Aims: To evaluate the checkerboard DNA–DNA hybridization method for detection and quantitation of bacteria from the internal parts of dental implants and to compare bacterial leakage from implants connected either to cast or to pre‐machined abutments.

Materials and methods: Nine plastic abutments cast in a Ni–Cr alloy and nine pre‐machined Co–Cr alloy abutments with plastic sleeves cast in Ni–Cr were connected to Branemark‐compatible implants. A group of nine implants was used as control. The implants were inoculated with 3 μl of a solution containing 10⁸ cells/ml of Streptococcus sobrinus. Bacterial samples were immediately collected from the control implants while assemblies were completely immersed in 5 ml of sterile Tripty Soy Broth (TSB) medium. After 14 days of anaerobic incubation, occurrence of leakage at the implant–abutment interface was evaluated by assessing contamination of the TSB medium. Internal contamination of the implants was evaluated with the checkerboard DNA–DNA hybridization method.

Results: DNA–DNA hybridization was sensitive enough to detect and quantify the microorganism from the internal parts of the implants. No differences in leakage and in internal contamination were found between cast and pre‐machined abutments. Bacterial scores in the control group were significantly higher than in the other groups (P<0.05).

Conclusion: Bacterial leakage through the implant–abutment interface does not significantly differ when cast or pre‐machined abutments are used. The checkerboard DNA–DNA hybridization technique is suitable for the evaluation of the internal contamination of dental implants although further studies are necessary to validate the use of computational methods for the improvement of the test accuracy.