Prevention and treatment of peri-implant diseases—The EFP S3 level clinical practice guideline

Number and design of included studies

Eleven observational studies including six case–control studies and five cohort studies (Carra et al., 2023).

Risk of bias

According to the Newcastle–Ottawa scale (NOS), eight studies were at low risk of bias and three studies were at high risk of bias.

Effect sizes and their clinical relevance

Pooled data analyses revealed a significantly lower rate of peri-implantitis (OR = 0.16; 95% CI [0.03; 0.96]; p = .004; I²: 0%; analysis based on two studies including 385 implants), and significantly lower marginal bone level changes over time (−0.36 mm; 95% CI [−0.65; −0.07]; p < .0001; I²: 95%; analysis based on six studies including 591 implants) in patients with good glycaemic control compared with poor glycaemic control. The mean difference (MD) in PD and BOP was not significantly different between the groups. With respect to dental implant survival, diabetes patients with poor glycaemic control were found to have a 7.59 increased risk of dental implant failure compared with patients with good glycaemic control (OR = 7.59; 95% CI [1.63; 35.3]; p = .01; I²: 0%; based on two studies including 524 implants). The estimated mean implant survival was 99% (95% CI [97.8%; 100%]; based on five studies including 253 dental implants) in patients with good glycaemic control and 95.6% (95% CI [91.4%; 99.8%]; based on five studies including 271 dental implants) in patients with poor glycaemic control.

The effect size of these findings is considered clinically relevant, but it must be highlighted that the results are based on a limited number of studies with small sample sizes, that the analyses were performed at the implant level only, and that the definition of good and poor glycaemic control was not consistent among the studies (i.e., good glycaemic control was defined as HbA1c between 6.1% and 8% in five studies, <7% in one study, and <6% in another study; poor glycaemic control was defined as HbA1c level ranging between 8.1% and 10% in 5 studies, as HbA1c > 8% in one study, and as HbA1c ranging between 7% and 9% in another study; three studies also included a group of very poorly controlled type-2 diabetes patients, as HbA1c > 9 or >10%).

Consistency

Consistency was found in the overall results, favouring good glycaemic control over poor glycaemic control. However, the definition of good and poor glycaemic control was not consistent among the available studies.

Balance of benefit and harm

Not assessed. However, glycaemic control in patients with diabetes is advised independently of implant therapy.

Overall certainty of the evidence

No study provided direct evidence. The results are inferred from studies with various working hypotheses that were not originally developed to test the effectiveness of a preventive measure on the occurrence of peri-implant diseases. Further research is needed to provide confidence in the estimated effect of glycaemic control on the risk of peri-implant diseases.

Number and design of included studies

Fourteen studies, 13 observational studies and 1 randomized clinical trial (RCT) (Carra et al., 2023).

Risk of bias

According to NOS: seven studies were at low risk of bias and six studies were at high risk of bias. According to RoB-II-RCT. one study was at some concern.

Effect sizes and their clinical relevance

Twelve studies compared patients regularly attending the recommended SPIC programme (adherent) versus non-attending patients or those attending SPIC visits irregularly. Pooled data analyses revealed that patients attending SPIC regularly were at significantly lower risk of presenting with peri-implant diseases (including both peri-implant mucositis and peri-implantitis) (OR = 0.42; 95% CI [0.24; 0.75]; p = .003; I²: 57%; analysis based on six studies including 736 patients) during the study follow-up period (ranging from 1 to 20 years). This was also observed for the specific diagnosis of peri-implantitis, both at the patient (OR = 0.45; 95% CI [0.30; 0.68]; p = .0002; I²: 51%; analysis based on six studies including 736 patients) and implant level (OR = 0.26; 95% CI [0.15; 0.46]; p < .0001; I²: 21%; analysis based on six studies including 1337 implants). No significant differences were observed between regular and irregular adherence to SPIC for the diagnosis of peri-implant mucositis.

In a sensitivity analysis excluding those studies that involved patients with a history of periodontitis, dental implants undergoing regular SPIC showed an OR = 0.23 (95% CI [0.08; 0.64]; p = .005; I²: 0%) of developing peri-implantitis compared to dental implants with no SPIC (based on two studies).

When dental implants were used as the statistical unit of analysis, those subjected to regular SPIC demonstrated a lower PD (MD: −0.48 mm; 95% CI [−0.67; −0.29]; p < .0001; I²: 32%; analysis based on five studies including 867 implants) and a reduced risk of exhibiting a MBL > 2 mm (OR: 0.4; 95% CI [0.25; 0.66]; p = .0003; I²: 73%; analysis based on three studies including 689 implants). Irregular SPIC was associated with a 3.76 increased risk of implant failure (95% CI [1.50; 9.45]; p = .005; I²: 0%) compared with regular SPIC.

