ruling in and ruling out diabetes, with FPG at or above this level very useful for ruling in diabetes, and FPG below this level moderately useful for ruling out diabetes. There was no evidence from four cohort studies of a strongly predictive FPG threshold for detecting IGT, but a level less than 7.0 mmol/L was moderately useful for ruling out IGT. A single retrospective cohort study investigating HbA1c to detect postnatal glucose intolerance reported HbA1c cut-offs ranging from 34 to 47 mmol/mol. A value greater than or equal to 39 mmol/mol was very useful for ruling in diabetes. HbA1c was not useful for ruling out diabetes.
Uptake of postnatal screening after a diagnosis of GDM is universally low. Rates of uptake in the UK are 28.2% of eligible women having undergone screening at 12 months postpartum and only 18.5% before six months.129 The authors of this systematic review suggested that strategies such as reminder services, screening co-ordinators and education of women and healthcare professionals to avoid underplaying the risks of GDM after delivery may be helpful strategies in improving uptake.
Changes in the haemoglobin levels and red cell turnover during pregnancy may make HbA1c unreliable in the weeks immediately after pregnancy. A cohort study investigated whether fasting glucose measured 24–72 hours after delivery may be helpful to rule out glucose intolerance.131 The study found that fasting levels at this point were lower than those at six weeks postpartum and therefore could not be used to exclude persisting glucose abnormalities in women with previous GDM. This was thought to reflect reduction in food intake during labour and ongoing effects of the lifestyle changes encouraged following a diagnosis of GDM. A small cohort study of 74 women in Ireland suggested that a combination of HbA1c and fasting glucose for later screening of women at four years postpartum would identify 75% of women diagnosed with abnormal glucose tolerance with an OGTT.130
Good practice
View definitionRecommendations on the choice and timing of testing for postnatal glucose intolerance are included at the end of section 6.2.
For women diagnosed with GDM it is important to identify any persisting abnormalities of glucose metabolism following the birth, such as IGT or progression to T2DM, to allow appropriate preventative measures or treatment strategies to be introduced. Early confirmation of dysglycaemia may be helpful to improve outcomes in future pregnancies. There are some theoretical disadvantages to testing too early after delivery with some studies suggesting false negative readings when women were tested using glucose values up to 72 hours after delivery131 and this may result in false reassurance and failure to intervene. The NICE guideline on diabetes in pregnancy: management from preconception to the postnatal period highlighted that if HbA1c was used up to 13 weeks after delivery an incorrect positive diagnosis could be made as the HbA1c during this period could reflect pregnancy-related hyperglycaemia.14
The NICE guideline identified 51 studies investigating when testing should be undertaken postnatally to identify glucose intolerance in women who have had GDM but are euglycaemic when they are transferred to community care.14 All evidence was rated as very low quality. For practical implementation, studies were categorised according to testing being performed:
· 0–13 weeks after birth
· more than 13 weeks and up to one year after birth
· more than one year after birth.
Testing from 0–13 weeks after birth
For this time interval, NICE reported that using a 75 g OGTT and WHO 1999 diagnostic criteria, diabetes was detected in a median percentage of 8.5% of women (range 1.3% to 50%). Impaired glucose tolerance was detected in a median percentage of 12.9% of women (range 2.5% to 15.3%) and IFG was detected in a median percentage of 6.9% of women (range 1.1% to 15.6%). During this time interval, the median percentage of women taking up the offer of 75 g OGTT was 49.8% (range 13% to 87%).
For this time interval, NICE reported that using an FPG measurement of at least 7.0 mmol/L (the threshold based on the 75 g OGTT applied using the WHO 1999 diagnostic criteria), diabetes was detected in a median percentage of 7.0% of women (range 1.6% to 11.5%) and IFG was detected in a median percentage of 9.3% of women (based on a single non-UK study). During this time interval, the median percentage of women taking up the offer of an FPG test was 53% (range 16% to 86%).
No evidence was identified for inclusion relating to testing for diabetes, IGT or IFG using an HbA1c measurement at up to 13 weeks after the birth.
Testing from more than 13 weeks and up to one year after birth
For this time interval, NICE reported that using a 75 g OGTT and WHO 1999 diagnostic criteria, diabetes was detected in a median percentage of 22.5% of women (range 9.2% to 48.1%). No evidence was identified relating to testing for IGT or IFG at more than 13 weeks and up to one year. During this time interval, the median percentage of women taking up an offer of a 75 g OGTT was 61.5% (range 52% to 73%).
For this time interval, no evidence was identified relating to testing for diabetes or IFG using an FPG measurement or for testing for diabetes, IFG or IGT using HbA1c.
Testing performed at more than one year after birth
For this time interval, NICE reported that using a 75 g OGTT and WHO 1999 diagnostic criteria, diabetes was detected in a median percentage of 12.5% of women (range 7.7% to 43.1%), IGT was detected in a median percentage of 23.8% of women (range 13.4% to 24.1%) and IFG was detected in a median percentage of 3.6% of women (in one cohort study). During this time interval, the median percentage of women taking up the offer of 75 g OGTT was 54% (range 45% to 85%).
For this time interval, NICE reported that using an FPG measurement of at least 7.0 mmol/L (the threshold based on the 75 g OGTT applied using the WHO 1999 diagnostic criteria), diabetes was detected in a median percentage of 12.4% of women (range 6.8% to 18%). No evidence was identified relating to testing for IFG. During this time interval, the median percentage of women taking up the offer of an FPG test was 68.5% (range 63% to 74%).
For this time interval, no evidence was identified relating to testing for diabetes, IGT or IFG using an HbA1c measurement at more than one year after the birth.
