The Microbiome of Peri-Implantitis: A Systematic Review of Next-Generation Sequencing Studies

(1) Introduction: Current evidence shows that mechanical debridement augmented with systemic and topical antibiotics may be beneficial for the treatment of peri-implantitis. The microbial profile of peri-implantitis plays a key role in identifying the most suitable antibiotics to be used for the treatment and prevention of peri-implantitis. This systematic review aimed to summarize and critically analyze the methodology and findings of studies which have utilized sequencing techniques to elucidate the microbial profiles of peri-implantitis. (2) Results: Fusobacterium, Treponema, and Porphyromonas sp. are associated with peri-implantitis. Veillonella sp. are associated with healthy implant sites and exhibit a reduced prevalence in deeper pockets and with greater severity of disease progression. Streptococcus sp. have been identified both in diseased and healthy sites. Neisseria sp. have been associated with healthy implants and negatively correlate with the probing depth. Methanogens and AAGPRs were also detected in peri-implantitis sites. (3) Methods: The study was registered with the International Prospective Register of Systematic Reviews (PROSPERO) (CRD42023459266). The PRISMA criteria were used to select articles retrieved from a systematic search of the Scopus, Cochrane, and Medline databases until 1 August 2023. Title and abstract screening was followed by a full-text review of the included articles. Thirty-two articles were included in the final qualitative analysis. (4) Conclusions: A distinct microbial profile could not be identified from studies employing sequencing techniques to identify the microbiome. Further studies are needed with more standardization to allow a comparison of findings. A universal clinical parameter for the diagnosis of peri-implantitis should be implemented in all future studies to minimize confounding factors. The subject pool should also be more diverse and larger to compensate for individual differences, and perhaps a distinct microbial profile can be seen with a larger sample size.


Introduction
Dental implants exhibit high success rates of up to 97% and above [1].However, contributory factors related to occlusal overloading and peri-implant tissue infection may lead to implant failure [2].Peri-implantitis is defined as an infection of the peri-implant tissues accompanied by suppuration and clinically significant progressive crestal bone loss after the adaptive phase, leading to decreased osseointegration and pocket formation [3,4].Peri-implantitis has a reported prevalence ranging from 6.6% to 51% [5][6][7][8][9].Various risk factors are associated with an increased risk of peri-implantitis.Prosthetic factors, including convex emergence profiles, submucosal crown margins, and excess cement in cemented implant prostheses, increase the risk of peri-implantitis [2,3].Systemic conditions such as diabetes mellitus and osteoporosis also increase the risk of peri-implantitis [10].Furthermore, smoking has been found to directly affect the bone surrounding the implant, thereby increasing the risk of peri-implantitis as well [11].Biofilm removal and control with instruments such as Gracey curettes, ultrasonic scalers, and air powder abrasive devices Antibiotics 2023, 12, 1610 2 of 30 have been employed with questionable success in the treatment of peri-implantitis since mechanical debridement also comes with its challenges, especially at the apically facing thread surfaces, as demonstrated by Steiger-Ronay et al. [12].Antimicrobials are also ineffective if mechanical debridement is inadequately performed, as mentioned previously [13,14].However, liquid desiccants have been reported to reduce the anaerobic bacteria load in diseased implants [15].To date, the treatment of peri-implantitis is similar to that of periodontitis [16].The prognosis of this condition is uncertain, and hence, determining the fundamental cause is important for preventive strategies and also targeted approaches [17].
The exact mechanism of microbial interaction in peri-implantitis is not clearly known [3].Initial studies reported that Staphylococcus aureus plays a role in the progression of the disease [18,19].However, the consensus on the predominance of S. aureus in peri-implantitis sites was contradicted by Belibasakis et al., as their study concluded the predominance of Treponema spp.and Synergistetes cluster A in peri-implantitis sites [19,20].
Koyanagi et al. reported a more diverse microbial profile compared to that of periodontitis [21], while other studies indicated similarity [22,23].A microbial profile consisting of aggressive and resistant microorganisms distinct from periodontitis has also been reported previously [24].Periodontally involved teeth act as reservoir for periopathogens which translocate to the implant sites, making chronic periodontitis an important risk factor for peri-implantitis [21,23,25,26].
Culture-dependent studies evaluating the microbiome of peri-implantitis have limited insights into the bacterial community [27,28], and more recent next-generation sequencing techniques may give us an insight into a more targeted approach to peri-implantitis treatment which, in turn, can improve the prognosis of this condition [29].The use of next-generation sequencing allows the identification of non-culturable species as compared to conventional methods [29].The detection of bacterial and fungal infections has been shown to be consistently accurate as compared to conventional methods [30].In addition, next-generation sequencing has been shown to be cost-effective for identifying the disease with a given high pretest probability, as compared to culture methods [31].
This systematic review aims to summarize and critically analyze the methodology and findings of studies that have utilized next-generation sequencing techniques to elucidate the microbial profiles of peri-implantitis.

