Diabetes and Pregnancy Part 1: Gestational Diabetes

This is the first of two articles about diabetes and pregnancy, and focuses on gestational as opposed to pre-existing diabetes. The case below sets the scene and poses a number of questions that the article will consider.

The effects of hyperglycaemia in pregnancy were first described by Bennewitz[1] who documented a case of a stillborn macrosomic infant in a 22-year-old woman in Berlin. She had symptoms of hyperglycaemia, but these disappeared after the delivery. In 1882, Duncan[2,3] described the outcomes of 22 pregnancies in 16 women. Amongst his conclusions was that hyperglycaemia could resolve after delivery.

Incidences

There are 700,000 pregnancies each year in England and Wales. Of these, 5% are complicated by hyperglycaemia (35,000), of which 12.5% have pre-existing diabetes (4,375) and 87.5% (30,625) gestational diabetes.[4] Gestational diabetes (GDM) was defined by the National Diabetes Data Group in 1979 as any degree of hyperglycaemia at onset or first recognition during pregnancy.[5] This includes undiagnosed pre-existing diabetes as well as people who develop diabetes transiently as a result of pregnancy. In the second group, hyperglycaemia occurs when pancreatic beta cell function is insufficient to overcome the physiological insulin resistance of pregnancy induced by pregnancy hormones such as human placental lactogen and adipokines. Individuals with a pre-disposition to insulin resistance are therefore at risk of gestational diabetes, and risk factors include ethnicity, family history, obesity and polycystic ovarian syndrome.[6-8] Prevalence rates vary widely depending on the diagnostic criteria used and populations studied. Data from the Born in Bradford Study 2007-2011 showed prevalence rates of 2% to 8·7% in white British women and 4% to 24% in south Asian women using six different criteria.[9] Stricter diagnostic criteria and increasing rates of obesity have resulted in escalating numbers, for example, Keiser Permanente of Colorado documented a doubling of numbers between 1994 and 2002.[10]

Why does it matter?

Short-term, gestational diabetes poses a number of risks for mother and baby, including pregnancy-induced hypertension, pre-eclampsia, shoulder dystocia, caesarean section and macrosomia. Maternal hyperglycaemia may also result in fetal hyperinsulinaemia.[11] It was not until 2006 that there was evidence that treatment reduced these risks.[12, 13] Longer term, women with GDM are at increased risk of diabetes, with the incidence of diabetes varying between 2.6 to over 70% depending on the groups studied and the length of follow up. The greatest risk is in the first five years.[14] A Swedish study found a 40% incidence of diabetes 10 years after women presented with GDM, representing a 10-fold increase in diabetes risk compared with women without a history of GDM.[15] The offspring of women with GDM have an increased incidence of obesity, Type 2 diabetes and metabolic syndrome, which is over and above that expected from a genetic predisposition providing indirect evidence of in utero programming as a result of maternal hyperglycaemia.[16,17]

Diagnosis

Despite the evidence above, and over 15,000 publications on gestational diabetes in the last decade, there is still no consensus as to how women should be screened and the glucose thresholds for diagnosis and treatment.

The Hyperglycaemia and Adverse Pregnancy Outcomes Study (HAPO),[18] a multicentred study involving over 25,000 women, demonstrated a continuous relationship between fasting as well as post prandial glucose and adverse pregnancy outcomes even within the accepted normal range. This was a landmark study published in 2008, and it was hoped would inform a consensus as to the diagnostic criteria for GDM. The International Association of Diabetes and Pregnancy Study Groups IADPSG) arrived at a consensus for diagnostic criteria based on fasting and post 75g glucose load.[19] Any one of the following secures a diagnosis of GDM: fasting glucose >5.1mmol/l, I hour >10mmol/l and 2 hour >8.5mmol/l. These criteria were adopted by the World Health Organisation in 2013, and the American Diabetes Association in 2014. The new criteria have been included in the 2014 update of the Scottish Intercollegiate guidance[20] on the management of diabetes, which has also formed the basis of the Irish guidance in this area.[21]

Concerns were expressed that the new criteria would increase the numbers with GDM by two-three times the current level, impacting on limited NHS resources. Meanwhile, the cost effectiveness at intervening at low levels of glucose intolerance remained uncertain.[22] NICE undertook a cost benefit analysis and changing the diagnostic criteria to either a fasting glucose >5.6mmol/l or two hours after 75G Glucose of >7.8mmol/l.[4] While many have welcomed the attempt to balance risk with limited resources, there is a concern that a number of women at increased risk are ‘falling through the net’[23] and also that there will still be a significant impact on workload as a result of adopting the NICE criteria. Locally, we have doubled numbers as a result of adopting the NICE criteria and increase in prevalence generally.

