Validity of hospital discharge records to identify pregestational diabetes in an Italian population

Pregnancy outcomes and maternal characteristics in women with pregestational and gestational diabetes: a retrospective study on 206,917 singleton live births

AIMS

To examine risk of adverse pregnancy outcomes, mothers' characteristics and incidence rate over time of pregestational type 1 (T1D), type 2 (T2D) or gestational diabetes (GDM).

METHODS

The study included all singleton live births born from women aged 15-45 year, in Tuscany, Italy from 2010 to 2018. Pregnancy outcomes were retrieved by certificates of care at delivery compiled by midwives. Pregestational diabetes and GDM were identified by regional administrative databases. Time course of pregestational diabetes and GDM across last decade was assessed by Poisson analysis. Logistic regression analysis was used to calculate adjusted odds ratios (OR; 95% CI) for maternal characteristics or neonatal outcomes.

RESULTS

Among 206,917 singleton live births, GDM was diagnosed in 21,613 pregnancies (10.46%) and pregestational diabetes in 979, being T2D more prevalent than T1D (606; 0.29% vs. 373; 0.18%). Pregestational T2D incidence progressively decreased over last decade, T1D remained stable while GDM progressively rose. Pre-pregnancy obesity, preterm deliveries or cesarean sections were common characteristics of pregestational diabetes and GDM. Risk of neonatal distress and neonatal malformations was higher in pregestational T1D. Risk of prior spontaneous abortions was higher in GDM and in pregestational T2D (OR: 4.19; 3.30-5.33), mostly treated with metformin. Risk of neonatal macrosomia was increased only in pregestational diabetes.

CONCLUSIONS

In our population, pregestational T2D was more prevalent than T1D. Neonatal complications were mostly associated with pregestational T1D. Increased risk of previous spontaneous abortions was the hallmark of pregestational T2D. GDM, even if sharing adverse outcomes with pregestational T2D, was unrelated to rise in risk of neonatal macrosomia.

A Population-Based Study of Diabetes During Pregnancy in Spain (2009-2015): Trends in Incidence, Obstetric Interventions, and Pregnancy Outcomes

(1) Background: We examined trends in incidence and outcomes in women with existing type 1 diabetes mellitus (T1DM), type 2 diabetes mellitus (T2DM) and gestational diabetes mellitus (GDM) compared with a control group without diabetes. (2) Methods: This was an observational, retrospective epidemiological study using the National Hospital Discharge Database. (3) Results: There were 2,481,479 deliveries in Spain between 2009 and 2015 (5561 mothers with T1DM, 4391 with T2DM, and 130,980 with GDM). Incidence and maternal age of existing diabetes and GDM increased over time. Women with T2DM were more likely to have obstetric comorbidity (70.12%) than those with GDM (60.28%), T1DM (59.45%), and no diabetes (41.82%). Previous cesarean delivery, preeclampsia, smoking, hypertension, and obesity were the most prevalent risk factors in all types of diabetes. Women with T1DM had the highest rate of cesarean delivery (Risk Ratio (RR) 2.34; 95% Confidence Interval (CI) 2.26–2.43) and prolonged maternal length of stay. Labor induction was higher in T2DM (RR 1.99; 95% CI 1.89–2.10). Women with T1DM had more severe maternal morbidity (RR 1.97; 95% CI 1.70–2.29) and neonatal morbidity (preterm birth, RR 3.32; 95% CI 3.14–3.51, and fetal overgrowth, RR 8.05; 95% CI 7.41–8.75). (4) Conclusions: existing and GDM incidence has increased over time. We found differences in the prevalence of comorbidities, obstetric risk factors, and the rate of adverse obstetric outcomes among women with different types of diabetes. Pregnant women with diabetes have the highest risk of adverse pregnancy outcomes.

Joint Effect of Maternal Tobacco Smoking and Pregestational Diabetes on Preterm Births and Congenital Anomalies: A Population-Based Study in Northern Italy

Smoking and pregestational diabetes (PGD) are recognized risk factors for adverse pregnancy outcomes, but to date, no population-based study has investigated their joint effects. Using hospital discharges, we identified all women with PGD delivering in Emilia-Romagna region during 2007-2010 matched 1 : 5 with parturients without diabetes. Our study endpoints were preterm births and congenital anomalies. We measured interaction between PGD and maternal smoking, by calculating excess prevalence and prevalence ratio due to interaction, relative excess risk due to interaction (RERI), attributable proportion (AP), and the synergy index (S). Analyses were performed in the overall study population and in the subgroup whose PGD was validated through diabetes registers. The study included 992 women with PGD (10.5% smokers) and 4788 comparison women (11.9% smokers). The effects of PGD and maternal tobacco smoking were greater than additive for both preterm birth (excess prevalence due to interaction = 11.7%, excess ratio due to interaction = 1.5, RERI = 2.39, AP = 0.51, S = 2.82) and congenital anomalies (excess prevalence due to interaction = 2.2%, excess ratio due to interaction = 1.3, RERI = 1.33, AP = 0.49, S = 5.03). Joint effect on both endpoints was confirmed in the subgroup whose PGD status was validated. In conclusion, we found that maternal tobacco smoking and PGD intensify each other's effect on preterm birth and congenital anomalies.