4. Summary of main results table.
| Overview Review Outcomes | High‐quality evidence | Moderate‐quality evidence |
Low‐quality evidence or very low‐quality evidence |
||||||
| Primary outcomes ‐ maternal | Benefit | Harm | No clear difference | Benefit | Harm | No clear difference | Benefit | Harm | No clear difference |
|
1.0 Hypertensive disorders of pregnancy (including pre‐eclampsia, pregnancy‐induced hypertension, eclampsia 1.1 Any hypertensive disorders of pregnancy, not defined |
Insulin versus oral therapy (Brown 2017d) |
Metformin versus glibenclamide (Brown 2017a) |
Glibenclamide versus placebo (Brown 2017a) Very low |
||||||
| 1.2 Pregnancy‐induced hypertension | Metformin versus glibenclamide (Brown 2017a) |
Glibenclamide versus placebo (Brown 2017a) Low |
|||||||
| Low‐ versus high‐carbohydrate diet (Han 2017) Very low* | |||||||||
| High‐ versus low‐unsaturated fat diet with matching calories (Han 2017) Very low* | |||||||||
| Ethnic specific diet versus standard healthy diet (Han 2017) Very low* | |||||||||
| Insulin regimen A versus B (Brown 2017d)* Low | |||||||||
| 1.3 Pregnancy‐induced hypertension or pre‐eclampsia combined | Glibenclamide versus placebo (Brown 2017a) Low | ||||||||
| Low‐moderate versus moderate‐high GI diet (Han 2017) Very low | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
| 1.4 Pre‐eclampsia | DASH1 diet versus control diet with matching macronutrient contents (Han 2017)* Insulin versus oral therapy (Brown 2017d) |
Metformin versus glibenclamide (Brown 2017a) Very low |
|||||||
| Energy‐ versus no energy‐restricted diet (Han 2017) Low | |||||||||
| High‐ versus low‐unsaturated fat diet with matching calories (Han 2017) Low* | |||||||||
| Soy protein versus no soy protein diet (Han 2017) Very low* | |||||||||
| Lifestyle intervention versus usual care or diet alone (Brown 2017b) Low | |||||||||
| Exercise versus control (Brown 2017c) Low | |||||||||
| Intensive management versus routine care (Han 2012) Low* | |||||||||
| Insulin type A versus B (Brown 2017d) Low* | |||||||||
| Self‐ versus periodic‐glucose monitoring (Raman 2017) Very low | |||||||||
| Post‐ versus pre‐prandial glucose monitoring (Raman 2017) Very low | |||||||||
| 1.5 Eclampsia | Low‐moderate versus moderate‐high GI diet (Han 2017) Very low | ||||||||
| 2.0 Caesarean section | Exercise versus control (Brown 2017c) Insulin versus oral therapy (Brown 2017d) Insulin type A versus B (Brown 2017d)* |
Induction of labour versus expectant management (Biesty 2018) Very low |
|||||||
| Glibenclamide versus placebo (Brown 2017a) Very low | |||||||||
| Metformin versus glibenclamide (Brown 2017a) Low | |||||||||
| Glibenclamide versus acarbose (Brown 2017a) Low | |||||||||
| Low‐moderate versus moderate‐high GI diet (Han 2017) Very low | |||||||||
| Energy‐ versus no energy‐restricted diet (Han 2017) Low | |||||||||
| DASH1 diet versus control diet with matching macronutrient contents (Han 2017) Low* | |||||||||
| Low‐ versus high‐carbohydrate diet (Han 2017) Low* | |||||||||
| High‐ versus low‐unsaturated fat diet with matching calories (Han 2017) Very low* | |||||||||
| Low‐GI diet versus high‐fibre moderate‐GI diet (Han 2017) Very low* | |||||||||
| Diet + diet‐related behavioural advice versus diet only (Han 2017) Very low* | |||||||||
| Soy‐ versus no soy‐protein diet (Han 2017) Very low* | |||||||||
| Ethnic specific diet versus standard healthy diet (Han 2017) Very low* | |||||||||
| Lifestyle intervention versus usual care or diet alone (Brown 2017b) Low | |||||||||
| Intensive management versus routine care (Han 2012) Very low* | |||||||||
| Strict2 versus less strict glycaemic control (Martis 2016a) Very low | |||||||||
| Insulin regimen A versus B (Brown 2017d) Very low* | |||||||||
