Answers are grounded in this paper and cite their sources.
Background/Objectives: Periodontitis and peri-implantitis are biofilm-associated inflammatory diseases in which conventional clinical and radiographic measures mainly reflect accumulated tissue damage, rather than ongoing molecular activity. Extracellular vesicles (EVs) carry proteins, nucleic acids and immune mediators that may reflect disease activity and treatment response. This systematic review evaluated human evidence on EVs in periodontitis and peri-implant diseases, asking which signals are consistent, whether EV measures add value to existing biomarkers, and which gaps must be addressed first. Methods: The review followed PRISMA 2020 and was registered in PROSPERO (CRD420261277287). PubMed/MEDLINE, Scopus and Web of Science were searched, with citation searching. Human studies of exosomes or specifically characterized EVs were eligible. Risk of bias was appraised with the Newcastle–Ottawa Scale or ROBINS-I. Synthesis was qualitative and structured by biomarker class, biofluid, consistency and isolation method. Results: Sixteen studies were included. The most consistent signal was increased EV concentration in gingival and peri-implant crevicular fluid, although the three supporting studies came from one research network. EV-associated complement proteins and cytokines implicated overlapping inflammatory pathways, indicating consistency at the pathway level, rather than at the level of individual molecules. However, no specific EV-associated microRNA was consistently validated across independent patient cohorts, and tetraspanin-associated findings varied according to the marker examined (CD9, CD63 or CD81). These findings suggest that the biological relevance of EV-associated inflammatory cargo is more consistently supported than the diagnostic reproducibility of individual EV biomarkers. Discrimination estimates were highest in the smallest cohorts, divergent findings coincided with differences in isolation method, and no study compared EV markers head-to-head with established biomarkers. No study followed healthy individuals to incident disease. Conclusions: EVs are biologically plausible candidates for disease characterization and treatment monitoring, but certainty is low for diagnosis and treatment monitoring and very low for therapeutic efficacy, and their incremental value over existing tools has not been tested. Harmonized host-EV methods, head-to-head comparisons and independent validation are priorities, followed by prospective risk-prediction cohorts.
No takes yet. Share an insight, caveat, or question.
Hashim et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: