OBJECTIVE: (I) To compare the oral microflora at implant and tooth sites in subjects participating in a periodontal recall program, (II) to test whether the microflora at implant and tooth sites differ as an effect of gingival bleeding (bleeding on probing (BOP)), or pocket probing depth (PPD), and (III) to test whether smoking and gender had an impact on the microflora.

MATERIAL AND METHODS: Data were collected from 127 implants and all teeth in 56 subjects. Microbiological data were identified by the DNA-DNA checkerboard hybridization.

RESULTS: PPD> or =4 mm were found in 16.9% of tooth, and at 26.6% of implant sites (P<0.01). Tooth sites with PPD> or =4 mm had a 3.1-fold higher bacterial load than implant sites (mean difference: 66%, 95% confidence interval (CI): 40.7-91.3, P<0.001). No differences were found for the red, orange, green, and yellow complexes. A higher total bacterial load was found at implant sites with PPD> or =4 mm (mean difference 35.7 x 10(5), 95% CI: 5.2 (10(5)) to 66.1 (10(5)), P<0.02 with equal variance not assumed). At implant sites, BOP had no impact on bacterial load but influenced the load at tooth sites (P<0.01).

CONCLUSION: BOP, and smoking had no impact on bacteria at implant sites but influenced the bacterial load at tooth sites. Tooth sites harbored more bacteria than implant sites with comparable PPD. The 4 mm PPD cutoff level influenced the distribution and amounts of bacterial loads. The subject factor is explanatory to bacterial load at both tooth and implant sites.