1
Epidemiology of Diabetes in Pregnancy

David Simmons

School of Medicine, Western Sydney University, Campbelltown, New South Wales, Australia

PRACTICE POINTS

• The World Health Organization (WHO) (3) has recommended that hyperglycemia first detected at any time during pregnancy should be classified as either:
  • - diabetes mellitus in pregnancy (DIP), or
  • - gestational diabetes mellitus (GDM).
• Pre-gestational diabetes is diabetes that had been diagnosed before pregnancy.
• The prevalence of pre-gestational diabetes has been increasing across the world over >40?years and has a prevalence of 1-5%. Approximately 0.3-0.8% of pregnancies are complicated by type 1 diabetes; the rest are type 2 diabetes, and a small fraction have rare forms of diabetes.
• DIP has a prevalence of 0.2-0.4%, mostly type 2 diabetes postpartum.
• WHO (3) criteria for GDM have now changed, involving a much lower fasting criterion (=5.1?mmol/l), the introduction of a 1?h value after a 75?g oral load (=10.0?mmol/l), and an increased diagnostic cutoff 2?h post load (=8.5?mmol/l). These criteria substantially increase the prevalence of GDM, in some populations to over 35%.
• Non-European ethnicity and obesity are the major risk factors for hyperglycemia in pregnancy; others such as a family history of diabetes, previous GDM, polycystic ovarian syndrome, age, and previous stillbirth or macrosomic infant are important.
• Pre-gestational diabetes and DIP contribute significantly to malformations.
• Total hyperglycemia in pregnancy contributes to adverse pregnancy outcomes on a population level, particularly shoulder dystocia.
• GDM is a precursor of up to 34% of type 2 diabetes in women.
• There is an association between maternal hyperglycemia in pregnancy and obesity, diabetes, and metabolic syndrome in the offspring.

Case History

A 32-year-old woman, G3P2, with no significant past medical history and no family history of diabetes, had a random glucose of 7.8?mmol/l at 8 weeks gestation with a normal oral glucose tolerance test (OGTT) (4.3, 7.6, and 7.4?mmol/l) at 11 weeks (1). Her pre-pregnancy BMI was 19.9?kg/m2. At 28 weeks, she presented acutely, afebrile but with severe general fatigue. A random plasma glucose was 27.2?mmol/l, blood pressure was 110/84?mmHg, and heart rate 106 beats/min. Ketones were 3+, arterial pH was 7.45, bicarbonate 12.1?mmol/l, and base excess -9.8?mmol/l (i.e., compensated metabolic acidosis). HbA1c was 125?mmol/mol (13.6%). Anti-glutamic acid decarboxylase (GAD) antibody was 25.0 (reference range 1-5). She was diagnosed as having type 1 diabetes and commenced insulin therapy. The rest of the pregnancy was uneventful, although total weight gain was only 3?kg and birth weight was 3006?g.

Questions to be answered in this chapter:

• What proportion of pregnancies are complicated by type 1 diabetes, type 2 diabetes, monogenic diabetes, or other rare forms of diabetes?
• What proportion of pregnancies are complicated by GDM?
• What type of patient develops hyperglycemia first detected in pregnancy?
• What is the public health impact of hyperglycemia in pregnancy?

Prevalence of Total Hyperglycemia in Pregnancy

Diabetes in pregnancy (DIP) and gestational diabetes mellitus (GDM) have been terms used in clinical medicine for over 100 years. In 2010 and 2013, respectively, the International Association of Diabetes and Pregnancy Study Groups (IADPSG) (2) and the World Health Organization (WHO) (3) reclassified hyperglycemia in pregnancy into three groups to incorporate all aspects of the range of raised glucose that can increase pregnancy complications:

The global prevalence of total hyperglycemia in pregnancy has recently been estimated to have been 16.9%, or 21.4 million, live births (women aged 20-49 years) in 2013 (4). The highest prevalence was in Southeast Asia at 25.0%, with 10.4% in North America and the Caribbean Region. Low- and middle-income countries are estimated to be responsible for 90% of cases.

Prevalence of Known Pre-Gestational Diabetes in Pregnancy

The prevalence of both type 1 and type 2 diabetes among reproductive-aged women has been increasing globally (5). In the USA, the incidence of type 1 and type 2 diabetes among those aged under 20 years is projected to triple and quadruple by 2050, respectively (5). An example of the growth in pre-gestational diabetes between 1999 and 2005 is shown for Southern California in Figure 1.1 (by age group), where age- and ethnicity-adjusted rates increased from 8.1/1000 in 1999 to 18.2/1000 by 2005 (6).

Figure 1.1 Pregnancies complicated by pre-gestational diabetes, 1999-2005 (per 1000), by age.

There are significant ethnic differences in prevalence. For example, in 2007-2010 among women aged 20-44 years across the USA, prevalence ranged from 2.7% (1.8-4.1%) among non-Hispanic whites, to 3.7% (2.2-6.2%) among Hispanic women, to 4.6% (3.3-6.4%) among non-Hispanic blacks (7). Prevalence rates are higher in other populations (4).

Prevalence of Type 1 Diabetes in Pregnancy

The prevalence of type 1 diabetes in pregnancy is less than in the nonpregnant population in view of the lower standard fertility ratio (SFR) (fertility rate in comparison with the wider population). The SFR in type 1 diabetes is 0.80 (95% CI: 0.77-0.82), and is particularly low among women with retinopathy, nephropathy, neuropathy, or cardiovascular complications (0.63, 0.54, 0.50, and 0.34, respectively) (8). The gap in fertility between women with and without type 1 diabetes has closed considerably over time, and it appears to be greatest for women who were diagnosed as a child, rather than as an adult (9).

The prevalence of type 1 diabetes in pregnancy increases with age, as shown in Table 1.1 for Norway (1999-2004) (10) and Ontario, Canada (2005-2006) (11).

Table 1.1 Prevalence (per 1000) of type 1 and type 2 diabetes in pregnancy, by age.

Besides women with preexisting type 1 diabetes, a small proportion of women with diabetes first diagnosed during pregnancy are found to have type 1 diabetes (see, e.g., the Case History for this chapter). In New Zealand in 1986-2005, 11/325 (3.4%) of women with new diabetes diagnosed postpartum had type 1 diabetes (12). Other women with GDM have autoimmune markers (islet cell antibody [ICA], GAD antibody [GADA], or tyrosine phosphatase antibody [IA-2A]) without necessarily overt DIP. Overall, the prevalence of such autoimmune markers ranges between 1 and 10%, and it is greatest in populations where the prevalence of type 1 diabetes is higher (13). In a Swedish study, 50% women with antibody positivity had developed type 1 diabetes, compared with none among the GDM control subjects (14).

Prevalence of Type 2 Diabetes in Pregnancy

While fertility rates in type 2 diabetes have not been reported, they would be expected to be low (particularly in view of the associated obesity, polycystic ovarian syndrome [PCOS], and vascular disease) (15). Nevertheless, the rates of type 2 DIP are increasing more rapidly than those of type 1 diabetes in pregnancy (16).

In addition to the increasing age-standardized prevalence and lowering of the age at onset of type 2 diabetes (driven by the obesity epidemic), demographic changes (e.g., ethnicity) may partly explain the changes in prevalence over time in individual locations. For example, in Birmingham, UK, in 1990-1998, the ratio of type 1 to type 2 diabetes was 1:2 in South Asians but 11:1 in Europeans (17). In the north of England in...