Results
From the initial search, 506 articles were identified after the elimination of duplicates.After performing the preliminary review of the title and abstracts, 32 articles were included for full-text screening.Based on the selection criteria, 32 studies were chosen to be included in the qualitative analysis (Figure 1).The Risk Of Bias In Non-randomized Studies-of Exposures (ROBINS-E) assessment of 32 articles is shown in Table 1.The Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach was used (Table 2) and revealed a low certainty of evidence for the outcomes of diversity and richness as well as the abundance of taxa.a .Out of the 32 studies reviewed, nine were of some concern, while four were at a high risk of bias based on the ROBINS-E assessment tool.b .Inconsistency is seen due to the heterogeneity across all 32 studies.c .Indirectness is seen due to the differences in the severity of peri-implantitis.Microbial compositions of different severities present heterogenous results.
The DNA extraction technique, sequencing technique, targeted region, and the reference database for each study are summarized in Table 4.The microbiome profile is depicted in relation to the diversity, richness, and taxa abundance in Table 5.
Among the 32 studies reviewed, seven studies found an increase in the microbial diversity of peri-implantitis sites as compared with healthy implant sites [20,23,33,38,43,44,52].Five studies did not report the diversity and richness of the samples collected [41,46,51,55,56,58].Five studies reported an increase in the microbial diversity in peri-implantitis sites as compared with periodontitis sites [21,32,35,39,57].Five studies reported a reduced microbial diversity in peri-implantitis sites compared with healthy implants in subgingival plaque [22,34,44,45,52].Additionally, four studies reported no significant difference in diversity between healthy implants and peri-implantitis samples [23,33,37,50].[21,52,56].Sanz-Martin et al. reported higher levels of Eubacterium in a healthy implant, when a diseased implant was also present in the same oral cavity [20].Two studies found high levels of Bacteroidetes and Firmucutes in PI sites as compared to HI sites [20,46].Three authors found higher levels of Bacteroides in diseased implants [32][33][34].Yu et al. demonstrated that F. fastidiosum SH03 and the Fretibacterium oral taxon SH01 were linked with plaque at healthy subgingival sites [48].This study concluded that there were no clear differences or similarities between Synergistetes communities found in diseased versus healthy sites or between periodontal/subgingival niches and peri-implant/submucosal niches [48].Another study by Yu et al. also showed that the prevalent and abundant bacteria were Streptococcus infantis/mitis/oralis (HMT-070/HMT-071/HMT-638/HMT-677) and Fusobacterium sp.HMT-203/HMT-698 in healthy implants and diseased implants [42].Another 18 phyla were found in low abundance, particularly the Aquificae, Chlamydiae, Gemmatimonadetes, Nitrospirae, TM6, Verrucomicrobia, and WPS2 phyla, which were present in <0.01% of the total reads for each of the four clinical site categories, with some being undetectable in one or more niches [42].Healthy implants demonstrated higher proportions of Actinomyces, Atopobium, Gemella, Kingella and Rothia and lower levels of Campylobacter, Desulfobulbus, Dialister, Eubacterium, Filifactor, Mitsukella, Porphyromonas, and Pseudoramibacter in one study [56].One study that underwent a pathogen-specific analysis for Archaea found that PI sites had a higher frequency of sites that were positive for Archaea [58].Filifactor was found to be abundant in peri-implantitis sites when compared with healthy implant sites, as shown by several studies [20,35,36,40,47,55,56].Three studies demonstrated that Parvimonas was the most abundant at peri-implantitis sites [21,55,57].

