Table 3.
Prognostic determinants included in the final model and their respective predictive performance
| Study | Prognostic factors included in the final model | Discrimination | Calibration | Classification | ||||
|---|---|---|---|---|---|---|---|---|
| C-statistics | AUC | Slope | Plot | HosLTest | Sensitivity | specificity | ||
| Bengtson 2022 | Weight, BMI, family history of type 2 diabetes, GDM in a prior pregnancy, GDM diagnosis < 24 weeks’ gestation, and fasting and 2-h plasma glucose at 2 days postpartum | 0.79 (0.72, 0.85) | 80.0% | 58.0% | ||||
| Man 2021 | Insulin treatment during pregnancy, BMI, fasting glucose level, HbA1c | 0.74 | ||||||
| Bartáková 2021 | FPG in mid-trimester OGTT above 5.1 mmol/L, obesity, family history of diabetes, instrumental delivery, personal history of hypothyroidism | 0.83 | 82.4% | 90.3% | ||||
| Joglekar 2020 | Age, BMI, pregnancy fasting glucose, postpartum fasting glucose, cholesterol, and triacylglycerols, circulating miR-369-3p measured at 12 weeks postpartum | 0.92 (0.84, 1.00) | 91.0% | 89.0% | ||||
| Muche 2020 | Advanced maternal age, overweight and/or obesity, high FPG at GDM diagnosis, and antenatal depression | 0.88 (0.82–0.94) | P = 0.759 | |||||
| Khan 2019 | Seven lipid metabolites (CE (16:0), NEFA (22:4), TAG 48:2 FA 16:1, CE (20:4), PE(P-18:0/18:1), CE (16:0), TAG 54:0 FA 16:0, TAG 50:1 FA 16:0) | 0.92 (0.89,0.95) | 87.0% | 93.0% | ||||
| Kondo 2018 | Basic model (i.e. age, family history of diabetes, BMI > = 25 kg/m2, and use of insulin during pregnancy) plus insulinogenic index/fasting immunoreactive insulin < 1.1 | 0.71 (0.59–0.83) | P = 0.2102 | |||||
| Allalou 2016 | PC ae C40:5, hexoses, branched-chain amino acids (BCAAs) (Val, Leu, Ile), and SM (OH) C14:1 | 0.83 (0.76–0.89) | 86.3% | 69.0% | ||||
| Ignell 2016 | Ethnic origin, 2-h glucose concentrations during pregnancy, BMI | 0.91 | 86.0% | 82.1% | 88.0% | |||
| Köhler 2016 | Insulin treatment during pregnancy, family history of diabetes, BMI in early pregnancy, and lactation | 0.76 | 1.13 | |||||
| Bartáková 2015 | HbA1c, FPG, 1-h post 75 g load glucose* | |||||||
| Lappas 2015 | Three lipids (CE 20:4, PE(P-36:2), and PS 38:4), six risk factors (age, BMI, and levels of pregnancy fasting glucose, postnatal fasting glucose, triacylglycerol, and total cholesterol) | 0.86 (0.86, 0.87) | 59.0% | 89.9% | ||||
| Cormier 2015 | Age, BMI, explained-variance GRS | 0.67 (0.60–0.73) | ||||||
| Kwak 2012 | Age, pre-pregnancy BMI, family history of diabetes, blood pressure, fasting glucose and fasting insulin concentration, wGRS | 0.77 | ||||||
| Kjos SL 1995 | The area under OGTT glucose curve at 4–14 weeks postpartum, Gestational age at diagnosis of GDM, the area under OGTT glucose curve during pregnancy, and the highest fasting serum glucose centration during pregnancy* | |||||||
HosLTest Hosmer–Lemeshow test, BMI body mass index, GDM gestational diabetes mellitus, FPG fasting plasma glucose, oGTT oral glucose tolerance test, CE(16:0) cholesteryl ester (16:0), NEFA(22:4) non-esterified fatty acids (22:4), TAG 48:2 FA 16:1 triacylglycerol 48:2, fatty acid 16:1, CE(20:4) cholesteryl ester (16:0), PE(P-18:0/18:1) phosphatidylethanolamine, TAG 54:0 FA 16:0 triacylglycerol 54:0, fatty acid 16:0, TAG 50:1 FA 16:0 triacylglycerol 50:1, fatty acid 16:0, HbA1c hemoglobin A1c, wGRS weighted genetic risk score, PC ae C40:5 phosphatidylcholine acyl-alkyl C40:5, SM (OH) C14:1 C14:1-OH sphingomyelin
*The predictive performance is not reported