Monitoring diabetes before, during and after pregnancy

Recommendations for testing and diagnosis of diabetes in pregnant women have been the subject of much debate (see: “A lack of evidence hampers consensus on how to test for gestational diabetes”). Despite the lack of consensus on testing during pregnancy to identify women with gestational diabetes, the role of testing early in pregnancy and postpartum in women with previous gestational diabetes is much clearer:^{4, 5}

• There is good evidence that hyperglycaemia in early pregnancy resulting from undiagnosed diabetes (usually type 2) results in adverse pregnancy outcomes and that treatment of women with diabetes during pregnancy improves the health of mother and child
• There is good evidence that women with a history of gestational diabetes are at increased risk of future type 2 diabetes

There is, therefore, a sound evidence base to test for undiagnosed diabetes in women who become pregnant, in order to identify those who can benefit from intervention, and for women who have had gestational diabetes to be monitored postpartum and be offered advice and support to reduce their future risk of type 2 diabetes.

Ministry of Health guidelines now recommend that all women should be tested for undiagnosed diabetes early in pregnancy (prior to 20 weeks’ gestation) using HbA_1c. A schedule of oral glucose tolerance tests (OGTT) between 24 to 28 weeks’ gestation to detect gestational diabetes is also recommended, with the specific testing regimen dependent on the HbA_1c test result from early pregnancy (Figure 1).⁴

Testing for undiagnosed diabetes in early pregnancy

Key practice points:

• It is now recommended that all pregnant women undergo testing for pre-existing diabetes
• Use HbA_1c prior to 20 weeks’ gestation to detect pre-existing diabetes
• Women with an HbA_1c ≥ 50 mmol/mol should be referred to a diabetes in pregnancy clinic (Figure 1).
• Women with an HbA_1c between 41 and 49 mmol/mol should be encouraged to adopt lifestyle measures to reduce their risk of adverse pregnancy outcomes; local DHB protocols may vary as to whether to refer these women to a diabetes in pregnancy clinic

Testing for pre-existing diabetes can be performed using HbA_1c. Physiological changes which occur during pregnancy cause red blood cell turnover to increase and HbA_1c levels decline, so HbA_1c should be performed prior to 20 weeks’ gestation to improve accuracy.⁴ Whenever testing for diabetes using HbA_1c, clinicians should keep in mind that some clinical conditions can affect HbA_1c levels and give misleading results (see: Factors affecting the reliability of HbA_1c testing).

HbA_1c test is most easily done as part of the first antenatal blood test screen. If a patient is seen in general practice after they have enrolled with a lead maternity carer (LMC), check that the first antenatal screen, including HbA_1c, has been completed.

HbA_1c testing in early pregnancy will identify women with probable pre-existing diabetes (HbA_1c ≥50 mmol/mol) but also creates a new diagnostic entity – that is women with an HbA_1c of 41 to 49 mmol/mol. The HAPO study showed that mothers with elevated glycaemia below the threshold for diagnosing diabetes are at risk of some adverse pregnancy outcomes so these women should be encouraged to adopt lifestyle measures to reduce their risk of developing gestational diabetes and the adverse pregnancy outcomes associated with elevated glycaemia (see: “Lifestyle approaches”).

A recently published opinion piece argues that women with an HbA_1c ≥ 41 mmol/mol should be referred immediately for management to a diabetes in pregnancy clinic (rather than just those with an HbA_1c ≥ 50 mmol/mol).¹⁵ There is as yet no evidence from randomised controlled trials, however, that earlier pharmacological intervention in these pregnancies improves outcomes (see: “Research into gestational diabetes testing in New Zealand”).

Figure 1: Screening and testing pathways for diagnosing diabetes in pregnancy⁴

Caring for patients with pre-existing diabetes who become pregnant

With rates of both type 1 and type 2 diabetes increasing in New Zealand, and maternal age at pregnancy increasing, it is becoming more likely that clinicians will have under their care women with diabetes who become pregnant. In addition to blood glucose monitoring with the aim of meeting recommended treatment targets (see: “What are the treatment targets?”) these women usually require extra testing during pregnancy, in particular for retinopathy and nephropathy.

Retinopathy testing

Diabetic retinopathy can progress during pregnancy.⁵ Women with diabetes who become pregnant should undergo retinal photography during the first trimester, unless they have had this performed in the previous three months.⁵ Follow-up ophthalmology examinations during pregnancy may be indicated depending on the degree of retinopathy.