Phyla
The range of phyla was reported to be varied among the 25 studies.Koyanagi T et al. reported that Firmicutes (45.6%) is the most abundant phylum found in the subgingival plaque in peri-implantitis samples, followed by Bacteroidetes, Proteobacteria, Fusobacteria, Actinobacteria, TM7, Synergistetes, Spirochaetes, Tenericutes, Chloroflexi, and Deferribacteres [21].Three studies were in concordance in concluding that Bacteroidetes is one of the genera that is found in great abundance in peri-implantitis samples [20,21,46].The abundance of Synergistetes was reported to be higher in diseased samples in four studies in comparison to in healthy samples [20,21,23,33].Spirochaetes was identified in diseased samples in three studies [20,21,46], with one study reporting that Spirochaetes increased significantly as peri-implantitis became more severe [20].

Genus
Numerous changes were reported at the genus level (Table 5), with many of them focusing on several genera which are the most abundant in the peri-implant sites.One study reported that there was a preponderance of Veillonella in diseased peri-implant mucosal tissues [45].However, there are also studies that have suggested that Veillonella is significantly reduced in samples with an increasing peri-implantitis severity [20,53].Veillonella was also associated with healthy implant sites in other studies [20,47,55,56].Several authors have found that Prevotella spp.are significantly more abundant at peri-implantitis sites [23,34,36,39,53,54].Kumar et al. and Daubert et al. found that healthy implants showed higher levels of these two microorganism species [22,45], which was also supported by Apatzidou et al., who showed their greater abundance in diseased samples [23].Other than Veillonella and Prevotella, most studies also pointed out that Porphyromonas was commonly associated with diseased implants [20,23,51,53,56].Several studies pointed out that Fusobacterium was present in high levels in peri-implantitis samples [21,37,41,46,[55][56][57]. Five studies reported that Streptococcus was more abundant in healthy plaque samples as compared to its abundance in diseased samples [20,22,23,44,45].Yu et al. also found that Streptococcus was found in both healthy implants and peri-implantitis sites [42].On the contrary, Kumar et al. concluded that peri-implantitis samples demonstrated a higher level of Streptococcus [22].A study reported that Propionibacterium, Paludibacter, Staphylococcus, Filifactor, Mogibacterium, Bradyrhizobium, and Acinetobacter are unique to peri-implant sites [47].In addition, Actinomyces spp.has been reported to be prevalent in peri-implantitis sites [22,52,53].However, da Silva et al. reported higher levels of Actinomyces spp. in healthy implants [56].

Microbiome Complex
Apart from the genera and phyla levels, Al-Ahmad et al. and Kim et al. reported that Porphyromonas gingivalis and Tannerella forsythia of the red complex are highly associated with peri-implantitis [32,46].A study reported positive correlations with certain red and orange complex bacteria but a negatively correlation with blue complex bacteria in periimplantitis samples [20].Furthermore, another study reported that Bacteroidetes, Chloroflexi, Spirochaetes, Synergistetes, and TM7 positively corresponded with the pocket depths [23].

