The Microbiome of Peri-implantitis: Is It Unique?

Flavia R. F. Teles, DDS, MS, DMSc

A better characterization of the peri-implant microbiome can improve the understanding of the etiology of peri-implant diseases. Ultimately, more detailed information about the peri-implantitis microbiome will lead to better strategies for prevention, supportive therapy, and risk assessment, as well as early diagnosis of peri-implantitis and timely intervention, all of which are critical for the longterm retention of implants.

Dental implants have become an increasingly accessible therapeutic alternative for the reestablishment of function and esthetics in a patient’s dentition. According to the American Academy of Implant Dentistry, more than 3 million Americans have dental implants, a number that is increasing by 500,000 per year. Thus, it is not surprising that the US and European markets for dental implants are expected to reach $4.2 billion by 2022.¹ However, the widespread use of implants has also led to an increase in the number of cases of biofilm-mediated peri-implant diseases, particularly peri-implantitis. Although long-term longitudinal studies indicate that implant therapy has success rates of 95% to 99%,^2,3 recent publications have shown a prevalence of peri-implantitis of at least 20%.⁴ The major challenge brought by the increasing prevalence of peri-implantitis is the lack of effective treatments. The limited success might be explained, in part, by the assumption of similarities in the etiology and pathogenesis of periodontal and peri-implant diseases. Yet, implant sites represent specific environments, given the presence of artificial surfaces (ie, implant, cement, crown). Furthermore, implant sites may also be influenced by the patient’s medical/dental history (eg, smoking, diabetes, periodontitis)⁵ and the type of reconstruction (full or partial). Finally, fundamental differences exist between periodontal and peri-implant sites regarding their anatomy, ability to respond to bacterial challenge, and histopathologic presentation.^6,7

The first investigations of the peri-implant microbiome date from the 1980s and utilized primarily microbial cultivation and microscopy.^8-10 Early observations indicated that diseased implant sites harbored motile rods, fusiform bacteria, and spirochetes, while samples from successful implants contained only a small number of coccoid cells and very few rods. The microbiota of diseased implant sites consisted of primarily Gram-negative anaerobic rods, with Fusobacterium species and Prevotella intermedia being regularly detected, often found at high levels. In contrast, samples from healthy implant sites yielded very low cultivable counts and consisted predominately of Gram-positive cocci.^11,12 In addition, it became increasingly clear that peri-implant infection was associated with a distinct microbiota when compared with implant failures due to trauma. Samples from implants failing with infection revealed high levels of spirochetes and motile rods and harbored many periodontopathic species, including Parvimonas micra, Fusobacterium species, enteric Gram-negative rods, and yeasts. In contrast, implants failing from suspected traumatic etiology demonstrated a morphotype profile consistent with periodontal health and cultivable microbiota, predominated by streptococci.¹³ During the 1990s and early 2000s, additional cultivation studies indicated that failing implants showed a complex microbiota comprising oral pathogenic species, such as Porphyromonas gingivalis, Campylobacter rectus, and P. intermedia, as well as Aggregatibacter actinomycetemcomitans and Capnocytophaga species, albeit in lower levels. In addition, non-oral organisms, such as staphylococci, as well as yeasts, appeared to be common features of the peri-implantitis microbiota.^14-16

The late 1990s and early 2000s marked a transition from culture-based studies to cultivation-independent investigations. Even though cultivation studies had provided valuable foundational knowledge of the microbiota associated with peri-implant health and disease, they were difficult to conduct. They required time, cost, and expertise, thereby limiting the number of patients and samples that could be feasibly analyzed. In addition, such studies required that bacterial cells remained viable between the collection of clinical samples and their processing at the laboratory. This aspect was further complicated by the fact that certain organisms are more fastidious than others, require varying growth conditions, or simply could not be cultivated. Hence, the checkerboard DNA-DNA hybridization technique quickly became the mainstay method for studies of the peri-implant microbiota. It provided a cost-effective platform for the detection and enumeration of the 40 most common periodontal bacterial species in multiple samples simultaneously. Most importantly, its extensive use worldwide in studies of the periodontal microbiome and oral microbial ecology generated a solid conceptual framework of the microbial basis of periodontal health and disease.¹⁷