| Insulin versus exercise (Brown 2017d) Very low* | |||||||||
| Insulin versus diet (Brown 2017d) Very low* | |||||||||
| Post‐ versus pre‐prandial glucose monitoring (Raman 2017) Very low* | |||||||||
| Self‐ versus periodic‐ glucose monitoring (Raman 2017) Low | |||||||||
| Telemedicine versus standard care glucose monitoring (Raman 2017) Very low | |||||||||
| Continuous‐ versus self‐monitoring (Raman 2017) Very low | |||||||||
|
3.0 Development of type 2 diabetes 3.1.1 OGTT3 Test) for diagnosis of type 2 diabetes at one to two weeks postpartum or at four to 13 months postpartum |
High‐ versus low‐unsaturated fat diet with matching calories (Han 2017) Very low* | ||||||||
| 3.1.2 OGTT3 for diagnosis of type 2 diabetes at three months postpartum | Low‐GI diet versus high fibre moderate‐GI diet (Han 2017) Very low* |
||||||||
| 3.1.3 Diagnostic test and time frame not defined | Lifestyle intervention versus usual care or diet alone (Brown 2017b) Low |
||||||||
| 3.1.4 OGTT3 test 6‐8 weeks postpartum | Insulin versus oral therapy (Brown 2017d) | ||||||||
| 3.1.5 Up to 15 years follow‐up. Diagnostic test not defined | Insulin versus diet (Brown 2017d) Very low* | ||||||||
| 4.0 Perineal trauma/tearing | Lifestyle intervention versus usual care/diet alone (Brown 2017b) | Induction of labour verus expectant management (Biesty 2018) Low* | |||||||
| Metformin versus glibenclamide (Brown 2017a) Low | |||||||||
| Glibenclamide versus placebo (Brown 2017a) Very low | |||||||||
| Continuous‐ versus self‐ monitoring (Raman 2017) Very low* | |||||||||
| Post‐ versus pre‐prandial glucose monitoring (Raman 2017) Very low* | |||||||||
| 5.0 Postnatal weight retention or return to pre‐pregnancy weight | Exercise versus control (Brown 2017c) (at follow‐up, timing not defined) |
Lifestyle intervention versus usual care or diet alone (Brown 2017b) (at 12 months post partum) Low |
Lifestyle intervention versus usual care or diet alone (Brown 2017b) (at 6 weeks post partum) Low |
||||||
| Low‐GI diet versus high‐fibre moderate‐GI diet (Han 2017) (at 3 months post partum) Very low* | |||||||||
| Lifestyle intervention versus usual care or diet alone (Brown 2017b) (at 7 months post partum) Very low | |||||||||
| Insulin versus oral therapy (Brown 2017d) (up to 1‐year postpartum) Low | |||||||||
| 6.0 Postnatal depression | Lifestyle intervention versus usual care or diet alone (Brown 2017b) Low | ||||||||
| 7.0 Induction of labour | Lifestyle intervention versus usual care or diet alone (Brown 2017b)* Insulin versus oral therapy (Brown 2017d) |
Glibenclamide versus placebo (Brown 2017a) Very low | |||||||
| Metformin versus glibenclamide (Brown 2017a) Low | |||||||||
| Low‐moderate versus moderate‐high GI diet (Han 2017) Low | |||||||||
| Energy‐ versus no energy‐restricted diet (Han 2017) Low | |||||||||
| Exercise versus control (Brown 2017c) Very low$ | |||||||||
| Intensive management versus routine care (Han 2012)* Very low | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
| 8.0 Large‐for‐gestational age (LGA) (defined as > 90th percentile in all included reviews) | Lifestyle intervention versus usual care or diet alone (Brown 2017b) | Insulin versus oral therapy (Brown 2017d) | Intensive management versus routine care (Han 2012) Low* |
Induction of labour versus expectant management (Biesty 2018) Low |
|||||
| Glibenclamide versus placebo (Brown 2017a) Very low | |||||||||
| Metformin versus glibenclamide Brown 2017a) Low | |||||||||
| Glibenclamide versus acarbose (Brown 2017a) Very low$ | |||||||||
| Myo‐inositol versus placebo4 (Brown 2016a) Very low$ | |||||||||
| Low‐moderate versus moderate‐high GI diet (Han 2017) Very low | |||||||||
| Energy‐ versus no energy‐restricted diet (Han 2017) Low | |||||||||