Peri-Implantitis with Periodontitis
Granulicatella adiacens (phylum Bacillota) was identified in two-thirds of peri-implantitis sites; these two species were also detected at periodontitis sites but not in healthy implants [57].Shiba et al. found that the microbial composition at the genus level was diverse among the samples for each disease and between both samples from each individual, although the predominant species were similar [49].Two studies showed that the periodontitis microbial community is more diverse than peri-implantitis sites [25,47].Interestingly, three studies found the opposite, whereby periodontitis samples yielded lower diversities than peri-implantitis samples [21,22,57].Aleksandrowicz et al. demonstrated that Archaea was found in diseased implants and teeth [41].Furthermore, they were found in abundant levels at periodontitis sites when compared to peri-implantitis sites [41].

Peri-Implantitis with Peri-Implant Mucositis
Shi et al. reported no differences in diversity between peri-mucositis sites as compared to peri-implantitis sites, but they found an increased microbial richness in peri-mucositis sites [36].Sousa et al. reported a decreased abundance of Bradyrhizobium in peri-mucositis sites and peri-implantitis sites [47].One study concluded that the microbial profile associated with peri-implantitis was also present with a moderate relative abundance at peri-mucositis sites.This study also found that the Shannon index of peri-mucositis was lower than that of peri-implantitis [52].Tsigarida et al. reported subtle differences between the peri-mucositis and peri-implantitis microbiomes, and these subtle differences were between the transition from health to disease [50].Streptococci and Rothia were associated with peri-mucositis, while Fusobacterium and Treponema were associated with peri-implantitis, as shown by Polymeri et al. [37]

Heterogeneity of Studies
Significant heterogeneity can be identified in the methodologies of the selected studies.The ROBINS-E tool was used to assess the quality of the 32 nonrandomized cohort observational studies.The ROBINS-E tool (Table 1) showed that nine studies had some concerns, while four studies were at a high risk of bias.Table 4 illustrates the heterogenicity of the gene sequencing techniques utilized.Figure 2 illustrates the diversity reported in terms of the Shannon's indexes reported by five studies [21,25,36,37,57].Figure 3 illustrates the heterogeneity regarding the location (Figure 3a), database used (Figure 3b), and case definition criteria (Figure 3c) of the studies reviewed.heterogenicity of the gene sequencing techniques utilized.Figure 2 illustrates the diversity reported in terms of the Shannon's indexes reported by five studies [21,25,36,37,57].Figure 3 illustrates the heterogeneity regarding the location (Figure 3a), database used (Figure 3b), and case definition criteria (Figure 3c) of the studies reviewed.heterogenicity of the gene sequencing techniques utilized.Figure 2 illustrates the diversity reported in terms of the Shannon's indexes reported by five studies [21,25,36,37,57].Figure 3 illustrates the heterogeneity regarding the location (Figure 3a), database used (Figure 3b), and case definition criteria (Figure 3c) of the studies reviewed.

Discussion
This systematic review comprehensively reviews the current available evidence on the microbiome of peri-implantitis.Variations in the study methods, sample collection, and study design were observed.However, the review focuses on studies employing the 16S r RNA gene sequencing technique to summarize meaningful observations from the available evidence.
Ten of the studies reviewed showed that the microbial diversity of peri-implantitis is distinct and usually higher than that at healthy implant sites [14,15,17,19,24,26,28,34,38,39].The alpha diversity considers the richness (number of taxa) and evenness (relative abundance) of species within a sample/community; the beta-diversity quantifies the identities of taxa involved between samples/communities [49].Changes

