Risk indicators for peri-implantitis. A narrative review
For figures, tables and references we refer the reader to the original paper.
Over the last decades, the use of dental implants has become a common way of replacing teeth. For the “healthy” patient, survival rates for oral implants have been reported to be 91.5%, even after 15 years’ follow-up, at least if the implants are placed under favorable conditions (Dierens et al. 2012). Also, the amount of marginal bone loss is limited in most patients as confirmed by a large-scale meta-analysis on implants ≥5 years in function (Laurell & Lundgren 2011). Varying rates of technical, biological, and esthetic complications can occur (Roccuzzo et al. 2010, 2012, 2014; Jung et al. 2012; Pjetursson et al. 2012a; Romeo & Storelli 2012). Peri-implantitis is clinically defined by the presence of redness and swelling of the mucosa, bleeding and/or suppuration on probing, deepening of the pockets adjacent to the dental implants, and loss of the implant-supporting bone (Lindhe & Meyle 2008). Peri-implantitis has been reported to occur in the range of 1.4% to 53.5% (Buser et al. 2012; Fardal & Grytten 2013). The large differences in the reported prevalence rates may be explained by differences in the definition of peri-implantitis, the inclusion of convenience samples, and the prevalence of maintenance care. Following a consensus conference, it was stated that “the prevalence of peri-implantitis over a 5–10 year period following implant placement has been reported to be in the order of 10% of implants and 20% of patients” (Klinge et al. 2012) and in a recent meta-analysis with data from nine studies and a total of 1497 participants, the frequency of peri-implantitis was reported to be 18.8% on a subject level and 9.6% on an implant level (Atieh et al. 2013).
To obtain long-term success of dental implant therapy, it is essential that the patient can maintain healthy peri-implant tissues. If microorganisms are allowed to develop pathogenic complexes around the implants, inflammation will occur (Pontoriero et al. 1994). When healthy dentate individuals restored with dental implants underwent an experimental undisturbed plaque accumulation over 3 weeks, it was demonstrated that peri-implant soft tissues developed a stronger inflammatory response to experimental plaque accumulation compared with that of the gingiva at teeth. The experimentally induced gingivitis and peri-implant mucositis were reversible as assessed by biomarkers, whereas clinically, 3 weeks of resumed plaque control did not yield pre-experimental levels of gingival and peri-implant mucosal health (Salvi et al. 2012). In a recent systematic review, on risk indicators for peri-implant mucositis, it was concluded that plaque accumulation adjacent to dental implants would result in the development of peri-implant mucositis (Renvert & Polyzois 2014). If such an inflammation persists, it may result in the resorbtion of the implant-supporting bone. Long-term clinical studies have demonstrated a significant correlation between poor oral hygiene and peri-implantitis (Lindquist et al. 1996; Ferreira et al. 2006).
Even if the infectious etiology of peri-implantitis is well established, data on the microbial composition of the infection at dental implants indicate that this infection may be more complex than what has been reported in cases of periodontitis. The infection at dental implants has been described as a poly-microbial anaerobic infection but may differ in bacterial composition to what has been reported in periodontitis lesions (Koyanagi et al. 2010, 2013; Charalampakis et al. 2012; Maruyama et al. 2014; Persson & Renvert 2014).
The etiology of peri-implantitis is multifactorial, and some individuals seem to be more prone to its development than others. An individual's specific susceptibility may increase the risk for the development of peri-implantitis and ultimately loss of the dental implant. Conditions that are related to the development of the disease are considered to be risk factors. As defined by Genco et al. (1996), a risk factor is “an environmental, behavioural, or biological factor that if present directly increases the probability of a disease occurring and, if absent or removed reduces that probability”. The documentation of risk factors requires longitudinal data and demonstration of causality.
Such risk factors can be either local or general. Local risk factors are factors that may influence the bacterial composition and the bacterial load at implants. Examples of such factors are roughness of the implant surface (Teughels et al. 2006), inaccessibility for oral hygiene (Serino & Ström 2009), remaining untreated periodontal pockets (Quirynen et al. 1996; Sumida et al. 2002; Aoki et al. 2012), deep peri-implant pockets (Mombelli et al. 1995; Renvert et al. 2007), the implant supra-structure connection (Mawhinney et al. 2015), the surface structure of the implant (Renvert et al. 2011), and the absence of keratinized tissue adjacent to the implant (Lin et al. 2013).
