Glucose intolerance in pregnancy and risk of early-onset type 2 diabetes: a population-based cohort study

Background The risk of type 2 diabetes among women with glucose intolerance during pregnancy that does not meet gestational diabetes criteria requires further investigation. We aimed to explore the associations between various degrees of gestational glucose intolerance and the risk of type 2 diabetes in young adulthood. Methods For this population-based cohort study, the national Israeli conscription database was linked to Maccabi Healthcare Services (MHS), the second-largest state-mandated health provider in Israel. We included 177 241 women who underwent a pre-recruitment evaluation at adolescence (age 16-20 years), 1 year before mandatory military service, and later underwent, from Jan 1, 2001, to Dec 31, 2019, two-step gestational diabetes screening with a 50 g glucose challenge test (GCT) based on a threshold of 140 mg/dL (7 center dot 8 mmol/L), followed as needed by a 100 g oral glucose tolerance test (OGTT). Abnormal OGTT values were defined according to the Carpenter-Coustan thresholds: 95 mg/dL (5 center dot 3 mmol/L) or higher in the fasting state; 180 mg/dL (10 center dot 0 mmol/L) or higher at 1 h; 155 mg/dL (8 center dot 6 mmol/L) or higher at 2 h; and 140 mg/dL (7 center dot 8 mmol/L) or higher at 3 h. The primary outcome was incident type 2 diabetes in the MHS diabetes registry. Cox proportional hazards models were applied to estimate adjusted hazard ratios (HRs) with 95% CIs for incident type 2 diabetes.Findings During a cumulative follow-up of 1 882 647 person-years, and with a median follow-up of 10 center dot 8 (IQR 5 center dot 2-16 center dot 4) years, 1262 women were diagnosed with type 2 diabetes. Crude incidence rates of type 2 diabetes were 2 center dot 6 (95% CI 2 center dot 4-2 center dot 9) per 10 000 person-years in women with gestational normoglycaemia, 8 center dot 9 (7 center dot 4-10 center dot 6) per 10 000 person-years in women with an abnormal GCT and normal OGTT, 26 center dot 1 (22 center dot 4-30 center dot 1) per 10 000 person-years in women with one abnormal OGTT value (in the fasting state or 1 h, 2 h, or 3 h post-challenge), and 71 center dot 9 (66 center dot 0-78 center dot 3) per 10 000 person-years in women with gestational diabetes. After adjustment for sociodemographic characteristics, adolescent BMI, and age at gestational screening, the risk of type 2 diabetes was higher, compared to the gestational normoglycaemia group, in women with an abnormal GCT and normal OGTT (adjusted hazard ratio [HR] 3 center dot 39 [95% CI 2 center dot 77-4 center dot 16]; p<0 center dot 0001), in women with one abnormal OGTT value (9 center dot 11 [7 center dot 64-10 center dot 86]; p<0 center dot 0001), and in women with gestational diabetes (24 center dot 84 [21 center dot 78-28 center dot 34]; p<0 center dot 0001). The risk of type 2 diabetes was modestly increased in women with isolated elevated fasting glucose (adjusted HR 11 center dot 81 [95% CI 8 center dot 58-16 center dot 25]; p<0 center dot 0001), and in women with gestational diabetes and an abnormal fasting glucose (38 center dot 02 [32 center dot 41-44 center dot 61]; p<0 center dot 0001). Interpretation Gestational glucose intolerance, including conditions not meeting gestational diabetes criteria of the two-step strategy, confers a high risk of type 2 diabetes in young adulthood. These conditions should be recognised as risk factors for type 2 diabetes, especially among women with abnormal fasting glucose concentrations during pregnancy.