Discussion
This systematic review comprehensively reviews the current available evidence on the microbiome of peri-implantitis.Variations in the study methods, sample collection, and study design were observed.However, the review focuses on studies employing the 16S r RNA gene sequencing technique to summarize meaningful observations from the available evidence.
Ten of the studies reviewed showed that the microbial diversity of peri-implantitis is distinct and usually higher than that at healthy implant sites [14,15,17,19,24,26,28,34,38,39].The alpha diversity considers the richness (number of taxa) and evenness (relative abundance) of species within a sample/community; the beta-diversity quantifies the identities of taxa involved between samples/communities [49].Changes in oxygen and nutrient concentrations associated with the deepening of a pocket around an implant may be responsible for the shift in the microbial diversity [32].Figure 2 shows the Shannon's indexes reported by five studies, as not all studies reported indices [21,25,36,37,57].These variations in the diversity can be explained by the heterogenicity of various factors such as the location of the study (Figure 3a), the reference database (Figure 3b), and the case criteria definition (Figure 3c).A variation in the genomic database can introduce conflicting results, as one study showed that even the use of a single database within a study can implicate systematic errors during the mapping process which subsequently affects genomic analyses [59].In addition to that, the sample collection method and the type of sample collected are other confounding factors that may produce conflicting findings.
The studies that included in the current review originate from different countries (Figure 3a), for example, Japan [21,49,55,57], China [36,42,48,52,60], United States of America [22,25,45,50], United Kingdom [47], Germany [38,43,46,53], and The Netherlands [37].It is significant to note that certain sections of the globe are not represented here.This may also be due to the exclusion of articles written in other languages.Hence, the current data may be significantly influenced by the diet and genetic make-up of the individuals from the representative countries [61].The characterization of oral dysbiosis in different ethnicities and races presents significant challenges due to variations across multiple studies [62][63][64].This is due to the highly varied diet, nutrition and lifestyle practices present over several generations in different geographical locations [65,66].
The case definition for peri-implantitis varied across the studies reviewed (Figure 3c).For example, Koyanagi et al. used a criteria of a probing depth (PD) ≥5 mm with bleeding on probing (BOP) and/or suppuration and bone loss >3 threads up to half of the implant length, while Apatzidou et al. diagnosed subjects as having peri-implantitis when there was PD ≥ 6 mm, BOP and/or suppuration, and radiographic bone loss of ≥2 mm in at least one implant surface after one year of loading [21,23].However, it is evident that the disease severity may vary, even with the employment of the above criteria, hence making it difficult to combine or compare the results of certain studies.Standardizing the methodological quality of microbiome studies has been previously suggested as a necessary step in this direction.
Even though few studies included criteria related to the systemic status of the patient, drugs taken, previous history of other oral diseases like periodontitis and the age of the patient into consideration, the varied criteria set across studies makes a meaningful comparison irrelevant.It would be greatly beneficial for future investigations into the microbiome of the oral cavity to follow a standardized protocol to facilitate comparability between studies [67].
Despite being considered an extension of peri-implantitis and the presence of common bacteria, peri-implant mucositis has been reported to have a distinct microbial profile in some studies [68,69].However, a few studies were not able to provide a conclusive result on this aspect [36,37,47,50,52].The diversity in peri-implant mucositis has been reported to be higher than at healthy implant sites [36] but lower than in peri-implantitis [52].Moreover, the immune cell profiles of both entities seem to differ as well.Enhanced neutrophil and B-cell responses have previously been identified for peri-implantitis lesions when compared to peri-implant mucositis lesions under experimental conditions.The shift in the microbiome profile may also be explained by the increase in frequency and the number of bleeding sites subsequent to biofilm accumulation surrounding the implants [70].
The association of Veillonella sp. with healthy implant sites is well-correlated with its reduced prevalence in deeper pockets and severe disease progression [20,43,46,55].Streptococci spp.have been identified in both diseased [21,22,53,56] and healthy sites [20,23,45].Neisseria sp. have been associated with healthy implants and negatively correlates with the probing depth [20,40,43,44], suggesting that Neisseria sp. could have been replaced by other colonizers or may exert a protective effect.Species of the genus Neisseria are well-established primary colonizers of the dental plaque of natural teeth but are not well known for their presence in dental implants.On the contrary, three studies reported high levels of Neisseria sp. in peri-implantitis sites, which contradicts other studies [22,51,54].Considering the common occurrence of these species in the oral cavity and the possibil-ity of transfer from a diseased to a healthy site or vice versa leads to the lack of a clear understanding of its role in the initiation and the progression of the disease.
Numerous studies have identified Fusobacterium sp. as the dominant species in periimplantitis [20,21,46].Studies have also reported the presence of the genus Treponema at peri-implantitis sites of increasing severity [20,43].However, Kumar et al. reported higher levels of the genera Treponema and Prevotella at healthy implant sites, which is the opposite to what other studies have found [22].Peri-implantitis sites have also seen an abundance of species from the phylum Synergistetes [20,23,46].Porphyromonas sp. have been reported at peri-implantitis sites by multiple studies [20,21,23].
A distinct microbial pattern could not be identified across all the 25 studies reviewed, possibly due to the abovementioned factors.Sahrmann et al. also found that there was an absence of a characteristic bacterial profile at peri-implantitis sites [71].Both the current review and the review by Sahrmann et al. had a consensus that there was considerable heterogeneity in the studies reviewed [71].The red complex is frequently identified at periimplantitis sites, as are putative pathogens of the orange and yellow complex.Furthermore, it seems that the relative abundance of each complex changes with an increasing disease progression severity.The blue complex was also reported to be negatively correlated with peri-implantitis sites, suggesting its protective effect.The red complex was also more abundant at implant sites for subjects who smoked, which correlates well with our current understanding that smoking is a risk factor for peri-implantitis.The studies have findings that contradict one another, and this makes it difficult to obtain a characteristic microbial profile for peri-implantitis.However, it is evident that the microbiome of peri-implantitis is unique and distinct from that of periodontitis.
Carvalho et al. found that peri-implantitis lesions were associated with the presence of S. epidermidis, P. gingivalis, T. forsythia, T. denticola, F. nucleatum, and P. intermedia [72].The review included culture-dependent studies in the analysis.On the contrary, the current systematic review only included studies that utilized next-generation sequencing due to its improved detection limit [30,73].Additionally, Carvalho et al. reported that a definitive conclusion regarding the microbiome of peri-implantitis could not be reached due to the nature of the studies analyzed.Next-generation sequencing methods have shown that the microbiome of peri-implantitis is distinct from that of periodontitis.Non-culturable species such as Fusobacterium and the Treponema sp.HMT-257 have been detected in peri-implantitis lesions [74,75].The current systematic review demonstrates that, even with the inclusion of only next-generation sequencing studies, a distinct and unique microbial community pattern could not be identified.
The current review is limited by the studies' number of participants, with the highest being 139 in a study by Aleksandrowicz et al. [41].This suggests that the results may not be generalized to the clinical setting due to the small sample size.This review is also limited by the heterogeneity presented across all studies reviewed.Hence, a characteristic microbial profile cannot be determined for future targeted therapies.

