Discussion
At follow-up, with the metabolic control, many parameters improved significantly, with respect to the onset, such as glycated hemoglobin (-19%), body mass index (BMI), and also nutritional behaviors, such as normal calorie intake (+6%), carbohydrate intake (-12%), extra portion request (-4%), and meals distribution during the day. Moreover, glycated hemoglobin decrement correlated with both total and rapid absorption carbohydrate intake (Spearman's rho = 0.288, 95% CI 0.066-0.510, p = 0.013), showing as the nutritional behavior supported the insulin therapy efficiency. The next-generation sequencing (NGS) analysis of microbiota revealed abundance differences for Ruminococcus bromii and Prevotella copri (higher at onset, p < 0.001) and the genera Succinivibrio and Faecalibacterium (lower at onset, p < 0.001), as a consequence of nutritional behavior, but it was not the only changing driver. The qRT-PCR analysis showed significant variations, in particular for Bacteroidetes and Bifidobacterium spp. (+1.56 log gene copies/g stool at follow-up, p < 0.001). During the year, in 11% of the patients, severe clinical episodes occurred (hypoglycemic or ketoacidosis). The likelihood of a severe hypoglycemic episode was modulated when the Methanobrevibacter smithii amount increased (odds ratio 3.7, 95% CI 1.2-11.4, p = 0.026). Integrated evaluation, including nutritional behavior and microbiota composition, could be considered predictive of the metabolic control management for children cohort with a recent diagnosis of T1D.
Methods
This study aimed to produce a quantitative comparison of the behavior, nutrition habits, and gut microbiota composition between the onset and the 1-year follow-up in 35 children with T1D.