The NICE guideline development group concluded, based on the evidence reviewed, that fasting glucose appears to be the most reliable test for identifying women with dysglycaemia or who may progress to T2DM after delivery in women diagnosed with GDM in pregnancy. They recommended that this should take place ideally between 6 and 13 weeks postnatally which allows early recognition and treatment. The group recognised that a fasting glucose test may not be practical for a women with a new baby who may be breastfeeding and may also be challenging to provide and so pragmatically suggested that a (non-fasting) HbA1c could be offered. As this would need to be delayed until at least 13 weeks after delivery to avoid false positive results which reflect hyperglycaemia during pregnancy, the longer the gap before this test was performed, the greater risk of delaying diagnosis of glucose intolerance. NICE highlighted that whilst OGTT may increase the diagnosis rates significantly after 13 weeks, the practical considerations meant that uptake of screening may be further reduced. NICE highlighted the lack of uptake of screening and suggested that uptake rates and barriers to uptake of screening should be monitored and explored.
From evidence identified after the publication of the NICE guideline, a single-cohort study conducted in South Africa investigated the utility of postpartum in-hospital glucose evaluation to identify women at risk of developing diabetes.131 Fasting plasma glucose levels measured 24–72 hours after delivery were significantly lower compared with both antenatal diagnostic measures (after 24 weeks’ gestation) and postnatal OGTT 4–12 weeks postpartum. None of the women identified with hyperglycaemia using OGTT 4–12 weeks postpartum had abnormal fasting glucose levels at 24–72 hours after delivery. The authors note that early postnatal glucose testing failed to identify high-risk individuals and did not demonstrate that in-hospital fasting glucose measurement could help to direct resources to those most in need of surveillance.
A systematic review of barriers and facilitators of attending postnatal screening for T2DM identified 11 primary studies and three systematic reviews of qualitative and quantitative design which provided evidence categorised into seven themes by the review authors.129 Barriers were noted to be:
· the OGTT test
· competing demands on maternal time
· a lack of education and information
· risk perception and fear
· knowledge amongst healthcare professionals
· problems with continuity and co-ordination of care, eg, poor communication between professionals, including from secondary to primary care.
Interventions which may improve uptake included:
· the use of reminders
· increasing awareness of GDM and the risk of subsequent T2DM, by education
· introduction of a more user-friendly and convenient blood glucose test than the OGTT.
Recommendation
View definitionFasting plasma glucose and HbA1c should not be used to determine glucose status before six weeks after delivery as levels may not be representative.
Recommendation
View definitionExplain to women who were diagnosed with gestational diabetes about the risks of gestational diabetes in future pregnancies and offer them testing for diabetes when planning future pregnancies.
Recommendation
View definitionFor women who were diagnosed with gestational diabetes and whose blood glucose levels returned to normal after the birth:
· Offer lifestyle advice (including weight management, diet and exercise).
· Offer a fasting plasma glucose test 6–13 weeks after the birth to exclude diabetes (for practical reasons this might take place at the 6-week postnatal check or timed to co-ordinate with their baby vaccination schedule).
· If a fasting plasma glucose test has not been performed by 13 weeks, offer a fasting plasma glucose test, or an HbA1c test if a fasting plasma glucose test is not possible, after 13 weeks.
· Do not routinely offer a 75 g two-hour OGTT.
Recommendation
View definitionFor women having a fasting plasma glucose test as the postnatal test:
· Advise women with a fasting plasma glucose level below 6.0 mmol/L that:
o they have a low probability of having diabetes at present
o they should continue to follow the lifestyle advice (including weight management, diet and exercise) given after the birth
o they will need an annual test to check that their blood glucose levels are normal
o they have a moderate risk of developing type 2 diabetes, and offer them advice and guidance in line with SIGN 172: Prevention, early recognition and treatment, and remission of type 2 diabetes.
· Advise women with a fasting plasma glucose level between 6.0 and 6.9 mmol/L that they are at high risk of developing type 2 diabetes, and offer them advice, guidance and interventions in line with SIGN 172: Prevention, early recognition and treatment, and remission of type 2 diabetes.
· Advise women with a fasting plasma glucose level of 7.0 mmol/L or above that they are likely to have type 2 diabetes, and offer them a diagnostic test to confirm diabetes.
Recommendation
View definitionFor women having an HbA1c test as the postnatal test:
· Advise women with an HbA1c level below 39 mmol/mol (5.7%) that:
o they have a low probability of having diabetes at present
o they should continue to follow the lifestyle advice (including weight management, diet and exercise) given after the birth
o they will need an annual test to check that their blood glucose levels are normal
o they have a moderate risk of developing type 2 diabetes, and offer them advice and guidance in line with SIGN 172: Prevention, early recognition and treatment, and remission of type 2 diabetes.
· Advise women with an HbA1c level between 39 and 47 mmol/mol (5.7% and 6.4%) that they are at high risk of developing type 2 diabetes, and offer them advice, guidance and interventions in line with SIGN 172: Prevention, early recognition and treatment, and remission of type 2 diabetes.
· Advise women with an HbA1c level of 48 mmol/mol (6.5%) or above that they have type 2 diabetes and refer them for further care.
In most centres in Scotland women with GDM have HbA1c measured 3 months after delivery and are offered entry to the A Healthier Future: type 2 diabetes prevention, early detection and intervention framework.
Recommendation
View definitionRates of uptake of screening should be monitored and the effects of strategies, such as education of women and healthcare professionals, and introduction of screening co-ordinators, should be tested to evaluate improvement in uptake.
Recommendation
View definitionStrategies to improve uptake of screening are vital to allow early interventions and improve metabolic outcomes, for example trying to co-ordinate with other postpartum milestones such as vaccinations.