RESUMEN
OBJECTIVES: The aim of this in vitro study was to identify and quantify up to 38 microbial species from human saliva penetrating through the implant-abutment interface in two different implant connections, external hexagon and tri-channel internal connection, both with conventional flat-head or experimental conical-head abutment screws. MATERIAL AND METHODS: Forty-eight two-part implants with external hexagon (EH; n = 24) or tri-channel internal (TI; n = 24) connections were investigated. Abutments were attached to implants with conventional flat-head or experimental conical-head screws. After saliva incubation, Checkerboard DNA-DNA hybridization was used to identify and quantify up to 38 bacterial colonizing the internal parts of the implants. Kruskal-Wallis test followed by Bonferroni's post-tests for multiple comparisons was used for statistical analysis. RESULTS: Twenty-four of thirty-eight species, including putative periodontal pathogens, were found colonizing the inner surfaces of both EH and TI implants. Peptostreptococcus anaerobios (P = 0.003), Prevotella melaninogenica (P < 0.0001), and Candida dubliniensis (P < 0.0001) presented significant differences between different groups. Means of total microbial count (×104 , ±SD) for each group were recorded as follows: G1 (0.27 ± 2.04), G2 (0 ± 0), G3 (1.81 ± 7.50), and G4 (0.35 ± 1.81). CONCLUSIONS: Differences in the geometry of implant connections and abutment screws have impacted the microbial leakage through the implant-abutment interface. Implants attached with experimental conical-head abutment screws showed lower counts of microorganisms when compared with conventional flat-head screws.