General risk factors are factors related to the individual and factors that may influence the patient's susceptibility to infection. Risk factors that are often highlighted in this aspect are a history of periodontal disease (Daubert et al. 2015; Renvert et al. 2014), genetic predisposition (Laine et al. 2006), smoking habits (Rinke et al. 2011), the individual's general health status (Renvert et al. 2014), diseases like diabetes mellitus (Daubert et al. 2015), and the individual's motivation to attend supportive care visits and willingness to perform adequate oral hygiene measures (Rinke et al. 2011; Roccuzzo et al. 2012). To avoid the development of peri-implantitis, it has been proposed that patients supplied with dental implants should be enrolled in a structured maintenance program in a similar way as periodontal patients (Rinke et al. 2011; Costa et al. 2012; Pjetursson et al. 2012b).
At present, there are limited scientific data on different risk factors for peri-implantitis. The aim of this review is to examine the existing evidence in identifying risk factors in the etiology of peri-implantitis.
Literature search
For the different subject headings, a literature search was performed in MEDLINE via PubMed database of the US National Library of Medicine, for articles published until October 2014 using Medical Subject Heading search terms + free text terms and in different combinations. To be included in the study, studies had to (i) be written in the English language, (ii) be published in an international peer-reviewed journal, and (iii) have a clear definition for peri-implantitis or the presence of clinical data/assessed parameters, which the reviewers could reliably translate to peri-implantitis. An additional search was also performed to find additional relevant publications.
Local risk factors influencing the bacterial composition and/or the bacterial load at implants
Oral hygiene
As soon as the dental implant is exposed in the oral cavity, microbial colonization of the exposed implant surface occurs (Quirynen et al. 2006; Fürst et al. 2007). In a prospective study by Lindquist et al. (1996), poor oral hygiene was associated with peri-implant bone loss. In a subsample from the original 47 individuals included in the study (13 individuals with good oral hygiene and 14 individuals with poor oral hygiene), bone loss around implants differed at 10 years. A mean bone loss of 0.65 mm was found in the group with good oral hygiene compared to 1.65 mm in the group with bad oral hygiene. Improper accessibility for oral hygiene at the implant sites was related also to peri-implantitis (Serino & Ström 2009). In this study, 48% of the implants presenting peri-implantitis were those with no accessibility/capability for proper oral hygiene. In this part of the review, papers identifying the level of plaque accumulation related to the presence of peri-implantitis were assessed.
Results
Ferreira et al. (2006) analyzing risk factors for peri-implant disease reported very poor oral hygiene to be a risk factor for peri-implant disease (OR 14.3 (95% CI: 9.1–28.7) P = 0.002 (Table 1). In a cross-sectional study, Roos-Jansåker et al. (2006c) reported that the presence of plaque was explanatory for the outcome event peri-implant mucositis but not for peri-implantitis. In a 5-year follow-up study, Costa et al. (2012) reported plaque to be a significant factor for the development of peri-implantitis.
Table 1. Plaque as a risk indicator for peri-implantitis
Discussion
Plaque accumulation adjacent to dental implants is clearly associated with the development of peri-implant mucositis (Renvert & Polyzois 2014). Peri-implant mucositis may, however, not always result in the development of peri-implantitis, but in cases where peri-implantitis develops, an infection has been demonstrated to be a prerequisite for the disease (Ericsson et al. 1992; Leonhardt et al. 1993; Pontoriero et al. 1994; Salvi et al. 2012). The present review identified two papers that reported plaque accumulation to be associated with peri-implantitis (Ferreira et al. 2006; Costa et al. 2012). In a cross-sectional paper by Roos-Jansåker et al. (2006c), an association between plaque and peri-implant mucositis, but not with peri-implantitis, was reported. As peri-implantitis takes time to develop and the presence of plaque at teeth and/or implants may vary over time, a cross-sectional design may be a less suitable study design to evaluate such a relationship.
Microbial biofilm composition
Aggregatibacter actinomycetemcomitans (A. actinomycetemcomitans), Porphyromonas gingivalis (P. gingivalis), and Tannerella forsythia (T. forsythia) are key pathogens in the process of periodontitis (Haffajee & Socransky 1994) and probably also play a significant role in the development and progression of peri-implant mucositis and peri-implantitis (Mombelli et al. 1987; Leonhardt et al. 1993; Shibli et al. 2008; Mombelli & Decaillet 2011). The presence of periodontopathic bacteria has been proposed as a risk indicator for peri-implant mucositis (Renvert & Polyzois 2014). The microenvironment (i.e. sulcus/pockets) around implants may favor the colonization of anaerobic Gram-negative bacteria. An interaction within the biofilm may also contribute to the aggregation of microorganisms allowing for the destruction of the peri-implant tissues. The microbiota associated with peri-implant disease may be described as a poly-microbial anaerobic infection (Koyanagi et al. 2010, 2013; Mombelli & Decaillet 2011; Al-Radha et al. 2012; Charalampakis et al. 2012; Dabdoub et al. 2013; Maruyama et al. 2014; Persson & Renvert 2014), and microorganisms often associated with periodontal disease such as T. forsythia, P. gingivalis, Treponema denticola (T. denticola), Prevotella nigrescens (P. nigrescens), Prevotella intermedia (P. intermedia), and Fusobacterium nucleatum (F. nucleatum) have all been isolated around implants with peri-implant disease. In 2011, Mombelli & Décaillet presented a comprehensive review of the literature on the characteristics of biofilms in peri-implantitis. The authors concluded that peri-implantitis may be viewed as a mixed anaerobic infection and that in most cases, the composition of the flora is similar to the subgingival flora at chronic periodontitis. However, studies have also indicated that microorganisms not frequently found in periodontitis patients (Staphylococcus aureus, Enerobacteriaceae, Candida albicans, and Pseudomonas aeruginosa) are present in areas with a clinical diagnosis of peri-implantitis (Alcoforado et al. 1991; Leonhardt et al. 1999; Renvert et al. 2007; Salvi et al. 2008).