The differences between the Scottish and NICE guidance are summarised in Table 1, and creates a confusing situation for local specialist teams and primary care. With 5% of pregnancies affected by hyperglycaemia, an individual practice is likely to see some patients every year with gestational or preexisting diabetes, and so familiarity with the local guideline is needed. A significant responsibility lies with the midwifery team to take an accurate history so that women requiring an assessment of glucose tolerance can be identified. In addition, women who are known to have pre-diabetes are likely to fulfil the diagnostic criteria for gestational diabetes and should be referred at the earliest opportunity.

Treatment of GDM

Treatment targets are listed in Table 1. It was estimated that using the pre- IDPSG and NICE criteria, 70- 85% of women with gestational diabetes could be managed with dietary advice and lifestyle measures[24] and therefore involvement of a specialist dietitian is essential. Women are advised to eat food low in the glycaemia index and to pair foods, such as adding protein to carbohydrate, to delay the absorption of the latter and so reducing the post prandial glucose rise. A common problem for this motivated group is excessive restriction of their diet and a perception that gestational diabetes is their fault. It is important to support women to make positive changes in their diet without calorie restricting and inducing starvation ketosis. The Irish guidance emphasises this point drawing attention to the fact that severe calorie restriction can also be associated with small for dates infants.[23]

Women are encouraged to be active to improve insulin sensitivity; for example, NICE suggests encouraging women to walk for 30 minutes after meals.[4] This strategy has been effective in our experience when women have a glucose rise in the evening after their biggest meal.

When targets are not met pharmacological treatment is needed. Even a decade ago, the first line treatment was insulin, but there have been a number of studies demonstrating the safety and efficacy of metformin and glibenclamide.[25-27] Up to 40% of women need insulin in addition. However, there is some evidence as to an increased incidence of macrosomia and neonatal hypoglycaemia with glibenclamide compared with insulin. A recent meta-analysis concluded that glibenclamide was “inferior to both insulin and Metformin”.[28] There is insufficient data to use any other oral hypoglycaemic agent or GLP agonists.

Metformin, unlike the other agents, freely crosses the placenta, and the impact on the child needs to be considered. The Metformin in Gestational Diabetes study[29] randomised women to metformin or insulin as the primary treatment and found no differences in perinatal outcomes. The cohort was Regular fetal surveillance is undertaken using ultrasound scanning every two-four weeks to measure growth, liquor volume, and may include Dopplers and markers of foetal wellbeing. The timing of delivery should be individualised, and while there are minor differences between guidelines delivery is brought forward to around 40 weeks due to the risk of stillbirth.

Postnatal management

For those without evidence of overt hyperglycaemia in the immediate postnatal period, until recently, most units have recommended an oral glucose tolerance after six weeks. But uptake is poor and varies widely, with one review finding rates varying between 34-73%.[31] Those who do not attend are at a higher risk of developing diabetes.[32] NICE recommends undertaking a fasting glucose from six weeks postpartum, and if this is not undertaken, an HbA1c at 13 weeks or beyond. A fasting glucose less than 6mmo/l or HbA1c less than 5.7% (39mmol/mol) makes it unlikely that the woman has diabetes, but remains at risk. A fasting glucose which is repeatedly 7mmol/l or above or Hba1c 6.5% (48mmol/mol) or above is suggestive of diabetes. Finally, a fasting glucose of 6-6.9 or HbA1c 5.7 – 6.4% (39-47mmol/mol) is likely to suggest prediabetes, fasting hyperglycaemia or impaired glucose tolerance, meaning the women are at very high risk of developing diabetes. It is hoped that this approach will improve the uptake of postpartum glucose testing. Undertaking a fasting glucose alone would miss those with impaired glucose tolerance, and the recommendation has sparked some controversy. Again, primary care teams will need to be aware of the local interpretation of the guidance.

Prevention

As stated above, women with GDM have a 10-fold increase of developing diabetes with up to 50% developing diabetes in the first 5 years.[14,15] Age, obesity, family history, ethnicity and level of glucose intolerance in pregnancy are risk factors for progression to diabetes. There is the potential to reduce these risks through diet, lifestyle and metformin. One of the landmark studies was the Diabetes Prevention Trial,[35] which recruited people with fasting hyperglycaemia and impaired glucose tolerance (i.e., pre-diabetes) and reduced diabetes risk by 58% in the lifestyle and 31% in metformin group. A subgroup analysis of 350 women with GDM[36] demonstrated the benefit from lifestyle and metformin, whereas the control group benefited from the former alone. This provides evidence for the benefits of intervention in women with a history of GDM to prevent diabetes and raises the question of the use of metfomin as well as lifestyle in diabetes prevention in this group.

Women with GDM should be screened annually for diabetes, and this can be undertaken using HbA1c. Irrespective of the result of the post natal glucose evaluation, women with a history of gestational diabetes should be signposted to local initiates whether they are locally developed or part of a national programme. This is particularly important for those with pre-diabetes, where the progression to diabetes can be prevented or delayed. On a more simplistic level, the birth of a child and the experience of gestational diabetes presents a window of opportunity for the broader community-based teams to engage with women to change their lifestyles as a family.