Using checkerboard DNA-DNA hybridization, Salcetti et al¹⁸ found greater detection frequencies of Prevotella nigrescens, P. micra, Fusobacterium nucleatum ss vincentii, and F. nucleatum ss nucleatum in plaque samples collected from failing implant sites as compared with healthy peri-implant sites. Hultin et al¹⁹ noted that peri-implantitis sites harbored high levels of periodontal pathogens, A. actinomycetemcomitans, P. gingivalis, P. intermedia, Tannerella forsythia, and Treponema denticola. Those findings were supported by results by Shibli et al.²⁰ The authors found significantly higher mean counts of P. gingivalis, T. denticola, and T. forsythia were observed in supra- and subgingival biofilm samples from peri-implantitis sites when compared with controls. The proportions of the pathogens from the red complex were elevated, while host-compatible microbial complexes were reduced in diseased, compared with healthy implants.

Another advantage of the checkerboard technique was its flexibility, in that it allowed for the development of whole genomic probes for bacteria other than the typically employed 40 periodontal species, thus permitting the investigation of oral taxa beyond the “standard” checkerboard probe panel and non-oral and medically important species. For the exploration of the role of “alternative” species in peri-implantitis, Persson et al²¹ utilized an expanded checkerboard DNA-DNA hybridization assay encompassing 79 different microorganisms. Yet, the most prevalent bacteria included oral taxa F. nucleatum, T. forsythia, and A. actinomycetemcomitans, but also non-oral species such as Helicobacter pylori and staphylococci. The group expanded the work in a subsequent study,²² in which microbiologic samples were collected from 166 patients with peri-implantitis and 47 individuals with healthy implants and analyzed by DNA-DNA checkerboard hybridization targeting 78 species. Nineteen bacterial species were found at significantly higher counts in peri-implantitis samples, including red (P. gingivalis, T. forsythia, T. denticola) and orange (C. rectus, Campylobacter showae, Campylobacter gracilis) complex species, A. actinomycetemcomitans and Treponema socranskii, and non-oral species H. pylori, Haemophilus influenzae, Staphylococcus aureus, and Staphylococcus anaerobius, suggesting a potential role for bacteria typically detected in infections of implanted medical devices.²³

Despite the advances warranted by checkerboard studies, they still had limitations, in that the technique represents a “close-ended” approach; it is limited to the detection of the taxa targeted by the probes used. Since it was realized that the oral cavity may harbor more than 700 taxa,²⁴ 35% of which have never been cultivated,25 it became clear that a substantial (and not less relevant) portion of the peri-implant microbiome was being systematically ignored. Thus, the late 2000s and early 2010s saw the rise of the use of open-ended techniques that focus on the amplification and sequencing of the 16S rRNA gene. This gene is a microbial marker with valuable properties: it is ubiquitous across all bacterial species and has conserved and variable regions that are ideal for specific and sensitive polymerase chain reaction (PCR) amplification; full sequences became increasingly available in curated databases.²⁶ Hence, 16S rRNA sequencing approaches are, at least in principle, agnostic and “unbiased” tools to study “all” the bacterial taxa present in a sample, regardless of their viability and culturability. Using cloning and sequencing of the 16S rRNA gene, the first iteration of this approach, Porphyromonas, Fusobacterium, and Filifactor species were shown to be abundant members of the peri-implant microbiome.^27,28 However, because of the laborious nature of this approach, it was soon surpassed by high-throughput next-generation sequencing (NGS) approaches.²⁹

