The association between hyperglycaemia and congenital malformations was first recognised over 40 years ago and was followed by the development of preconception clinics for women with diabetes. A fresh look at preconception care is needed as many studies were conducted during the late 1970s and early 1980s, before the introduction of regular home blood glucose monitoring and glycosylated haemoglobin assays, and when many patients with diabetes had microvascular complications. Recent observational studies and a meta-analysis suggest preconception care is effective with an approximately threefold reduction in the risk of malformations. There is now a worldwide epidemic of type 2 diabetes, but only few studies of preconception care have included women with type 2 diabetes. Furthermore, few studies have addressed the relationship between preconception care and perinatal morbidity. This article will review the evidence for preconception care in women with diabetes, evaluate different models of preconception care and discuss future strategies.
Over the last 35 years, it has been recommended that all women with diabetes should plan their pregnancies and access preconception care (PCC) before embarking on a pregnancy. However, pregnancy outcomes remain very poor for women with type 1 and type 2 diabetes with a two- to threefold increase in risk of malformations and a fourfold increase in perinatal death compared with women without diabetes. Most studies report that only a third of women receive PCC. Worldwide, type 2 diabetes is now the most common type of diabetes to complicate pregnancy and women with type 2 diabetes are more likely to enter pregnancy with obesity and taking potentially teratogenic medications.
National guidelines are consistent in recommending PCC (prepregnancy care) as the cornerstone for optimising pregnancy outcome in women with diabetes. However, many studies of PCC were performed in the 1970s and 1980s, before the advent of regular home blood glucose monitoring or measurement of glycosylated haemoglobin (HbA1c). At this time, patients with diabetes frequently had significant microvascular complications. The studies usually include few women with type 2 diabetes. Furthermore, theses studies usually only addressed the relationship between PCC and risk of malformations with few studies examining the relationship between PCC and perinatal morbidity. It is therefore timely to review the evidence behind the recommendations for PCC.
Development of PCC
Relationship between hyperglycaemia and poor pregnancy outcome
Molsted-Pedersen first described the high incidence of congenital malformations in women in 1964 with 6.4% of infants of their diabetic mothers showing a malformation compared with 2.1% of women without diabetes. More recent studies, in women with type 1 and type 2 diabetes, have confirmed that there is a two-to threefold increase in risk of malformations in women with diabetes compared with women without diabetes. Hyperglycaemia has been proposed as a possible mechanism with both animal and human studies supporting this hypothesis.
More recent studies have confirmed the relationship between hyperglycaemia and poor pregnancy outcome. A study from the United Kingdom of 158 pregnancies in women with type 1 diabetes showed a significant increase in both congenital malformations and spontaneous abortion in women with a booking HbA1c above 7.5% compared with women with a booking HbA1c below 7.5%. Pregnancies in women with an HbA1c at booking visit above 7.5% had a fourfold increase in spontaneous abortion rate (relative risk 4.0, 1.2–13.1) and a ninefold increase in the congenital malformation rate (relative risk 9.2, 1.1–79.9). A recent meta-analysis by Inkster and colleagues of seven observational studies in type 1 diabetes and six studies in type 1 and type 2 diabetes confirmed the relationship between poor pregnancy outcome and glycaemic control. They reported approximately threefold increases in spontaneous abortions, malformations and perinatal deaths in pregnancies with poorer glycaemic control. Results of this meta-analysis showed a 0.4–0.6 relative risk reduction of congenital malformation for each 1% fall in HbA1c.
A second meta-analysis of studies of HbA1c and congenital malformation also showed a stepwise fall in the risk of malformation with lowered HbA1c with a 3% risk of malformation for an HbA1c of 6%, a 6% risk for an HbA1c of 9% and a 12% risk for an HbA1c of 12%.
Development of PCC
The concept of PCC for women with diabetes was developed after Pedersen observed the relationship between glucose control and malformations and described how “the occurrence of hypoglycaemic reactions and insulin coma during the first trimester was low in mothers with malformed infants, indicating a poor compensation of the diabetes at that time.” It was recognised that malformations occurred in the first few weeks of pregnancy, and that improvement in glucose control was needed before conception to impact on the risk of malformation. This led to the development of preconception clinics and regional diabetes and pregnancy programmes being established. A personal view of the development of PCC in Scotland in the late 1970s has been well described by Judith Steel.
Aims of preconception clinics – then and now
Diabetes and antenatal care in the late 1970s
It is important to look back at routine diabetes and antenatal care in the 1970s to fully understand both the aims of PCC in the earlier studies, why the development of this care was so important at that time and to interpret results of the studies.
In the late 1970s, routine diabetes care consisted of large diabetic clinics, often staffed by junior doctors with relatively limited diabetes expertise. Patients, usually with type 1 diabetes, controlled their diabetes with once- or twice-daily insulin injections and by limiting carbohydrate intake at mealtimes and snacks. They often had advanced diabetic complications. Urine tests for glycosuria were the only means for checking diabetes control outside the hospital setting. Home blood glucose monitoring, using coloured strips to be read visually, was introduced in the early-1980s, but these colour-read strips were relatively inaccurate for reading blood glucose levels above 9 mmol l −1 . Routine measurement of HbA1c was only developed from the mid-1980s as was the use of basal-bolus insulin regimes, the concept of the annual review with regular fundoscopy and urine checks for proteinuria and the introduction of the diabetes specialist nurse as an educator.
