Gestational Diabetes - PrEvention of PostPartum Diabetes Study

Survey of the field Gestational Diabetes Mellitus (GDM) is increasing in the world today and is strongly correlated to adverse outcomes both for the baby and the mother. The prevalence differs considerably depending on screening recommendations and diagnostic criteria and is increased among women with obesity. Both the mother and the child face higher health challenges, including manifest T2DM, over the years to come due to GDM. In turn, type 2 diabetes increase the risk of cardiovascular disease, nephropathy, retinopathy and neuropathy. Based on the results of the HAPO study, the International Association of Diabetes and Pregnancy Study Groups proposed new criteria for GDM in a consensus statement in 2012. Adopting a one-step strategy would at least double the number of women with the diagnose GDM. Sweden has started to implement the new diagnostic criteria step-wise in different counties. For diagnostic criteria, see Table 1.

Table 1. Diagnostic criteria for gestational diabetes

Gestational fP- glucose* 1-hour 2-hour glucose(venous) According to EASD 1991 ≥7.0 mmol/L ≥9 mmol/L

• 10mmol/L (capillary) According to WHO Since 2013 ≥5.1 mmol/L ≥10.0 mmol/L ≥8.5 mmol/L

1. and 2-hour glucose after ingestion of 75g glucose after an overnight fast. *fP-glucose= fasting plasma glucose

The metabolic demands of pregnancy can reveal a predisposition for T2DM, especially within 5 years postpartum. Women with GDM are eight times more likely to develop T2DM than women with normoglycemic pregnancies. Risk factors for development of manifest T2DM are common and well-known; 1) Overweight 2) Ethnicity 3) Low level of physical activity. 4) Heredity for T2DM. Women with first degree relatives with T2DM have an increased risk to develop T2DM.

Considering all the risk factors it is possible to influence both weight and physical activity and even if women with GDM are called for a follow up the first year after delivery, they are often left on their own thereafter to implement lifestyle changes and pursue a healthy life-style. This, while the child/children require care and often are the focus of the family. The planned intervention is based on already existing routine care added by information to the nurse at the Child Health Care Centre and a supportive website, specially developed for women at risk to develop T2DM.

Background to the exploratory aims Women with GDM have demonstrated aberrations in lipolytic markers in addition to disturbances in glucose metabolism, eg increased fasting blood free fatty acids (FFA) compared with healthy controls as well as an impaired anti-lipolytic effect of insulin. Abundance of FFA have also been implicated in the development of type 2 diabetes (T2DM) and are also an independent forerunner of T2DM in at-risk populations. An infusion of FFA has been shown to reduce insulin-stimulated uptake of glucose in skeletal muscle. However, the effect FFA on insulin secretion is not resolved, with both increase as well as decreased insulin release being reported. Because GDM subjects are at high risk of transition to T2DM, they constitute an attractive prediabetes group for studies of associations between FFA abnormalities and markers of insulin secretion and insulin resistance. Although, there are several reports on the role of FFA for insulin secretion in GDM no study with enough power has evaluated the relationship in well-matched groups of GDM women and healthy controls in carefully phenotyped subjects. In this study, the intension is to follow fasting FFA, fasting C-peptide, fasting insulin and repeated measurements of these variables for 2 hrs after an OGTT in GDM women out of pregnancy and compare effect of lifestyle intervention with that of care as usual during 5 years. Thus, metabolically healthy women without previous GDM who are out of pregnancy will be enrolled as a reference population. In particular, it is of great interest to test the hypothesis that fasting FFA could be a driver of insulin secretion in normal fasting glycaemia and whether weight reduction may suppress fasting insulin (evaluated in part by HOMA-IR) and FFA in parallel. This sheds light on fundamental metabolic questions in the pathogenesis of GDM and T2D.

Digital support and coaching Mobile technology has been used to treat a variety of clinical conditions, for lifestyle interventions regarding physical activity and diet as well as quitting smoking. For diabetes prevention there are a number of studies with varying methodology (mainly messages on the mobile phone) and most of them have a relatively short follow-up time and have focused on changing weight or HbA1c. In a review article from 2017, the interventions are often assessed as cost-effective. Three success factors are mentioned; automatic transmission of the patient's measurement data rather than manual entry by the patient, tailor-made personal messages or the addition of direct face-to-face contact, and the intervention being in a context where participants can interact with each other.

MyMOWO - The application and website used in this lifestyle intervention is MyMOWO (www.mymowo.com) which is an established company that offers physical training and activity with over 250 different training sessions of various intensity and scope through digital management and support. The service can be accessed using modern technology, such as mobile phone, tablet or computer, via webpage, application or social media. A licenced physiotherapist at the public health unit in Region Jämtland Härjedalen has together with MyMOWO developed special training programs for pregnancies and obesity. These groups also make it possible for MyMOWO's personal trainers to stay in contact with the participants. All participants in the intervention group will receive access to MyMOWO for free with automated support.

In this digital platform links will be given to information films regarding nutrition, the benefit of lifestyle and physical activity, stress, sleeping habits and tobacco. This from www.sundkurs.se. These are films by Prof. Mai-Lis Hellénius produced by the Karolinska Institute and the patient organization Hjärt-Lungfonden and they are used with permission in this project.

The planned intervention is based on already existing routine care added by an interactive digital service specially developed for women at risk to develop T2DM, information to the nurse at the Child Health Care Centre.

Characterization of the phenotype in the PEPP Diabetes study Medical history, socio-economic and lifestyle factors: Standard questionnaires will be used to gain information on previous hospitalisations, on smoking and alcohol habits and on physical activity, leisure time as well as at work. A well validated questionnaire will be used to report diet (MiniMeal-Q).