All studies reporting dental implant survival evaluated study samples that included a proportion of patients with a history of periodontitis. Overall, the estimated mean implant survival was 99.3% (95% CI [98.6%; 100%]) in the regular SPIC group (based on 564 implants) and 97.8% (95% CI [95.6%; 99.9%]) in the irregular SPIC group (based on 454 implants) (follow-up ranging from 4.5 to 20 years after implant loading).

The RCT that was evaluated compared four different SPIC protocols (including a 3-monthly SPIC with curettes, with sonic scalers or air-polishing, and with or without chlorhexidine varnish application) and found no significant differences between the groups in terms of PD, BOP, and survival at 1 year (Ziebolz et al., 2017).

When comparing patients with a history of generalized moderate-to-severe periodontitis presenting with deep residual pockets (>6 mm) during SPC, with patients who had a history of generalized moderate-to-severe periodontitis but without residual deep pockets, a significantly higher occurrence of peri-implantitis (3.5% vs. 15.2%, implant level analysis) was observed when deep residual pockets were present (Cho-Yan Lee et al., 2012).

Consistency

All selected studies were overall consistent, favouring regular SPIC over irregular SPIC.

Balance of benefit and harm

Not assessed. However, the importance and clinical relevance of SPIC should be reinforced, given that regular SPIC carries little risk compared with the benefits it brings.

Overall certainty of the evidence

Moderate. Results are inferred from studies with various working hypotheses that were not originally developed to test the effectiveness of a preventative measure on the occurrence of peri-implant diseases. Further research, including clinical trials with strict inclusion criteria, may have an impact on confidence in the estimated effect of regular versus irregular SPIC on the risk of peri-implant diseases.

Number and design of included studies

Four studies, including three case–control studies and one cohort study (Carra et al., 2023). Clear similarities between the three case–control studies conducted by the same research team were noted.

Risk of bias

According to NOS, the three case–control studies were at high risk of bias, and the cohort study was at low risk of bias.

Effect sizes and their clinical relevance

Only one study described the occurrence of peri-implant diseases as a clinical diagnosis, reporting a lower rate of peri-implant mucositis (43.9% vs. 48.6%) and peri-implantitis (19.7% vs. 30.5%) in former smokers compared with current smokers (Costa et al., 2022). The authors observed a direct association between cumulative smoking exposure and the risk for peri-implantitis, as well as with the time span since smoking cessation.

All studies reported significant clinical differences between former smokers, e-cigarette users, waterpipe smokers and current smokers. The former smoker category exhibited less peri-implant mucosal inflammation, PD and MBL compared with the other categories.

Consistency

There is insufficient evidence to determine whether cigarette smoking cessation decreases the risk for peri-implant diseases. There is little evidence to support the contention that using e-cigarettes or the habit of water pipe smoking is associated with a decreased risk for peri-implant diseases compared with cigarette smoking.

Balance of benefit and harm

Not assessed. However, because of the several harmful consequences of smoking, smoking cessation should be advised and promoted for every patient irrespective of implant therapy.

Overall certainty of the evidence

Low. No interventional studies were found to provide direct evidence. The results are inferred from studies with various working hypotheses that were not originally developed to test the effectiveness of smoking cessation on the occurrence of peri-implant diseases. Further research is very likely to have an impact on confidence in the estimate of the effects of cigarette cessation on the reduction of the risk of incident peri-implant diseases. Regarding the use of non-cigarette smoking, any estimate of effect is very uncertain.

Number and design of included studies

Six of the studies included in the SR and meta-analysis (three RCTs, one NRCT, one case–control and one cohort study) were considered. They compared peri-implant tissue health parameters between sites with a deficiency in PIKM and receiving an FGG to increase PIKM width versus no intervention. No study was specifically designed to assess the impact of FGG on the prevention of peri-implant diseases.

Risk of bias

According to RoB-II-RCT, the three RCTs presented some concerns. According to RoBins-NRTC, the selected study was at moderate risk of bias. According to NOS: the two studies were at low risk of bias.

