Machine Learning Stratification of Periodontal and Cerebrovascular Disease Status Using Salivary and Subgingival Lipopolysaccharide Activity.

Oral dysbiosis may contribute to systemic inflammation, but taxonomic composition alone does not capture the host-relevant inflammatory activity of microbial products. This study investigated whether salivary and subgingival lipopolysaccharide (LPS) activities could reflect host-microbiome interactions across periodontal and cerebrovascular disease states. Participants from the SECRETO Oral study were stratified by periodontal status and cryptogenic ischaemic stroke. LPS activity was quantified using a recombinant Factor C assay. Log-transformed mean oral LPS activity and the subgingival-to-salivary LPS activity ratio were evaluated alongside demographic and behavioural variables in an exploratory machine-learning framework. Integrated oral LPS activity increased across disease groups and was highest in participants with both periodontitis and stroke, while the LPS activity ratio differed across disease states, indicating altered niche distribution. Models combining demographic, behavioural and LPS-derived features discriminated disease groups better than models using either feature set alone. LPS-derived features therefore provided complementary, but not independently sufficient, discriminatory information. Oral LPS activity may represent a functional marker of microbial inflammatory burden across periodontal and cerebrovascular disease states. These exploratory findings require validation in independent cohorts with paired microbiome and endotoxin data.