Physiologic measures: The nurses will measure standard resting blood pressure two times 1 minute apart after five minutes rest with the subjects in a sitting position. (Right brachial artery, arm in heart level, cuff size adjusted for arm circumference, and reading the pressure at the closest 2 mmHg). Body mass index (BMI) and waist hip circumferences are measured using standard methods. All participants will be examined with a bioimpedance measurement at base-line, after one year and at follow-up. All participants will also have their oxygen uptake capacity (VO2max) measured with a bicycle test (Ekblom-Bak) and their physical activity measured with an accelerometer, and a pulswave-measurement will be performed at baseline, after one year and at end-of-study.

Clinical chemistry: Venous blood samples will be collected after an overnight fast. HbA1c, glucose, insulin and cholesterol (HDL, LDL, total cholesterol and triglycerides) will be analysed without delay. Aliqouts of blood, plasma and serum will be frozen immediately at -82 degrees C. A standard oral glucose tolerance test (OGTT) is done in all participants. Insulin secretion and insulin sensitivity is estimated using the HOMA indexes [37], and for insulin secretion also the increment during the first 30, 60 and 90 minutes of the OGTT. Further sera, plasma, and whole blood (for DNA) are stored in a biobank at Gothenburg University. First-voided urine on waking in the morning is collected to determine the albumin-creatinine ratio. Samples will be collected at baseline, after one year and at end-of study.

Statistical analyses:

Both women diagnosed with gestational diabetes according to the "old" (before 2013) and "new" criteria will be included and will be analysed both jointly and as separate subgroups (see table 1). Power calculation is based on data using the old criteria. Inclusion will continue until the number of women with gestational diabetes based on the old criteria is included according to the power calculation.

Power calculation: Around 30% of the participants are expected to convert to T2DM after 5 years without intervention. The effect of the intervention is expected to be around 35% with less individuals progressing to T2DM. According to the power calculation a total of 150 participants will be needed in each group for interindividual comparisons between the intervention and "care as usual" group. Recruitment is planned for three years. Participants will be randomized to either intervention or care as usual in equal numbers (150 in each group). The participants will be stratified according to age, BMI and ethnicity. Data from a number of 150 healthy women with normal with normal glucose tolerance will also be collected.

Individuals progressing to the primary endpoint will be compared with logistic regression and adjusted for baseline data concerning glucose variables, age, BMI, ethnicity, time in study and socioeconomic factors. Differences between changes in BMI, waist-circumference, insulin resistance and HbA1c will be calculated with general linear models (GLM) and adjusted for the same variables, Differences in quality of life will evaluated using Mann-Whitney test. Women who get pregnant during the study time will be examined as soon as the new pregnancy comes to the knowledge of the coordinator and be excluded from the final analyses but will asked to remain in the study.

Data from the women with normal glucose will be collected and matched with data from the women with GDM.

Initially the aim is to conduct a pilot study over one year to test the protocol.

Study time line

As participation in the follow-up after delivery at the health care unit is often low we plan to base the intervention on the regular and well-attended meetings at the Child Health Care Centre. The intervention will be based on information from the Maternity Clinic to Child Health Care Centre.

The maternity center and two months after delivery. The women will get information orally and in writing about the study. A metabolically healthy women will be matched and given information about the study.

4 months post-partum If the woman agrees to be contacted she will be referred to the study center and examined with a physical examination and blood samples.

Examination (by the study nurse):

Anthropometric measurements: Bloodpressure, body weight and height, hip and waist circumference. Bioimpedance and pulse-wave measurements.

Questionnaires: Diet, Life-style, socioeconomic data. Fitness: The Ekblom-Bak submaximal bicycle test. A accelerometer will be applied during ten days.

Blood sampling: Blood samples for plasma glucose, HbA1c, cholesterol (high densitiy lipoprotein, low density lipoprotein and triglycerides) an oral glucose tolerance test (OGTT) will be analysed at once. Samples of whole blood, serum and plasma will be kept and preserved in the biobank.

If informed consent is signed the women will be randomized to intervention or care as usual.

Matched controls from a group of women with a metabolically normal pregnancy will be examined in parallel.

Intervention: At all visits at the child health care centre women in the intervention group will be reminded concerning physical activity and diet and blood-samples will be taken for fasting plasma glucose, HbA1c, weight and waist-circumference and the nurse at the child care clinic will give feed-back concerning the tests. The participant will be supported concerning physical activity and diet at every visit at the clinic on-wards.

The participants in the intervention group will also be informed and given support through a health coach service in V:a Götaland aimed to support lifestyle changes, and by an application, MyMOWO to support lifestyle changes, increase their knowledge of lifestyle-related issues, improve physical activity and dietary habits. The "Hälso-coach" service may be available in other parts of the country, using other names and will be used if available.

At all visits at the child health care centre the participants in the intervention group will have their blood glucose, HbA1c, weight and waist-circumference measured. The nurse will give support concerning physical activity and diet and give feed-back concerning the test results. The participant will be reminded of physical activity and diet at every visit at the clinic on-wards (at most once a month). If the women is diagnosed with either IFG, IGT or have an HbA1c ≥42 mmol/mol the nurse responsible for diabetes care at the health care center will be contacted for action in accordance with standard of care. The women will keep the regular controls at the Child care clinic.

Care as usual: This group of GDM women will be provided care according to health care routine and referred to the health care centre.

This extensive examination will be repeated 18 months and four years after delivery.

Dataprotection Data will be secured in a protected database on a special server in the Vastra Gotaland. All data will be anonymized and coded. Only the responsible researchers will have the code.

Data collection will be monitored by an independent monitoring service from the Gothia Forum.