The aim of the present study was to characterise microbiota and inflammatory host response around implants and teeth in patients with peri-implantitis. We included 17 partly edentulous patients with a total of 98 implants, of which 45 showed marginal bone loss of more than three fixture threads after the first year of loading. Nineteen subjects with stable marginal tissue conditions served as controls. Oral hygiene, gingival inflammation, and probing pocket depth were evaluated clinically at teeth and implants. Microbiological and crevicular fluid samples were collected from five categories of sites: 1) implants with peri-implantitis (PI), 2) stable implants (SI) in patients with both stable and peri-implantitis implants, 3) control implants (CI) in patients with stable implants alone, 4) teeth in patients (TP) and 5) controls (TC). Crevicular fluid from teeth and implants was analysed for elastase activity, lactoferrin and IL-1 beta concentrations. Elastase activity was higher at PI than at CI in controls. Lactoferrin concentration was higher at PI than at SI in patients with peri-implantitis. Higher levels of both lactoferrin and elastase activity were found at PI than at teeth in patients. The concentrations of IL-1 beta were about the same in the various sites. Microbiological DNA-probe analysis revealed a putative periodontal microflora at teeth and implants in patients and controls. Patients with peri-implantitis harboured high levels of periodontal pathogens, Actinobacillus actinomycetemcomitans, Porphyromonas gingivalis, Prevotella intermedia, Bacteroides forsythus and Treponema denticola. These findings indicate a site-specific inflammation rather than a patient-associated specific host response.

This retrospective study was designed to verify the factors that influence implant failures. Six prosthodontic clinics in Sweden participated in the study, and together they included a total of 54 patients treated between January 1988 and December 1996. All patients were completely edentulous in the maxilla, and received either a fixed prosthesis or an overdenture supported by at least 4 implants (Brånemark System). Half of the patients belonged to the study group, and an inclusion criterion for this group was that they had lost at least half of their implants. To reduce bias, the patients in the control group were matched to the study group, i.e. they were selected so that both groups were as identical as possible. The results of the study indicate that the control group had a better initial bone support than the study group. Furthermore, the patients in the study group suffered from circumstances that could induce implant failure, such as bruxism, personal grief, depression, as well as addictions to cigarettes, alcohol and/or narcotics. On the study form the clinicians were asked to give their own opinion of the reason for implant failure. The answers given could easily be grouped into 5 different topics, and this experience can be useful to improve patient selection. This study suggests that there are certain factors of importance to consider to prevent a cluster phenomenon of implant failures i.e. lack of bone support, heavy smoking habits and bruxism.