In the first study to examine the peri-implant microbiome using NGS, Kumar et al³⁰ found that peri-implantitis sites harbored lower levels of Prevotella and Leptotrichia and higher levels of Actinomyces, Peptococcus, Campylobacter, non-mutans Streptococcus, Butyrivibrio, and Streptococcus mutans than healthy implants. A subsequent paper from the same group³¹ reported that Staphylococcus were significantly associated with implant infection and that red complex pathogens were found in only 37% of the peri-implantitis biofilms. Collectively, those findings are a clear departure from previous studies of the peri-implant microbiome; the reasons for that may include the use of distinct sample collection methods and the nature of the platforms employed. However, one cannot discard the possibility that, by being the first study of its kind, the platform, analytical pipeline, and reference database were in their infancy and thus more prone to errors. In fact, years later, using a similar platform, Mayurama and co-workers³² reported that T. forsythia, P. gingivalis, T. denticola, F. nucleatum, and T. socranskii were abundant and prevalent in most samples of peri-implantitis. NGS studies also allowed for the evaluation of the impact of smoking, a well-known risk factor for peri-implantitis,⁵ on the peri-implant microbiome. As demonstrated by Tsigarida et al,³³ smoking shapes the peri-implant microbiome even in clinical health, promoting a pathogen-rich community depleted of commensals.

Data from 11 studies selected in a recent systematic review aimed at determining the weight of evidence for microorganisms related to peri-implantitis concluded that the main bacterial species associated with peri-implantitis are recognized as periodontal pathogens (P. gingivalis, T. denticola, and T. forsythia) and pathogenic species (P. intermedia, C. rectus). These findings are supported by one of the author and colleagues’ recent studies,³⁴ where the microbiome of healthy (H, n = 32) and diseased (P, n = 35) peri-implant sites were compared and the core peri-implant microbiome determined using 16S rRNA sequencing. A clear distinction was observed between H and P samples at all taxonomic levels. Bacteroidetes, Spirochetes, and Synergistetes were significantly higher in P, while Actinobacteria prevailed in H. Porphyromonas and Treponema were significantly more abundant in P, and Rothia and Neisseria were higher in H. The core peri-implant microbiome contained Fusobacterium, Parvimonas, and Campylobacter, genera often associated with periodontal inflammation. T. denticola and P. gingivalis levels were significantly higher in P, as well as Filifactor alocis, Fretibacterium fastidiosum, and Treponema maltophilum. Collectively, the results indicate that the peri-implantitis microbiome is commensal-depleted and pathogen-enriched, harboring traditional pathogenic taxa (red and orange complex species) and newly proposed pathogens, such as F. alocis and F. fastidiosum and yet-uncultured organisms such as Desulfobulbus sp HOT 041.³⁵

Previous studies of the peri-implant microbiome have provided valuable insight into the contributions of the local microbial insult to peri-implant diseases. However, collectively they present limitations that need to be taken into consideration for the interpretation of their results and the design of future studies. The cross-sectional design, the need for a proper selection of cases and controls and incorporation of clinical variables, the heterogeneity in sample collection procedures, microbial processing, and analytical strategies are all examples of issues that compromise the ability to obtain reliable and generalizable information to advance the field and better inform clinical decisions.

Because of the increasing cost-effectiveness, 16S rRNA sequencing plat-forms have become the standard for studies of the peri-implant microbiome. However, the platforms are not without limitations. Even though 16S rRNA gives a broad overview of samples’ composition, it may not distinguish closely related species. Furthermore, this technique ignores most of the microbial genetic information contained in the sample because it focuses on a portion of one bacterial gene. Recent advances in the field expanded the portfolio of “omics” technologies available for the study of the oral microbiome.³⁶ It is now possible to study the full genomic potential of the local microbiome, via metagenomics studies, and explore not only “who is there,” but also “what are they doing,” by using metatranscriptomics,³⁷ proteomics, and metabolomics. These methods have recently provided valuable insights into the pathogenesis of periodontitis and promise to significantly enhance understanding of peri-implant diseases and unravel new strategies for prevention and treatment of these conditions.

This report suggests the microbiome associated with peri-implantitis resembles the reported periodontitis microbiota. It shows an enrichment of well-recognized pathogens as well as newly proposed pathogenic microorganisms, several of which have not yet been cultivated, at the expense of a depletion of host-compatible species. Future studies should be designed with proper selection of cases and controls and incorporation of state-of-the-art “omics” approaches, such as metagenomics, metatranscriptomics, proteomics, and metabolomics.

Flavia R. F. Teles, DDS, MS, DMSc
Research Associate Professor
Department of Periodontology
School of Dentistry
University of North Carolina at Chapel Hill
Chapel Hill, North Carolinas