The identification of microrna signatures associated with glycaemic control and pregnancy outcomes in pregnancies complicated by Type 1, Type 2, and gestational diabetes mellitus

Abstract

Background. Diabetes in pregnancy (DIP) is associated with short- and long-term adverse pregnancy outcomes for both mother and child. However, pregestational type 1 (T1DM) and type 2 (T2DM) are associated with more common and severe pregnancy outcomes compared to gestational diabetes (GDM). Maternal biochemical and epigenetic markers and knowledge of diabetes have been associated with glycaemic control and adverse pregnancy outcomes in women with DIP, therefore they offer potential to serve as markers. This may aid in reducing adverse outcomes and improve mother and child health. Aim. The aim of this study was to compare adverse pregnancy outcomes by DIP type and explore the candidacy of adiponectin, leptin, sex hormone binding globulin (SHBG), microRNAs (miRNAs) and diabetes knowledge to serve as markers of glycaemic control during pregnancy and perinatal outcomes in pregnancies complicated by T1DM, T2DM and GDM. Methods. A prospective study was conducted at the high-risk antenatal clinic at Steve Biko Academic Hospital, Pretoria, South Africa between May 2017 and April 2023. The study population consisted of 232 pregnant women with pregestational T1DM (n=27) or T2DM (n=78), GDM (n=58), and normoglycaemia (n=69). Maternal serum adiponectin, leptin and SHBG levels were measured using enzyme linked immunosorbent assays (ELISAs). Maternal serum miRNAs were measured using quantitative real-time PCR. The diabetes knowledge and perceptions questionnaire was developed in three phases and content validity was tested in 20 women with GDM using researcher-administered interviews. Results. Pregestational T1DM and T2DM were associated with an increased risk of preterm birth (p=0.002). Obesity was associated with a higher frequency of GDM (p=0.036), while body weight ≥ 80 kg was associated with caesarean section before the onset of labour (p<0.05). Lower maternal leptin levels were associated with large for gestational age (LGA; p=0.036), macrosomia (birthweight more than 4 kg; p=0.060) and preterm birth (PTB; p=0.004). Lower levels of maternal SHBG were associated with macrosomia (p=0.025) and levels were negatively correlated with neonatal birthweight (r=-0.263, p=001). No association between maternal adiponectin levels and neonatal birth outcomes was observed. Four miRNAs (miR-124-3p, miR-128-3p, miR-20a-5p and miR-210-3p) were associated with small for gestational age (SGA) and were able to predict SGA, miR-124-3p (AUC=0.815), miR-128-3p (AUC=0.760), miR-20a-5p (AUC=0.841) and miR-210-3p (AUC=0.779). MiR-210-3p was associated with macrosomia and demonstrated good predictive ability (AUC=0.779). MiR-222-3p was increased in women with good glycaemic control compared to women with poor glycaemic control during pregnancy. A comprehensive questionnaire for evaluating diabetes knowledge in South African pregnant women with GDM was developed, which performed well in terms of content validity and was able to assess knowledge of diabetes in pregnant women with GDM. Conclusion. To our knowledge, this is the first study to investigate the association between maternal adiponectin, leptin, SHBG and miRNAs with neonatal birth outcomes in South Africa. Our findings suggest that maternal leptin, SHBG and miRNAs may offer potential as biomarkers of neonatal birth outcomes and glycaemic control. Additionally, this is the first study to develop a questionnaire to evaluate diabetes knowledge in women with GDM in South Africa. The developed questionnaire may aid in identifying knowledge gaps in pregnant women with GDM, thereby enhancing education programs and developing interventions to improve glucose management and pregnancy outcomes in women with GDM. However, further studies in larger and multi-ethnic populations are warranted to explore the candidacy of biochemical and miRNA biomarkers for glycaemic control and neonatal birth outcomes. Additionally, the developed questionnaire requires a comprehensive validation process to evaluate construct validity, internal consistency, and test-retest reliability in future.