Effect sizes and their clinical relevance

Indirect evidence based on the evaluation of peri-implant health parameters in the short term showed a non-significantly different PPD between the PIKM-augmented and non-augmented sites but a significantly lower clinical soft tissue inflammation index (BOP/GI) (standardized mean difference [SMD] = −1.18; 95% CI [−1.85; −0.51]; p = .0006; I²: 69%) around the dental implants receiving FGG to augment PIKM. Concerning the mean MBL, based on data from four studies, a significant difference in favour of PIKM-augmented sites (SMD: −0.25; 95% CI [−0.45; −0.05]; p = .01; I²: 62%) was also noted. When excluding from pooled data analysis of cohort and case–control studies, the results were consistent with no statistical heterogeneity. No difference in PPD (SMD: −0.25; 95% CI [−0.63; −0.13]; p = .20; I²: 0%; based on 107 implants), whereas a significant difference in BOP (SMD: −1.5; 95% CI [−1.93; −1.06]; p < .0001; I²: 0%; based on 107 implants) and MBL changes (SMD: −0.33; 95% CI [−0.55; −0.11]; p = .003; I²: 0%; based on two studies, 66 implants) were noted between PIKM-augmented sites versus non-augmented sites.

Only two studies reported the occurrence of PIDs (Frisch et al., 2015; Roccuzzo et al., 2016). The first study defined peri-implantitis as the presence of BOP, PPD ≥ 5 mm, and a radiographic bone loss ≥3.5 mm (Frisch et al., 2015). During a mean follow-up of 12 years, three groups receiving FGG, CTG or no intervention were compared. No statistical differences were found between groups. The second study, a 10-year prospective cohort, observed a significantly higher rate of PIDs for dental implants with PIKM deficiency compared with implants surrounded by PIKM (51.4% vs. 12.7%; p < .0001) (Roccuzzo et al., 2016). The authors also reported a significantly lower soreness for implants surrounded by PIKM or placed in the alveolar mucosa receiving FGG compared with implants surrounded by alveolar mucosa and not receiving FGG (Roccuzzo et al., 2016).

Consistency

Results are based on heterogeneous studies with, most of the time, small sample sizes and short follow-ups. Consistency is low.

Balance of benefit and harm

Not assessed. However, the decision-making process concerning surgical procedures to augment PIKM should consider the general risks associated with periodontal and implant surgery.

Overall certainty of the evidence

Low. No study design provided direct evidence. Results are inferred from studies with various working hypotheses that were not originally developed to test the effectiveness of peri-implant soft tissue augmentation procedures on the prevention of peri-implant diseases over time.

Care must be taken regarding the interpretation of the study results due to the high clinical heterogeneity of the included studies. Most of the studies described clinical peri-implant outcomes in the short term (6–12 months follow-up), whereas only two observational studies reported the occurrence of peri-implant diseases over a 10-year (low risk of bias) and 12-year (high risk of bias) follow-up.

However, a reduced width of keratinized tissue is associated with an increased prevalence of peri-implantitis, plaque accumulation, soft tissue inflammation, mucosal recession, MBL and greater patient discomfort (Ramanauskaite et al., 2022). The effectiveness of increasing PIKM as a preventive measure for peri-implant diseases requires longitudinal studies designed with a long-term follow-up, to evaluate the outcome of interest (i.e., peri-implant diseases).

Number and design of included studies

Eight studies, including one NRCT and six RCTs (Carra et al., 2023).

Risk of bias

According to NOS, one study is at high risk of bias; according to RoB-II-RCT, two studies were at low risk of bias and four studies presented some concern. According to RoBins-NRTC, one study is at moderate risk of bias.

Effect sizes and their clinical relevance

Pooled data analyses were based on four studies, including 179 implants, and found no difference between CTG/FGG versus XCM for mean PPD, MBL and BOP. One CCT with a small sample size (19 patients) observed a 4.3% rate of peri-implantitis in the control group compared with 0% in the test group receiving CTG (partial split-mouth design) (Hosseini et al., 2020). Meta-analysis was performed pooling together two studies comparing CTG versus no intervention (Frisch et al., 2015; Hosseini et al., 2020), and including 37 implants in CTG-augmented sites versus 69 implants in non-augmented sites. It showed no significant difference between the two groups for the rate of incident peri-implantitis (OR = 1.97; 95% CI [0.20; 19.72]; p = .56; I²: 0%).

Consistency

Data are consistent, although based on a very limited number of studies.

Balance of benefit and harm

Not assessed. However, the decision-making process should balance the risks associated with the different surgical procedures aimed at increasing PIKM thickness against the risks of surgery and the additional related costs, in people with peri-implant mucosal health.

Overall certainty of the evidence

No study design provided direct evidence. Care must be taken regarding the interpretation of the results, due to the high clinical heterogeneity of the included studies, in particular the high variability of the timeline at which the augmentation procedure was performed (before or after dental implant placement, after dental implant loading, simultaneously to the dental implant placement, at the stage 2 surgery, etc.). Most of the studies described clinical peri-implant outcomes in the short term (6–12 months follow-up).