Materials and Methods
A systematic review of observational and case-control studies (PROSPERO) (CRD42023459266) investigating the microbiome of peri-implantitis lesions was performed on the Cochrane, Medline, and Scopus databases from inception until 1 August 2023 and reported according to the Preferred Reporting Items for Systematic Reviews and Meta-Analysis (PRISMA) [76].A focused question was formulated based on PECO (population, exposure, comparator, and outcome).The population included patients with at least one osseointegrated dental implant, the exposure was the diagnosis of peri-implantitis lesions, the comparator included healthy implants, periodontitis sites, as well as peri-implant mucositis sites, and the outcome measure was the bacterial composition obtained from samples taken from peri-implantitis sites, as assessed through next-generation sequencing.The question was as follows: Among patients with at least one osseointegrated dental implant, what would be the difference between peri-implantitis lesions, healthy implants, periodontitis, and peri-implant mucositis in terms of the bacterial composition obtained from samples as assessed via next-generation sequencing?
The search strategy involved a combination of the following key terms: peri-implantitis, inflammation, disease, infection, consequence, sequence analysis, RNA, 16S, metagenomics, metagenome, microbiota, and bacteria.The keywords were combined using the Boolean operators "AND" and "OR" in the strategic search.This systematic review followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) criteria [77].
The titles and abstracts were independently screened by two reviewers (K.C.G., R.K.M.) for eligible studies, followed by full-text reading.Data were extracted independently and in duplicate by the two reviewers (K.C.G., R.K.M.) into a data extraction form created following the Cochrane Handbook of Systematic Reviews of Interventions guidelines [76].Observational and case-control studies investigating the microbiome of peri-implant tissues through next-generation DNA sequencing methods were included.Culture-based studies, conference papers, review articles, studies regarding peri-implantitis associated with other systematic factors (diabetes mellitus, immune disorders, etc.), and articles that examined only specific microorganisms were excluded from this systematic review.Non-English language articles and research conducted on non-human specimens were also excluded.This was followed by full-text screening for eligibility.The complete search strategy used is shown in Table 6.Table 7 depicts the inclusion and exclusion criteria for the articles.

