Evaluating clinical utility of subgingival and salivary endotoxin activity levels as periodontal biomarkers.

Objectives The use of periodontal biomarkers for identification and monitoring of unique patient populations could foster better stratification of at-risk groups, increase access to treatment for those most in need, facilitate preventive measures and improve personalised care plans. The aim of this study was to examine the diagnostic and prognostic utility of oral lipopolysaccharides as bacterially-derived periodontal biomarkers. Methods Periodontal parameters were recorded, and saliva and subgingival plaque samples were collected at the beginning of the study from periodontally healthy volunteers and periodontitis patients, and three months after completion of conventional periodontal treatment in the periodontitis group. Endotoxin activity in the samples was measured using the recombinant factor C assay. Associations between clinical periodontal parameters and subgingival and salivary endotoxin activities were analysed using a multivariate regression model, while the ROC curve was applied to estimate the sensitivity, specificity and c-statistics for salivary and subgingival endotoxin activities as diagnostic biomarkers for periodontitis. Results Significant correlations were found between subgingival endotoxin activities, probing pocket depth and periodontal diagnosis, which were independent from patients' age, gender and smoking status. In addition, subgingival endotoxin levels had high specificity and sensitivity in detecting periodontal health and disease (0.91 and 0.85 respectively). Salivary endotoxin activity was positively associated with periodontal diagnosis, mean probing pocket depth, percentages of sites over 4 mm and full mouth bleeding score. However, it was inferior in discriminating patients with stable periodontium from those with periodontitis (sensitivity = 0.69, specificity = 0.61) compared to subgingival endotoxin activity. Conclusions Subgingival endotoxin activity has good diagnostic and prognostic values as a site-specific periodontal biomarker and is not influenced by the patient's age, gender or smoking status. In contrast, salivary endotoxin activity, as a patient-level biomarker, is dependent on patient's age, has poorer diagnostic and prognostic capability, but shows good correlations with disease susceptibility and both its extent and severity.