Oral microbiota diversity and composition significantly differed between obese and normal-weight children, and a Random Forest model based on 23 oral and 7 gut co-present genera achieved an area under the curve of 0.73 in distinguishing the two groups.
Key Findings
Results
Oral microbiota diversity and composition significantly differed between obese and normal-weight children.
Study enrolled 136 children total: 68 in the obese (OB) group and 68 in the normal-weight (NW) group, aged 6-17 years.
Participants were recruited from Beijing Children's Hospital with saliva and fecal samples collected.
16S rRNA gene sequencing was used to compare microbial diversity between groups.
No significant differences were observed between groups in age, gender, birth weight, or height, supporting comparability of the groups.
Results
LEfSe analysis identified Neisseria, Peptostreptococcus, and Oribacterium as potential discriminative genera in the obese oral microbiota.
Linear discriminant analysis effect size (LEfSe) was used to identify genera significantly different between NW and OB children.
These three genera were specifically enriched or discriminative in the OB group's oral microbiota.
Relative abundances of major bacterial phyla and genera were aggregated to identify differentially abundant genera between groups.
Results
A Random Forest model based on co-present oral and gut genera achieved an AUC of 0.73 in distinguishing obese from normal-weight children.
The model incorporated 23 oral and 7 gut co-present genera.
Co-present genera were defined as genera detected in both oral cavity and gut samples at >10% prevalence.
The area under the curve (AUC) of 0.73 indicates moderate discriminative ability between NW and OB children.
Results
Both Veillonella and Bifidobacterium were identified as key bacterial genera in both the oral cavity and the gut in relation to childhood obesity.
These genera were among the co-present genera showing distinct relative abundance patterns across oral and gut sites.
Co-present genera exhibited distinct relative abundance patterns across the two body sites.
These genera were identified as important features in the Random Forest classification model.
Results
Correlations were examined between key oral genera and clinical indicators in the study participants.
The study analyzed relationships between oral microbial genera and clinical measurements.
This analysis was part of exploring the mechanistic link between oral microbiota and obesity-related clinical parameters.
Specific clinical indicators examined were not detailed in the abstract but were part of the case-control study design.
Background
A significant proportion of oral microbiota is hypothesized to influence gut microbiota through swallowing, representing an oral-gut microbial linkage pathway.
The study explored co-presence patterns between oral and gut microbiota in relation to childhood obesity.
The oral-gut microbial linkage is hypothesized as a pathway through which oral microbes exert systemic effects.
Co-present genera were defined by >10% prevalence in both oral and gut samples to operationalize this linkage.
What This Means
This research suggests that the community of bacteria living in the mouths of children with obesity is measurably different from that of children at a healthy weight. The study collected saliva and stool samples from 136 children aged 6 to 17 years in Beijing, equally split between those with obesity and those of normal weight. Using genetic sequencing to identify bacteria, the researchers found that specific mouth bacteria — particularly Neisseria, Peptostreptococcus, and Oribacterium — were more characteristic of the obese group, while overall bacterial diversity in the mouth also differed between the two groups.
The study also explored whether bacteria found in both the mouth and the gut might play a role in obesity. Because we constantly swallow saliva, oral bacteria regularly travel to the digestive system, and the researchers hypothesized this could be one way mouth bacteria influence body-wide health. They identified 30 bacterial types present in both locations and used a machine learning model (Random Forest) built on these shared bacteria to distinguish obese from normal-weight children with moderate accuracy (AUC of 0.73). Two bacteria — Veillonella and Bifidobacterium — stood out as particularly important in both the mouth and the gut.
This research suggests that the mouth microbiome could serve as a useful window into understanding and potentially preventing childhood obesity. Since the mouth is easily accessible, future strategies might use oral bacteria as markers to identify children at risk for obesity or even target oral bacteria as part of prevention or treatment approaches. However, as a case-control study, this research shows associations rather than proving that oral bacteria cause obesity, and further research would be needed to understand the direction and mechanisms of these relationships.
Qi Q, Gao C, Yan Y. (2026). The association of oral microbiota and oral-gut microbial co-presence with obesity in Chinese children.. Frontiers in cellular and infection microbiology. https://doi.org/10.3389/fcimb.2026.1866102