Renal testing

Nephropathy during pregnancy is associated with an increased risk of pre-eclampsia, fetal growth restriction and pre-term birth. Ideally, women with pre-existing diabetes should have renal function tests performed in the three months prior to pregnancy, or if not, early in pregnancy at the first point of contact with the clinician.⁵ A protein:creatinine level of 30 mg/mmol reflects a daily protein excretion of 300 mg and is the recommended test for the presence of proteinuria in pregnancy.¹⁸ A serum creatinine level > 90 micromol/L accompanied by hypertension after 20 weeks’ gestation is diagnostic of pre-eclampsia.¹⁸

For further information on other routine laboratory testing during pregnancy, see: www.bpac.org.nz/BT/2011/July/pregnancy.aspx

Ministry of Health guidelines now recommend that oral glucose testing later in pregnancy (at 24 to 28 weeks) be tailored to the patient’s early HbA_1c results. This testing would usually be organised by the midwife. Most women will be at low risk of developing gestational diabetes and can undergo a 50 g oral glucose challenge test. This test has a good negative predictive value, so that women who test negative are at low risk of developing hyperglycaemia and the associated risks of adverse pregnancy outcomes. Therefore, for most women the burden of testing is limited to a one hour test. Women who test positive on the 50 g oral glucose challenge should undergo a 75 g two hour oral glucose tolerance test.

Women who are at increased risk of developing gestational diabetes (initial HbA_1c results of 41 – 49 mmol/mol) should proceed directly to a 75 g two hour oral glucose tolerance test without undergoing an initial 50 g oral glucose challenge test (Figure 1).

There is some disagreement as to whether women with additional risk factors should proceed straight to a 75 g oral glucose tolerance test even if their initial HbA_1c screening result is ≤ 40 mmol/mol. For example, there is concern that an obese woman may test negative on a 50 g glucose challenge as the glucose amounts are not adjusted for body weight, potentially missing a diagnosis.¹⁵

Women who have pre-existing diabetes or have probable undiagnosed diabetes detected with initial HbA_1c screening should be under the care of a diabetes in pregnancy team. Further oral glucose tolerance testing is unlikely to be indicated.

One scenario which is not covered by the Ministry of Health guidelines is which testing procedure should be performed for women who do not have an initial HbA_1c result available. Given that an HbA_1c test can be ordered with the first antenatal blood tests it is likely that women without an HbA_1c measurement in early pregnancy also have a lower level of engagement with health services, and as a result may be at increased risk of adverse pregnancy outcomes. A reasonable course of action would be to request that these women undergo a 75 g two hour oral glucose tolerance test.

Women with diabetes during pregnancy may require oral hypoglycaemic medicines or insulin

Women who have difficulty reaching blood glucose targets or who have high initial blood glucose levels at diagnosis are likely to require hypoglycaemic medicines, such as metformin or insulin injections (see: “What are the treatment targets?”). The regimen of medicines used to control the patient’s blood glucose levels will be determined by the diabetes in pregnancy clinic, tailored to their treatment preferences and the degree of hyperglycaemia.^{5, 20} Metformin is the preferred first-line treatment, as the risk of hypoglycaemia is lower than when using insulin and many patients prefer the ease of taking a tablet rather than using injections. Glibenclamide is recommended as a possible second-line oral hypoglycaemic medicine by the Ministry of Health and National Institute for Health and Care Excellence, where blood glucose control is insufficient with metformin and the mother is unwilling or unable to use insulin, or if she experiences intolerance to metformin.^{4, 5} Patients using only metformin have better outcomes than those using only glibenclamide.^{4, 5}

Women with gestational diabetes should self-monitor blood glucose

Self-monitoring and laboratory measurement of glucose levels during pregnancy are used as the key tests to guide treatment, not HbA_1c. While HbA_1c is useful for monitoring long-term blood glucose control in non-pregnant patients, it does not capture fluctuations in glucose concentrations, and evidence suggests that targeting treatment to postprandial glucose levels results in the best outcomes for both mother and child in gestational diabetes. Furthermore, physiological changes affect HbA_1c levels during pregnancy, causing mean levels to drop compared with non-pregnant women for the same degree of hyperglycaemia.^{4, 5}

What are the treatment targets?