| Low‐ versus high‐carbohydrate diet (Han 2017) Very low* | |||||||||
| High‐ versus low‐unsaturated fat diet with matching calories (Han 2017) Very low | |||||||||
| Low‐Gi diet versus high‐fibre moderate‐GI diet (Han 2017)* Very low | |||||||||
| Diet + diet‐related behavioural advice versus diet only (Han 2017) Very low* | |||||||||
| Ethnic specific diet versus standard healthy diet (Han 2017) Very low* | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
| Self‐ versus periodic‐glucose monitoring (Raman 2017) Low | |||||||||
| Continous‐ versus self monitoring blood glucose (Raman 2017) Very low | |||||||||
| Post‐ versus pre‐prandial glucose monitoring (Raman 2017) Very low* | |||||||||
| Insulin type A versus B (Brown 2017d) Low* | |||||||||
| Insulin versus diet (Brown 2017d) Very low* | |||||||||
| Insulin regimen A versus B (Brown 2017d) Very low* | |||||||||
| 9.0 Perinatal death (fetal and neonatal death) only | Insulin versus diet (Brown 2017d)* | Induction of labour versus expectant management (Biesty 2018) Very low |
|||||||
| Metformin versus glibenclamide (Brown 2017a) Very low | |||||||||
| Glibenclamide versus acarbose (Brown 2017a) Very low$ | |||||||||
| Energy‐ versus no energy‐restricted diet (Han 2017) Low | |||||||||
| Low‐ versus high‐carbohydrate diet (Han 2017) Very low* | |||||||||
| Lifestyle intervention versus usual care or diet alone (Brown 2017b) Low | |||||||||
| Exercise versus control (Brown 2017c) Very low$ | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
| Self‐ versus periodic‐glucose monitoring (Raman 2017) Very low | |||||||||
| Continuous‐ versus self‐monitoring blood glucose Raman 2017 Very low | |||||||||
| Insulin versus oral therapy (Brown 2017d) Low | |||||||||
| Insulin regimen A versus B (Brown 2017d) Very low* | |||||||||
| 10.0 Death or serious morbidity composite (as defined in reviews, e.g. perinatal or infant death, shoulder dystocia, bone fracture or nerve palsy) | Exercise versus control (Brown 2017c) Insulin versus oral therapy (Brown 2017d) |
Metformin versus glibenclamide (Brown 2017a) Low |
Insulin regimen A versus B (Brown 2017d) Very low* | Ethnic specific diet versus standard healthy diet (Han 2017) Very low* |
|||||
| Lifestyle intervention versus usual care or diet alone (Brown 2017b) Very low | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
|
11.0 Neonatal hypoglycaemia 11.1 Neonatal hypoglycaemia not defined |
Lifestyle intervention versus usual care or diet alone (Brown 2017b) |
Myo‐inositol versus placebo4 (Brown 2016a) Low |
Induction of labour versus expectant management (Biesty 2018) Very low* | ||||||
| Glibenclamide versus placebo (Brown 2017a) Very low | |||||||||
| Energy restricted diet versus no energy restricted diet (Han 2017) Very low | |||||||||
| Low‐carbohydrate diet versus high‐carbohydrate diet (Han 2017) Very low* | |||||||||
| Ethnic specific diet versus standard healthy diet (Han 2017) Very low* | |||||||||
| Exercise versus control (Brown 2017c) Very low$ | |||||||||
| Self‐versus periodic‐glucose monitoring (Raman 2017) Low | |||||||||
| Insulin versus diet (Brown 2017d)* Very low | |||||||||
| 11.2. Neonatal hypoglycaemia defined | Metformin versus glibenclamide (BGL < 2.2 mmol/L; < 40 mg/dL) (Brown 2017a) Low |
||||||||
| Glibenclamide versus acarbose (BGL < 2.2 mmol/L; < 40 mg/dL) (Brown 2017a) Very low$ | |||||||||
| Soy‐ versus no soy‐protein diet (BGL < 1.7 mmol/L (< 30.6 mg/dL) (Han 2017) Very low* | |||||||||
| Intensive management versus routine care evidence (BGL < 1.7 mmol/L in two consecutive measurements (one trial) and as BGL < 1.94 mmol/L (one trial)) (Han 2012) Very low* | |||||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low Defined as <2.6 mmol/L in one trial | |||||||||
