ECE2023 Eposter Presentations Diabetes, Obesity, Metabolism and Nutrition (355 abstracts)
The role of microbiome features in postprandial blood glucose response in patients with gestational diabetes
Polina Popova 1,2 , Eugenii Tikhonov 3,4 , Artem Isakov 1 , Angelina Eriskovskaya 1 , Evgenii Pustozerov 2 , Elena Vasukova 2 , Aleksandra Tkachuk 2 , Anna Anopova 1 , Maria Kokina 1 , Andrey Degilevich 1 , Soha Zgairy 5 , Elad Rubin 5 , Carmel Even 5 , Sondra Turjeman 5 , Tatiana Pervunina 6 , Elena Grineva 2 & Omry Koren 5
1Almazov National Medical Research Centre, World-Class Research Center for Personalized Medicine, Saint Petersburg, Russia; 2Almazov National Medical Research Centre, Institute of Endocrinology, Saint Petersburg, Russia; 3Federal State Budgetary Scientific Institution "Institute of Experimental Medicine", Saint Petersburg, Russia; 4Almazov National Medical Research Centre, Sankt-Peterburg, Russia; 5Bar-Ilan University, Azrieli Faculty of Medicine, Safed, Israel; 6Almazov National Medical Research Centre; Institute of Perinatology and Pediatrics, Saint Petersburg, Russia
Background and Aims: The gut microbiome has been shown to differ between healthy individuals and those with diabetes and even in women with gestational diabetes mellitus (GDM). This raises the question of its role in postprandial glycemic response (PPGR). We aimed to evaluate the impact of microbiome features in PPGR in women with GDM and healthy pregnant women.
Methods: We obtained stool samples for 96 pregnant women (65 GDM, 31 control), previously recruited for GEM-GDM study (NCT03610178), who consented to continuous glucose monitoring (CGM) for 7 days (31.5±3.1 weeks) and provided relevant food diaries. After additional filtering for quality diaries and number of food intake records per person, 45 women were selected for the statistical analysis. CGM data were analyzed with records of 720 meals. Stool samples were collected within 1-2 weeks after recruitment (28.8±3.6 weeks).16S rRNA gene sequence analysis was carried out after sequencing on a MiSeq platform. 780 bacterial features were selected after the following filters were applied: by the frequency of occurrence of the feature in the samples (at least three) and by the number of copies of the bacterial feature (at least 20). The Shapley additive explanations method was implemented to evaluate feature importance.
Results: When microbiome data were added to the input parameters of the model, the greatest contribution to PPGR was made by the abundance of bacteria of the genera Ruminococcus and Bacteroides. Their greater relative abundance was associated with an increase in the peak glycemic level after a meal. The contribution of bacterial traits to the accuracy of PPGR prediction depended on the choice of prediction algorithm and method of diary filtering. Models built with the addition of microbiome data showed better prediction accuracy in cross-validation datasets and on new patient data in predicting peak glycemic levels and 1-hour postprandial glucose levels. Inclusion of the microbiome data slightly increased the accuracy for predicting peak glycemic levels (mean absolute error 0.484 mmol/l vs 0.497 mmol/l for the models with and without microbiome data) and correlation between actual and predicted values (R=0.718 vs R=0.704) based on the full dataset (n=96), but did not change the accuracy in the more strictly filtered dataset (n=45).
Conclusions: Bacteria of the genera Ruminococcus and Bacteroides made the greatest contribution to peak postprandial glycemic level in pregnant women with GDM.
Article tools
My recent searches
My recently viewed abstracts
Authors
Endocrine Abstracts
ISSN 1470-3947 (print) | ISSN 1479-6848 (online)
© Bioscientifica 2024 |
Privacy policy |
Cookie settings
BiosciAbstracts
Bioscientifica Abstracts is the gateway to a series of products that provide a permanent, citable record of abstracts for biomedical and life science conferences.
Find out more