Although obstetric scanning had been developed in the early 1980s, there were no early dating scans until later. During the 1970s and early 1980s, it was routine practice to deliver babies of women with diabetes at 36 weeks to reduce the risk of late stillbirth. Women often had poor glycaemic control and long menstrual cycles, leading to inaccuracies in dating the pregnancy.
The aims of PCC in the 1980s are shown in Table 1 . At that time, screening for diabetic complications was often haphazard, and assessment of the suitability of the patient for pregnancy in relation to the presence of diabetic complications was important. The lack of dating scans combined with the recommendation of delivery at 36 weeks gestation meant that accurate timing of conception was important to reduce the risk of very premature delivery.
| 1976– 1990 | 2010 |
|---|---|
|
|
|
|
|
|
| |
| |
|
|
| Screen for rubella immunity | Screen for rubella immunity |
| |
| Teach patient to monitor basal temperature to time conception | |
| Treat any gynaecological problem / diagnose and treat infertility early |
|
Components of PCC in the twenty-first century
By contrast, in 2010, aims of PCC include working in partnership with the woman and her partner with an emphasis on detailed discussion of risks so they can make informed choices about pregnancy ( Table 1 ). The use of folic acid supplements to reduce the risk of neural tube defects is now an essential component of PCC. With the epidemic of type 2 diabetes, the avoidance of potentially teratogenic medications is increasingly important, along with the management of obesity.
Aims of preconception clinics – then and now
Diabetes and antenatal care in the late 1970s
It is important to look back at routine diabetes and antenatal care in the 1970s to fully understand both the aims of PCC in the earlier studies, why the development of this care was so important at that time and to interpret results of the studies.
In the late 1970s, routine diabetes care consisted of large diabetic clinics, often staffed by junior doctors with relatively limited diabetes expertise. Patients, usually with type 1 diabetes, controlled their diabetes with once- or twice-daily insulin injections and by limiting carbohydrate intake at mealtimes and snacks. They often had advanced diabetic complications. Urine tests for glycosuria were the only means for checking diabetes control outside the hospital setting. Home blood glucose monitoring, using coloured strips to be read visually, was introduced in the early-1980s, but these colour-read strips were relatively inaccurate for reading blood glucose levels above 9 mmol l −1 . Routine measurement of HbA1c was only developed from the mid-1980s as was the use of basal-bolus insulin regimes, the concept of the annual review with regular fundoscopy and urine checks for proteinuria and the introduction of the diabetes specialist nurse as an educator.
Although obstetric scanning had been developed in the early 1980s, there were no early dating scans until later. During the 1970s and early 1980s, it was routine practice to deliver babies of women with diabetes at 36 weeks to reduce the risk of late stillbirth. Women often had poor glycaemic control and long menstrual cycles, leading to inaccuracies in dating the pregnancy.
The aims of PCC in the 1980s are shown in Table 1 . At that time, screening for diabetic complications was often haphazard, and assessment of the suitability of the patient for pregnancy in relation to the presence of diabetic complications was important. The lack of dating scans combined with the recommendation of delivery at 36 weeks gestation meant that accurate timing of conception was important to reduce the risk of very premature delivery.
| 1976– 1990 | 2010 |
|---|---|
|
|
|
|
|
|
| |
| |
|
|
| Screen for rubella immunity | Screen for rubella immunity |
| |
| Teach patient to monitor basal temperature to time conception | |
| Treat any gynaecological problem / diagnose and treat infertility early |
|
Components of PCC in the twenty-first century
By contrast, in 2010, aims of PCC include working in partnership with the woman and her partner with an emphasis on detailed discussion of risks so they can make informed choices about pregnancy ( Table 1 ). The use of folic acid supplements to reduce the risk of neural tube defects is now an essential component of PCC. With the epidemic of type 2 diabetes, the avoidance of potentially teratogenic medications is increasingly important, along with the management of obesity.
Is PCC effective?
PCC and congenital malformations in type 1 and type 2 diabetes
The majority of studies of PCC were carried out 20–30 years ago and have usually evaluated the effect of PCC on the risk of malformation. Most studies have included only women with type 1 diabetes. However, a small number, including a large recent study from the United Kingdom, have also included up to 40% women with type 2 diabetes. Results from these studies, including years of study, numbers of patients and type of diabetes, study setting and HbA1c values, are shown in Tables 2 and 3 . The majority of these studies have demonstrated that PCC is associated with improved glycaemic control in early pregnancy and a significant reduction in major congenital malformations. Furthermore, a meta-analysis of 14 studies of PCC and risk of minor and major malformations, which included 1192 offspring of mothers with PCC and 1459 offspring of mothers with no PCC, has shown that lack of PCC was associated with a threefold increase in the risk of major and minor congenital malformation.
