Microbial Network Signatures Associated With Outcomes of Peri‐Implantitis Therapy

ABSTRACT Aim To investigate the microbial network signatures associated with outcomes following surgical therapy of peri‐implantitis. Materials and Methods Microbiological data from a 1‐year randomised controlled trial of 28 patients (28 implants) were analysed. Based on a composite outcome (probing depth ≤ 5 mm, minimal bleeding on probing and no further marginal bone loss) at 1 year, implant sites were categorised into a successful ( n  = 10) or a non‐successful group ( n  = 18). Peri‐implant crevicular fluid was collected at baseline and at 3, 6 and 12 months and analysed using 16S rRNA gene sequencing. Results A significant compositional shift from baseline to 1 year was observed in the successful group ( p  = 0.001), with a significant increase in Rothia aeria ( p  = 0.048), Haemophilus parainfluenzae ( p  = 0.033) and Neisseria oralis ( p  = 0.025) over time. Network analysis revealed that the successful group exhibited a Porphyromonas gingivalis ‐associated module enriched with high‐centrality commensals at baseline, and significantly greater network edge density ( p  = 0.005) and connectivity ( p  = 0.018) at 1 year. Spearman correlation analyses at 1 year identified significant negative correlations between P. gingivalis and commensals including H. parainfluenzae , Streptococcus species and Veillonella undefined ( p   Conclusion In this exploratory secondary analysis of a small cohort, favourable outcomes of peri‐implantitis therapy were associated with increased microbial network connectivity and commensal centrality. Trial Registration Korean Clinical Research Information Service number KCT0007900