| Continuous‐ versus self‐monitoring blood glucose (Raman 2017) Very low Defined as ≤ 2.5 mmol/L in one trial | |||||||||
| Post‐ versus pre‐prandial glucose monitoring (Raman 2017) Very low* Defined as ≤ 30 mg/dL requiring glucagon or dextrose infusion in first four days after birth | |||||||||
| Insulin versus oral therapy (Brown 2017d) Low Definitions varied between trials. | |||||||||
| Insulin type A versus B (Brown 2017d) Very low* | |||||||||
| Insulin versus diet (Brown 2017d) Very low* | |||||||||
| Insulin versus exercise (Brown 2017d) Very low* | |||||||||
| Insulin regimen A versus B (Brown 2017d) Very low* | |||||||||
|
12.0 Adiposity (including skinfold thickness measurements (fat mass g) 12.1 Neonate |
Lifestyle intervention versus usual care or diet alone (Brown 2017b) (whole‐body neonatal fat mass) Low* |
Insulin versus oral therapy (Brown 2017d) (skinfold sum) Very low* Insulin versus oral therapy (Brown 2017d) (% fat mass) Very low$ |
|||||||
| 12.2 Childhood | Lifestyle intervention versus usual care or diet alone (Childhood BMI at 4 to 5 years of age (one trial); 7 to 11 years of age (one trial; 5 to 10 years of age (one trial)) (Brown 2017b) | Insulin versus oral therapy (Brown 2017d) (% fat mass) Low | |||||||
| Lifestyle intervention versus usual care or diet alone (at 4 to 5 years of age) (Brown 2017b) (BMI z score) Very low | |||||||||
| 13.0 Diabetes type 2 child as later infant/childhood/ adulthood | No data reported for this outcome in any of the included reviews | ||||||||
| 14.0 Neurosensory disability in later childhood | Insulin versus oral therapy (any mild developmental delay, hearing and visual impairment) (Brown 2017d) Low |
||||||||
| 15.0 Number of antenatal visits or admissions | Lifestyle intervention versus usual care or diet alone (Brown 2017b) | Soy‐ versus no soy‐protein diet (Han 2017) Very low* | |||||||
| Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low | |||||||||
| Self‐ versus periodic‐glucose monitoring (Raman 2017) Very low | |||||||||
| Insulin versus oral therapy (Brown 2017d) Low* | |||||||||
| 16.0 Length of postnatal stay (mother) | No data reported for this outcome in any of the included reviews | ||||||||
| 17.0 Length of postnatal stay (baby) including NICU/SCBU | Diet + diet‐related behavioural advice versus diet only (Han 2017) Very low* | ||||||||
| Continuous‐ versus self‐monitoring blood glucose (Raman 2017) Very low* | |||||||||
| Insulin versus oral therapy (Brown 2017d) Very low* | |||||||||
| 18.0 Costs associated with the treatment | Lifestyle intervention versus usual care or diet alone (Brown 2017b)* The cost data are based on narrative data |
Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low* |
Self‐ versus periodic‐monitoring Telemedicine versus standard care for glucose monitoring (Raman 2017) Very low* Insulin versus oral therapy (Brown 2017d) Very low* |
||||||
*The GRADE judgement was made by two authors of this overview
$The GRADE judgment was amended from the original review by authors of this overview
1 DASH is an acronym for Dietary Approaches to Stop Hypertension
2Strict intensity of glycaemic control (stricter) defined in review as: pre‐prandial 5.0 mmol/L (90 mg/dL) and one hour post‐prandial 6.7 mmol/L (120 mg/dL) and less strict glycaemic control (liberal) defined in review as: pre‐prandial 5.8 mmol/L (104 mg/dL) and one hour post‐prandial 7.8 mmol/L (140 mg/dL)
3OGTT is an acronym for Oral Glucose Tolerance Test
44 g myo‐inositol + 400 µg folic acid orally per day and exercise and dietary advice versus placebo 400 µg folic acid orally per day and exercise and dietary advice
NICU ‐ neonatal intensive care unit
SCBU ‐ special care baby unit
BMI ‐ body mass index
LGA ‐ large for gestational age
GI ‐ gastrointestinal
BGL ‐ blood glucose level