BACKGROUND: Implant failure and biologic complications such as periimplantitis are not completely avoidable. Are there any genetic and microbiologic parameters that could be used to identify patients at risk for implant failure, preferably prior to treatment? This would result in improvement of the diagnostics, treatment decision, and risk assessment. PURPOSE: The aims of this retrospective study were to describe (1) the absolute failure rate of Branemark System implants (Nobel Biocare AB, Goteborg, Sweden) consecutively installed over a 10-year period in partially edentulous patients treated for periodontal disease prior to implant treatment and under regular professional maintenance, (2) the rate of interleukin-1 (IL-1) polymorphism in those patients who experienced at least one implant failure during the first year of function, and (3) the prevalence of periodontal pathogens in dental and periimplant sites with and without signs of inflammation. MATERIAL AND METHODS: Of 766 patients, 81 encountered at least one implant failure; 22 patients were clinically examined and were tested genetically for IL-1 genotypes. The presence of Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis, and Prevotella ni-grescens was analyzed. RESULTS: The absolute implant survival rate for the whole population was 95.32%; 10.57% of the patients encountered an implant loss. Implant loss in the examined group (n = 22) was 32 of 106 (30.1%); 10 (45%) of the 22 patients were smokers, and 6 (27%) of the 22 patients were IL-1 genotype positive. Patients positive for IL-1 genotype were not more prone to implant loss; however, a significant synergistic effect with smoking was demon-strated. Between patients who were IL-1 genotype positive and those who were IL-1 genotype negative, the differences in regard to bleeding on probing or periodontal pathogens did not reach statistical significance. CONCLUSION: The overall implant failure rate in a population treated and maintained for periodontal disease is similar to that of healthy subjects. A synergistic effect found between smoking and a positive IL-1 genotype resulted in a significantly higher implant loss. This indicates that further research with a larger patient group should focus on multifactorial analysis for adequate risk assessment.