Recommended treatment targets for self-monitored (capillary fingerprick) blood glucose levels are:^{4, 20}

• Pre-prandial (fasting): ≤ 5.0 mmol/L
• Post-prandial, either ≤ 7.8 mmol/L at one hour or ≤ 6.7 mmol/L at two hours

Ministry of Health guidelines recommend that women with gestational diabetes should aim to have > 90% of blood glucose measurements in a week fall within the targets for glucose levels. If more than 10% of measurements fall outside of these ranges, treatment should be reassessed.⁴

Research suggests that the closer to normal blood glucose is during pregnancy, the lower the risk of maternal and neonatal complications. The treatment of all types of diabetes is a balancing act between attaining good glycaemic control while minimising the risk of hypoglycaemia. Maintaining blood sugars that are too low can cause intrauterine growth restriction.^{5, 20}

After pregnancy: general practitioners should reassess glycaemic status

Following a pregnancy affected by gestational diabetes, maternal hyperglycaemia may either resolve completely, or persist – either as intermediate hyperglycaemia or as established diabetes.

Women with previous gestational diabetes have an approximately six to eight-fold higher risk of developing type 2 diabetes than women who have been pregnant without diabetes, and may be at increased risk of developing type 1 diabetes.^{4, 5} Five year incidence rates of type 2 diabetes of 18% to 50% have been reported in women with a history of gestational diabetes.²¹ The best approach is for preventive measures to begin as soon as the mother can manage.

HbA_1c testing at three months postpartum or later is recommended

Clinicians should aim to assess the glycaemic status of all women who have had gestational diabetes. Research suggests that many women with a history of gestational diabetes in New Zealand are not subsequently tested.

Oral glucose tolerance testing at six weeks after birth has been, until recently, the recommended way to assess glycaemic status following gestational diabetes, and this is still recommended in many overseas’ guidelines. However, the Ministry of Health guidelines now recommend using HbA_1c at three months after birth, with the possible addition of fasting blood glucose. It is hoped that this change will facilitate greater uptake of testing given that many patients find the oral glucose tolerance test inconvenient. This change also means that three month postnatal testing uses the same test as later annual monitoring, so that clinicians are better able to determine if glycaemic control has deteriorated.⁴

Annual testing thereafter using HbA_1c is recommended (Table 3). Opportunistic or scheduled patient appointments, such as seeing the clinician when they bring their infant for vaccinations, can be a good time to test the mother for diabetes, or clinicians can set up an electronic reminder to ensure HbA_1c testing is performed at appropriate intervals. Patient reminders such as a letter, email or text are likely to improve rates of postpartum testing for diabetes.²²

Testing HbA_1c at three months postpartum has low sensitivity but high specificity for detecting type 2 diabetes in women who have had gestational diabetes compared to a 75 g oral glucose tolerance test. It is likely to detect those with the highest levels of glycaemia and most in need of treatment. Fasting blood glucose testing can also be performed at three months postpartum at the clinician’s discretion, which increases the sensitivity of three month testing for detecting patients with type 2 diabetes.⁴

Table 3: Tests and appropriate follow-up at three months postpartum for women who have had gestational diabetes⁴

* In most cases this will be type 2 diabetes; occasionally patients with early stage type 1 diabetes or rarely monogenic diabetes (caused by single gene mutations) may be encountered.

Patient and clinician actions that can prevent future diabetes

Lifestyle measures are the cornerstone of preventing and treating type 2 diabetes and clinicians should encourage women with a previous history of gestational diabetes to adopt a diet and exercise regimen which can reduce their risk of future diabetes. In addition to lifestyle measures, other advice and treatments which clinicians can offer include:

Breastfeeding: Mothers who breastfeed usually lose more weight than mothers who do not, and various cohort and observational studies have found that mothers who have breastfed have reduced risks for a number of metabolic diseases across their lifetime, including obesity, diabetes, hypertension, hyperlipidaemia and cardiovascular disease.²³

Metformin: Women with an HbA_1c of 41 – 49 mmol/mol who have not been successful in reducing their level of glycaemia with lifestyle measures could be offered metformin.⁴ Data from the United States Diabetes Prevention Program study suggests a number needed to treat (NNT) of seven patients for ten years to prevent one case of type 2 diabetes.²⁴ Clinicians may wish to consult an endocrinologist or diabetes specialist when considering this treatment approach.