Objective To investigate the value of salivary microbial and metabolic profiles for the adjunctive assessment of chronic periodontitis accompanied by type 2 diabetes mellitus (T2DM).

Methods Eighty patients with chronic periodontitis who were treated at the Departments of Stomatology and Endocrinology of Changzhou First People’s Hospital between 2023 and 2025 were enrolled. They were divided into a chronic periodontitis-alone group (CP group, n = 40) and a chronic periodontitis with T2DM group (DM-CP group, n = 40). Periodontal clinical parameters, including plaque index (PLI), sulcus bleeding index (SBI), probing depth (PD) and attachment loss (AL), as well as systemic glycemic indices, including fasting blood glucose (FBG) and glycated hemoglobin (HbA_1c), were recorded for all participants. Unstimulated whole-saliva samples were collected and subjected to microbiome and metabolome analyses using 16S rRNA gene sequencing and liquid chromatography-mass spectrometry (LC-MS)-based metabolomics, respectively.

Results PLI (P = 0.007), PD (P ＜ 0.001), HbA_1c (P ＜ 0.001) and FBG (P ＜ 0.001) were higher in the DM-CP group than in the CP group. After adjustment for age and sex, T2DM status remained associated with greater PD (β = 0.766, P ＜ 0.001) and a higher PLI (β = 0.330, P = 0.008). The 16S rRNA gene sequencing results showed reduced salivary microbial richness and increased relative abundances of genera such as Streptococcus in the DM-CP group, with a statistically significant difference in β-diversity between the two groups (P = 0.001). LC-MS analysis revealed distinct metabolic profiles between the two groups. The differential metabolites were mainly enriched in pathways such as steroid hormone biosynthesis and bile secretion, and metabolites such as neurokinin A were more abundant in the DM-CP group. Several characteristic bacterial genera and metabolites were associated with PD and HbA1c. The exploratory random forest model showed that the AUC of the microbiome metabolome combined model was 0.941(95%CI: 0.799-1.000), with a sensitivity of 1.00 and a specificity of 0.78. Although the AUC was numerically higher than that of the microbiome-only or metabolome-only models, the differences were not statistically significant (both P ＞ 0.05).

Conclusions T2DM may affect the salivary microbial and metabolic profiles of patients with chronic periodontitis. The combined salivary microbiome and metabolome features have value for the adjunctive assessment of chronic periodontitis with T2DM and may provide a reference for noninvasive clinical screening.