References

1. Bennetwitz HG, 1824. Diabetologia 1989; 32: 625
2. Duncan JM, Trans Obstet Soc Lond 1882; 24: 256
3. Jorge H. Mestman, The Endocrinologist 2002; 12: 224–242
4. National Institute for Clinical Excellence (2015). Diabetes in pregnancy; management from preconception to postnatal period NICE guideline NG3
5. National Diabetes Data Group. Classification and diagnosis of diabetes mellitus and other categories of glucose intolerance. Diabetes 1979; 28: 1039-1057
6. Teh WT, Teede HJ, Paul E et al. Australian and New Zealand Journal of Obstetrics and Gynaecology 2011; 51: 26–30
7. Pu J, Zhao B, Wang E et al. Paediatric and Perinatal Epidemiology 2015;29(5):436-443
8. Solomon C, Willett W, Carey V et al. JAMA 1997; 278(13): 1078-1083
9. Farrar D, Fairley L, Santorelli G et al. Lancet Diabetes and Endocrinology 2015: 3:795-804
10. Debelea D, Snell-Bergeon JK, Hartsfield CL. Diabetes Care 2005; 28(3):579-584
11. American Diabetes Association. Gestational Diabetes Mellitus. 2003; Diabetes Care 26 (Suppl 1): S103-105.
12. Crowder CA, Hillier JE, Moss JR et al. N Eng J Med 2005; 352:2477-2486
13. Landon MB, Spong CY, Thom E et al. N Eng J Med 2009; 361:1339-1348
14. Kim C, Newton K and Knopp R. Diabetes Care 2002; 25(10): 1862-1868
15. Lauenborg J, Torben H, Jensen DM et al. Diabetes Care 2004; 27(5): 1194-1199
16. Dabelea D and Pettitt DJ. J Pediatr Endocrinolo Metab 2001; 14(8):1085-1091
17. Dabelea D, Hanson R, Lindsay R et al. Diabetes 2000; 49(12):2208-2211
18. O’Sullivan JB and Mahan CM. Diabetes 1964; 13: 278-285.
19. WHO. Expert Committee on Diabetes Mellitus: second report. World Health Organisation Technical Report Series, 1980 accessed via http://apps.who.int/iris/bitstream/10665/41399/1/ WHO_TRS_646.pdf [last accessed 2nd October 2016]
20. Metzger BE, Lowe LP, Dyer AP et al. N Eng J Med 2008; 358: 1991-2002
21. Metzger BE, Gabbe SG, Persson B et al. Diabetes Care 2010; 33: 676-682
22. Scottish Intercollegiate Guidelines Network. Management of Diabetes. A national clinical guideline. 2010 updated May 2014. Accessed via http://www.sign.ac.uk/pdf/sign116.pdf [last accessed 2nd October 2016]
23. Health Service Executive Guidelines for the management of pre-gestational and gestational diabetes mellitus from the pre-conception to the postnatal period. Health Service Executive 2010. Accessed via http://www.hse.ie/eng/about/Who/ONMSD/NMPDU/NMPDDN/Diabetes%20in%20Pregnancy.pdf [ last accessed 2nd October 2016}
24. Cundy T, Ackerman E and Ryan EA. BMJ 2014; 348: g1567
25. Meek C, Lewis HB, Patient C et al. Diabetologia 2015: 58: 2003-2012.
26. American Diabetes Association. Management of diabetes in pregnancy. Diabetes Care 2015; 38 (Suppl 1): S77-S79
27. Kelley KW, Carroll DG and Meyer A. Drugs in Context 2015; 4: 212282 DOI:10.7573/dic.212282 {last accessed 23rd September 2016]
28. Sivalingam VN, Myers J, Nicholas S et al. Human Reproduction Update 2014: 6: 853-868
29. Misiak K, Darbinian JA, Ching JY et al. American Journal of Obstetrics and Gynecology 2016; 214(1) supplement: s92
30. Baisells M, Garcia- Patterson A, Sola I et al. BMJ 2015; 350:h102 doi: 10.1136/bmj.h102
31. Rowan JA Hague WM, Gao W et al. N Engl J Med 2008; 358:2003-2015.
32. Rowan JA, Rush EC, obolonkin V et al. Diabetes Care 2011;24: 2279-2284
33. Tovar A, Chasan-Taber L, Eggleston E and Oken E. Prev Chrni Dis 2011; 8(6): A124 accessed via https://www.ncbi.nlm.nih.gov/pmc/articles/PMC3221566/pdf/PCD86A124.pdf {last accessed 3rd October 2016]
34. Venkataraman H, Sattar N and Ponnusamy S. Lancet 2015; 3(10): 754-756.
35. The Diabetes Prevention Program Research Group. New Engl J Med. 346: 393-403.
36. Ratner RE, Christophi CA, Metzger BE et al. The Journal of Clinical Endocrinology and Metabolism 93: 4774-47

Dr Paru King Consultant in Diabetes and Endocrinology, Derby Teaching Hospitals NHS Foundation trust.