The prevalence and subtypes of septal defects in offspring of mothers with pre-existing or gestational diabetes

Abstract Background Maternal diabetes, pre-existing or gestational (GDM), affects approximately 6% of all pregnancies in Europe and is associated with adverse outcomes for mother and child. Previous studies of its potential association with offspring atrial and ventricular septal defects (ASD, VSD) have been limited by their size, definitions, and population base. These defects, particularly perimembranous VSDs, may require surgery. Purpose To assess the prevalence of ASD and VSD in newborns exposed to maternal diabetes compared to unexposed newborns in a large population-based cohort, systematically assessed using transthoracic echocardiography (TTE). Methods TTE was performed within 60 days of birth in newborns included in 1) a large general population sample with known maternal diabetes status (2016-2018) and 2) additional infants born to mothers with diabetes during pregnancy (2022-2024). VSDs were classified as muscular, perimembranous, or subarterial. ASDs were defined by defect size (≥ 4mm), location in the inferior 1/3 of the septum, or multiple communications. ASD prevalence was assessed in a sub-population with sufficient images of the septum eligible for evaluation. Maternal diabetes was categorized as pre-existing (type 1 or 2) or GDM. The distribution of ASDs and VSDs among the exposure groups was evaluated with Kruslkal-Wallis test. Relative risk and 95% confidence interval were calculated for the two exposed groups, with the unexposed as reference. Fisher’s Exact test was used to assess the association between exposure and outcome. Results The study population included 26,188 newborns, 378 (1.4%) exposed to pre-existing diabetes, and 1,192 (4.6%) exposed to GDM. Median age at TTE was 11 days (interquartile range: 7–15), and 48% of the newborns were female. The sub-population eligible for ASD evaluation included 13,406 newborns, 307 (2.3%) exposed to pre-existing diabetes and 867 (6.5%) exposed to GDM. There was no significant difference in overall VSD prevalence between the three groups (table 1), (p=0.84). Perimembranous VSDs, however, were more prevalent in newborns exposed to pre-existing diabetes compared to unexposed (1.3% and 0.1% respectively), corresponding to a relative risk of 5.4 (95% confidence interval: 2.3 -12.6, p0.01) (table 1). There was no difference in the prevalence of subtypes between newborns exposed to GDM and unexposed newborns. There was no significant difference in ASD prevalence between the three groups (table. 1), (p=0.29). Conclusion Newborns exposed to pre-existing diabetes had a fivefold increased risk of perimembranous VSDs compared to unexposed newborns. This is the VSD type least likely to close spontaneously. We found no increased risk of septal defects among newborns exposed to GDM compared to unexposed newborns. A follow-up study will clarify the clinical implications of our findings and the potential benefits of screening newborns born to mothers with pre-existing diabetes for VSDs.Table 1.