Objectives: The purpose of this study was to determine the relationship of persistent root canal bacteria, identified with sensitive molecular techniques, with changes in periapical radiographic status following one year as determined using the Periapical Index (PAI).
Methods: Twenty-eight endodontic specimens were included: 20 from cases with necrotic pulps and 8 from cases requiring retreatment for persistent disease. Specimens were obtained at the second endodontic visit, following chemomechanical preparation and irrigation with NaOCl, 17%EDTA, and finally with 2% CHX at both visits, with an interappointment Ca(OH)2 medicament. PCR with broad range 16S rDNA bacterial primers was performed, and for positive reactions, the products were cloned into the vector pCR 2.1-TOPO TA and transformed into E. coli One-Shot TOP10. Plasmid DNA from clones was purified and sequenced. Phylogenetic analysis was performed to reach final identity. Sequences with less than 98% similarity with published sequences were excluded. Radiographs were taken post-operatively and again at 12 months using polyvinylsiloxane stents. Calibrated endodontists independently scored the radiographs and a consensus PAI determined healed versus non-healed periapical status.
Results: Overall, 18 (64.3%) of the cases healed at 1 year (PAI: 1 or 2). Eight cases (29%) did not have detectable bacteria in their post-operative specimens; 7 of them (88%) healed, whereas 11 (55%) of the cases with bacteria healed (Fisher's Exact test; p=0.12). Cases treated to the radiographic apex were less likely to have detectable bacteria (p=0.003) and more likely to heal (p=0.049). When bacteria persisted, two phylotypes were significantly more prevalent in non-healed cases: Enterococcus casseliflavus/flavescens/gallinarum ( p=0.039) and Bacillus sp. clone DQ128250 (p=0.041).
Conclusions: Incomplete elimination of bacteria from the root canal may adversely affect clinical outcome. Certain bacterial phylotypes are more frequently associated with non-healed cases. (Supported by grant DE015320-01-A1 from NIDCR).