Consensus approach to differential abundance analysis detects few differences in the oral microbiome of pregnant women due to pre-existing type 2 diabetes mellitus

Oral microbiome dysbiosis has been proposed as a potential contributing factor to rising rates of diabetes in pregnancy, with oral health previously associated with an increased risk of numerous chronic diseases and complications in pregnancy, including gestational diabetes mellitus (GDM). However, whilst most studies examining the relationship between GDM and the oral microbiome identify significant differences, these differences are highly variable between studies. Additionally, no previous research has examined the oral microbiome of women with pre-existing type 2 diabetes mellitus (T2DM), which has greater risks of complications to both mother and baby. In this study, we compared the oral microbiome of 11 pregnant women with pre-existing T2DM with 28 pregnant normoglycaemic controls. We used shotgun metagenomic sequencing to examine buccal swab and saliva rinse samples at two time points between 26 and 38 weeks of gestation. To reduce variation caused by the choice of differential abundance analysis tool, we employed a consensus approach to identify differential taxa and pathways due to diabetes status. Differences were identified at the late time point only. In swab samples, there was increased Flavobacteriaceae , Capnocytophaga , Capnocytophaga gingivalis SGB2479, Capnocytophaga leadbetteri SGB2492 and Neisseria elongata SGB9447 abundance in T2DM as well as increased Shannon diversity and richness. In rinse samples, there was an increased abundance of Haemophilus , Pasteurellaceae , Pasteurellales and Proteobacteria . In contrast to studies of the oral microbiome in T2DM or GDM that use a single differential abundance analysis tool, our consensus approach identified few differences between pregnant women with and without T2DM.