Inclusion Criteria Exclusion Criteria
Observational and case-control studies investigating the microbiome of peri-implant tissues through next-generation DNA sequencing methods.Human studies in English Culture-based studies, conference papers, review articles, studies regarding peri-implantitis associated with other systematic factors (diabetes mellitus, immune disorders, etc.) Articles that examined only specific microorganisms.Non-English language articles and research conducted on non-human specimens.
The relevant studies were assessed with the Risk Of Bias In Non-randomized Studiesof Exposures (ROBINS-E) tool [78].

Conclusions
The study of the microbiome with next-generation sequencing allows more insight into the possible casual relationships between the bacteria and diseased state and not just culturable or cultivatable species.A unique and distinct microbial pattern could not be identified due to the vast heterogeneity present across all studies.The authors propose that future studies should investigate the microbial profile of peri-implantitis based on the severity of the disease to further provide insight into the progression and alteration of the microbial community within the peri-implant pocket.
A universal clinical parameter for the diagnosis of peri-implantitis should be implemented in all future studies to minimize the confounding factors.The subject pool should also be more diverse and larger to compensate for individual differences, and perhaps, a distinct microbial profile may be seen with a larger sample size.The studies reviewed also show that different groups of bacteria exist in the pockets at different stages of the diseases.This may imply that, with a complete microbial profile, an accurate estimation of the disease progression and monitoring can be performed.Furthermore, this also allows targeted drug therapies towards selective microorganisms that are strongly associated with peri-implantitis.

Table 1 .
The Risk Of Bias In Non-randomized Studies-of Exposures (ROBINS-E) assessment.
L: low risk of bias; S: some concerns; H: high risk of bias.

Table 2 .
Grading of Recommendations, Assessment, Development, and Evaluation (GRADE) approach.

Table 3 .
Characteristics of the population and the results derived from the included studies.

Table 4 .
Summary of techniques of DNA extraction, amplification, and sequencing.

Table 5 .
Microbial profiles from the retrieved studies showing the diversity and richness and the abundance of taxa.Holdemanella and Cardiobacterium PM > PI: Oribacterium, Staphylococcus, and Ramlibacter Solobacterium moorei and Prevotella denticola P: F. nucleatum, P. stomatis and Leptotrichia sp.

Table 5 .
Cont.Koyanagi et al. revealed that implants with peri-implantitis had a higher abundance of Eubacterium spp.when compared to healthy implants, and this finding is also supported by Zheng et al. and Kroger et al. [21,43,52]; da Silva et al. found that healthy implants demonstrated lower proportions of Eubacterium compared to peri-implantitis sites, while Koyanagi et al. and Zheng et al. concluded that peri-implantitis sites had significantly higher proportions of Eubacterium PI: Peri-implantitis; HI: healthy implants; P: periodontitis; PM: peri-mucositis.a : Shannon's index; b : Chao1 index; c : Principal Coordinate Analysis (PCoA); d : permutational multivariate analysis of variance (PERMANOVA); e : InvSimpson's index; f : weighted Unifrac distance analysis; g : number of operational taxonomic units (OTUs).

Table 7 .
Inclusion and exclusion criteria used for the studies screened.