This section of the present review focused on the comparison of microbiota at healthy implants and implants with a clinical diagnosis of peri-implantitis and on papers comparing the microbiota at peri-implantitis and periodontitis sites.
Results
A total of 15 papers presented in Table 2 were included in this part of the review. The microbiota at healthy and diseased implants as well as at healthy teeth and teeth with periodontitis has been evaluated in many studies employing different methods of bacterial detection. Leonhardt et al. (1999) and Botero et al. (2005), both using culturing technique, compared the microbiota at implants diagnosed with peri-implantitis to healthy implants. The authors reported the two clinical conditions to have distinct different bacterial profiles. In areas diagnosed with peri-implantitis, enteric rods, staphylococci, and yeasts were frequent findings.
Using DNA–DNA hybridization methods for the diagnosis of periodontal pathogens, Hultin et al. (2002) reported higher amounts of bacteria in sites with peri-implantitis compared to healthy sites, whereas Ebadian et al. (2012) reported no significant differences between healthy and peri-implantitis sites. Casado et al. (2011), using PCR technique for the detection of periodontal pathogens, found no significant difference in the detection of A. actinomycetemcomitans, P. gingivalis, P. intermedia, and T. denticola between healthy and diseased groups. Comparing the microbiota at implants diagnosed with peri-implantitis with teeth with periodontitis, significant differences were found for the prevalence of P. gingivalis, T. forsythia, P. intermedia, or T. denticola as determined by the PCR technique (Albertini et al. 2014).
Zhuang et al. (2014) analyzing P. gingivalis, T. denticola, A. actinomycetemcomitans, F. nucleatum, P. intermedia, and Staphylococcus aureus, using quantitative real-time polymerase chain reaction in a group of individuals with healthy gingiva, periodontitis, healthy peri-implant tissue, and peri-implantitis, reported F. nucleatum to be more abundant in periodontitis sites (P = 0.023).
Shibli et al. (2008) analyzing 36 species with DNA–DNA hybridization technique reported peri-implantitis sites to present with higher bacterial counts. For P. gingivalis, T. denticola, and T. forsythia, this difference was significant (P < 0.05). In contrast, Renvert et al. (2007) analyzing 40 species using DNA–DNA hybridization technique reported few differences related to implant status but found deeper pockets at implants to be correlated with higher levels of Eikenella corrodens (P < 0.05), F. nucleatum sp. vincentii (P < 0.05), P. gingivalis (P < 0.05), and Parvimonas micra (P < 0.01).
In a large study, using DNA–DNA hybridization technique, examining 166 peri-implantitis cases and 47 individuals with clinically healthy implants, a clear difference was observed between healthy implants and implants diagnosed with peri-implantitis. Additionally, a cluster consisting of T. forsythia, P. gingivalis, T. socranskii, Staphylococcus aureus, Staphylococcus anaerobius, Streptococcus intermedius, and Streptococcus mitis was reported to comprise 30% of the total microbiota at peri-implantitis sites (Persson & Renvert 2014).
