Rudy Bilous, MD, FRCP
Richard Donnelly, MD, PHD, FRCP, FRACP
Pregestational diabetes is the most common medical problem that complicates pregnancy, affecting 1 in 264 births (0.38%) in the UK in 2002–3, with over 25% occurring in women with type 2 diabetes. The Kaiser Permanente database recorded an average rate of 1.3% in California from 1999 to 2005, increasing from 0.81% to 1.82% of all pregnancies over this time. Diabetes can cause problems for both the mother and fetus and despite recent advances in antenatal care, the outcome in terms of perinatal health and survival remains significantly less good than for pregnancy in the absence of diabetes. Gestational diabetes usually occurs in the second half of pregnancy and is becoming more common because of increasing maternal obesity and age, although absolute rates are highly dependent on the population under study and reflect the background risk of type 2 diabetes….
Effects of pregnancy on the woman with diabetes
In women with pre-existing diabetes, glycemic control can worsen and insulin requirements usually increase during pregnancy, on average by 40% after the 18th week (range 0%–300%) (Box 27.1). This is because pregnancy induces insulin resistance through the diabetogenic effects of placental hormones (mainly growth hormone), cytokines such as TNF-α, and progesterone, the effects of which are maximal in the second and third trimesters. Hyperglycemia as a result of insulin resistance and consequent enhanced lipolysis is probably favorable in the woman without diabetes as it encourages nutrient transfer to the growing fetus, but in diabetes it can be seen as a form of accelerated starvation and predisposes to ketosis. As pregnancy progresses and the diaphragm is pushed upwards, there is a relative increase in alveolar ventilation with a consequent respiratory alkalosis and compensatory renal tubular loss of bicarbonate. There is a fall in serum bicarbonate and loss of acid-buffering capacity, which partly explains why DKA can occur in pregnancy at relatively modest hyperglycemia or even normoglycemia.
Because of the stringent glycemic targets, there is a real risk of hypoglycemia; 41% of 323 women with diabetes in The Netherlands who were pregnant in 1999–2000 reported at least one severe episode, and one died in a road traffic accident almost certainly as a result of hypoglycemia. Women who are driving should be advised to follow carefully guidance on blood glucose monitoring (see Chapter 30).
Effects of maternal diabetes on the pregnancy
Maternal diabetes can affect the fetus adversely by causing developmental malformations and altered islet cell development (increased insulin secretion) and by accelerating growth (macrosomia). Pre-eclampsia is more common in pregnancies with diabetes (particularly in women with nephropathy); there was a reported > 12-fold increase compared to pregnancy without diabetes in The Netherlands in 1999–2000. Perinatal mortality rate has fallen in recent years, but is still 3–4 times that in non-diabetic pregnancy (32/1000 in the Confidential Enquiry into Maternal and Child Health (CEMACH) report in the UK in 2002–3). Stillbirth rates remain about five times those of the non-diabetic population and are linked to standard of glycemic control throughout pregnancy; in the CEMACH enquiry less than 50% of women with a stillbirth ever achieved an HbA1c 70% of women with a normal pregnancy outcome recorded an HbA1c < 7.0% (53 mmol/mol) at some stage. Maternal mortality is also significantly higher than in non-diabetic pregnancies; there were two deaths in the Dutch cohort (0.6% of 323) and five within 1 year of delivery in the CEMACH cohort of 3733, but only three of these may have had a diabetes link.
Diabetes is teratogenic, particularly in the first 8 weeks’ gestation, when the major organs are forming. Major congenital malformations occurred in 144 of the CEMACH cohort, about twice the non-diabetic rate, and there was no difference by type of diabetes; the major defects affected the heart (42%), musculoskeletal (17%) and nervous (13%) systems. Malformation rate is closely related to hyperglycemia in early pregnancy; pre-pregnancy median HbA1c was 8.35% (68 mmol/mol) (interquartile range (IQR) 7.1–10.2) in women whose babies had a major malformation, compared to 7.8% (62 mmol/mol) (IQR 6.8–9.0) in those who had a normally formed baby alive at 28 days. Although pre-conceptual care and tight glycemic control are associated with lower rates (for obvious reasons, there are no randomised controlled trials in this area), 84% of the Dutch cohort pregnancies were planned and 75% achieved a first-trimester HbA1c < 7.0%, despite which their congenital malformation rates were over three times those seen in non-diabetic women. In the CEMACH, 25% of women whose babies had a major malformation had a first-trimester HbA1c < 7.0%. Part of the problem is that HbA1c is only a measure of average glycemia; peak blood glucose or variation may be more important, and very few women have complete normoglycemia before or at conception and during embryogenesis. Nonetheless, the Dutch researchers reported an odds ratio of 0.34 (95% CI 0.13 – 0.88) for major malformations in women with diabetes with a planned versus an unplanned pregnancy.