Several recent publications have compared the microbiota in peri-implantitis and periodontitis sites using sequences of the 16S rRNA gene for the detection of bacterial species (Koyanagi et al. 2010, 2013; Cortelli et al. 2013; Dabdoub et al. 2013; Tamura et al. 2013; Maruyama et al. 2014). Koyanagi et al. (2010, 2013) found the microbiota at peri-implantitis sites to be more diverse than at teeth with periodontitis. In 81 subjects, the microbiota at neighboring teeth and implants were compared. Twelve sites demonstrated periodontitis and 20 sites peri-implantitis. A total of 523 species were identified. It was reported that 60% of individuals shared <50% of all species between their periodontal and peri-implant biofilms and 85% of individuals shared <8% of abundant species between tooth and implant. Staphylococcus and Treponema were found to be significantly associated with diseased implants, but not teeth. The authors concluded that periodontitis and peri-implantitis represent microbiologically distinct ecosystems (Dabdoub et al. 2013). Cortelli et al. (2013) reported a higher bacterial frequency at teeth and that the bacterial frequency was higher in peri-implantitis sites than healthy sites (P < 0.05). Tamura et al. (2013) using the 16S rRNA gene for the analyses found that peri-implantitis sites had approximately 10-fold higher mean colony-forming units than healthy implant sites. They concluded that periodontopathic bacteria are not the only periodontal pathogens active in peri-implantitis. A total of 235 species was identified by Maruyama et al. (2014) using sequences for the 16S rRNA gene for the analyses of the microbiota. They reported P. gingivalis, T. denticola, and T. forsythia to be prevalent in most samples in both diseases, but compared with periodontitis, peri-implantitis-associated bacterial communities had significantly higher levels of the genera Olsenella, Sphingomonas, Peptostreptococcus, and unclassified Neisseriaceae.
Table 2. Biofilm as a risk indicator peri-implantitis clinical studies
Discussion
Based on the selected papers on the presence of a subgingival pathogenic microbiota as a risk indicator for peri-implant disease, it seems appropriate to conclude that the microbiota at the diseased teeth and implants demonstrate a complex and predominant anaerobic microbiota. This conclusion is in concordance with the results from a recent review by Mombelli & Decaillet (2011). Some of the selected papers in the present review have analyzed pathogens previously associated with periodontal disease using the DNA–DNA hybridization technique (Hultin et al. 2002; Casado et al. 2011; Ebadian et al. 2012; Cortelli et al. 2013), limiting the possibility to find species not considered to be periodontal pathogens. Others have, using the same technique, analyzed a broader panel of microorganisms (Renvert et al. 2007; Shibli et al. 2008; Persson & Renvert 2014), increasing the potential to find also non-periodontitis-associated microorganisms in the samples. Generally, higher amounts of microorganisms are reported from sites diagnosed as peri-implantitis compared to healthy sites (Hultin et al. 2002; Shibli et al. 2008; Ebadian et al. 2012; Cortelli et al. 2013) and when the analysis encompassed a broader panel of microorganisms, species not associated with periodontitis were commonly found (Renvert et al. 2008; Persson & Renvert 2014). The presence of microorganisms not primarily associated with periodontitis such as Staphylococcus aureus and Enterobacteriaceae and Candida albicans has also been reported as frequent findings at implants with peri-implantitis in studies using culturing techniques for analysis (Leonhardt et al. 1999; Botero et al. 2005; Albertini et al. 2014). As titanium implants may provide an ecological niche different from that of teeth, the biofilm formed may include other and possibly currently unknown bacteria. Recent papers employing the 16S rRNA gene clone library to make a broader analysis of the microbiota indicate that periodontitis and peri-implantitis may to some degree share microorganisms, but also that the microbial composition between the diseases differ (Koyanagi et al. 2010, 2013; Tamura et al. 2013; Maruyama et al. 2014). Accordingly, based on papers with a broader approach to analyze samples from sites with a clinical diagnosis of peri-implantitis, it seems like the microbiota associated with peri-implantitis demonstrate distinct differences to the one seen at periodontitis. Microorganisms like Staphylococcus, Enteric rods, and Candida are often found in cases of peri-implantitis.
The significant correlation observed between parts of the subgingival microbiota in peri-implant lesions and natural teeth, however, suggests that natural teeth may act as a reservoir for the colonization of pathogens in partially edentulous patients (Botero et al. 2005; Renvert et al. 2007; Koyanagi et al. 2010, 2013; Albertini et al. 2014; Maruyama et al. 2014). The geographic proximity may not be sufficient to determine the inhabitants of the microenvironment (Dabdoub et al. 2013).
The fact that probing depth at implants has been highlighted as an ecological factor for microbial growth that may favor certain microorganisms (Renvert et al. 2007) indicates that initially deep pockets at implants may serve as a risk indicator for peri-implant disease. The number of sites with residual probing depths (≥5 mm) following periodontal therapy in periodontitis-susceptible patients supplied with dental implants has also been reported to be a risk indicator for peri-implantitis (Pjetursson et al. 2012b). As the probing depth has been reported to be of importance for the bacterial profile, it would be of importance to analyze the bacterial profile of chronic periodontitis and peri-implantitis cases selected so that they show similar probing depths and clinical signs of disease (bleeding on probing and/or pus).