Number and design of included studies

Three studies were selected: two RCTs and one case–control study (Carra et al., 2023).

Risk of bias

According to NOS, one study was at low risk of bias. According to RoB-II-RCT, two studies were at some concern.

Effect sizes and their clinical relevance

Due to the heterogeneity in reporting outcome measures, no analysis of pooled data was possible. One study found a significant difference in favour of a counter-rotational powered toothbrush in terms of peri-implant mucosal inflammation and implant survival compared with manual toothbrushing (Truhlar et al., 2000). One case–control study indicated that the frequency of tooth brushing (at least twice a day vs. at most once a day) had no impact on peri-implant PD, MBL and BOP (Alhakeem et al., 2023).

Consistency

The three studies included were inconclusive regarding the type of toothbrush to use (e.g., powered or manual toothbrush), or the frequency of toothbrushing that was most effective in maintaining peri-implant health.

Balance of benefit and harm

Not assessed in the studies were considered. However, advising patients about OH and promoting OH behaviour improvements (in terms of techniques and frequency) carry little risk compared with the benefit it brings.

Overall certainty of the evidence

Low.

Number and design of included studies

A total of 15 studies were included in this SR (Stiesch et al., 2023). The studies included were of prospective (n = 10) and retrospective (n = 5) design reporting on a single treatment group (n = 9) or multiple treatment groups (n = 6), conducted in a university (n = 12) or private practice (n = 3). All studies that fulfilled the inclusion criteria regarding patient number (≥20 patients) and follow-up time (≥3 years) were focused on the medium- to long-term outcomes of peri-implantitis treatment. None of the studies meeting the inclusion criteria were specifically designed to evaluate or compare different SPIC protocols or SPIC frequencies and only one study was designed to evaluate the effect of SPIC on the secondary prevention of peri-implantitis.

Risk of bias

Most studies (87%) were assessed as having a low risk of bias, two studies (13%) showed some concerns, mainly regarding the inclusion of participants (lack of randomization information), treatment standardization or definition of treatment success and disease recurrence. There was considerable heterogeneity between studies with respect to study design including: peri-implantitis case definitions, outcomes reported, outcome definitions for success and disease recurrence, peri-implantitis treatment methods and supportive care protocols.

Effect sizes and their clinical relevance

Definitions for peri-implantitis, treatment success and recurrence of disease varied considerably across the 15 studies, contributing significantly to the heterogeneity of the data. While all definitions of peri-implantitis included clinical parameters such as BOP, PD and radiographic bone loss, the defined thresholds for bone loss and PD were heterogeneous.

Definitions for success were reported by 13 of the studies but also varied between studies. Therefore, a quantitative assessment of implant- and patient-level success was not possible. In nine studies, success was defined as PD < 5 mm with no BOP or suppuration and no further bone loss. In one study, success was defined as PD < 4 mm with no BOP or suppuration and no mobility. One study defined success as PD reduction, favourable soft tissue parameters and BOP decrease. Another study defined success as no further bone loss of >1.0 mm and no implant removal, and a further study defined success as radiographic evidence of >25% bone fill.

The definition of disease ‘recurrence’ also varied significantly between the studies. In eight studies, ‘further bone loss’ was defined as one important criterion for recurrence, together with implant loss (two studies). In four studies, BOP was a criterion for recurrence and in one study disease recurrence included clinical outcomes not meeting the success criteria.

Consistency

The review found that peri-implant tissue stability reported at the patient level and at the implant level varied widely and that recurrence of peri-implantitis was reported in up to 65.2% of treated implants receiving SPIC in studies with a follow-up of 3 years or more. While the SR (Stiesch et al., 2023) aimed to identify the most effective supportive care protocol in maintaining peri-implant tissue stability after peri-implantitis treatment, no comparison of protocols could be made. Furthermore, as the studies were not specifically designed to evaluate supportive care protocols, detailed information regarding supportive care was lacking. Therefore, it was not possible to make any conclusion regarding the most effective supportive care protocol. However, the protocols included similar preventive and therapeutic principles of supportive periodontal care as described in the EFP S3-level treatment guideline for stages I–III periodontitis (Sanz et al., 2020). Regular removal of plaque from the treated implant was common to all protocols described. Several studies also specified the provision of full-mouth professional plaque removal and the reinforcement of OH instructions.

Balance of benefit and harm

The results of this review confirm that SPIC may result in peri-implant tissue stability after peri-implantitis treatment. However, disease recurrence may occur, requiring additional treatment or, in some cases, implant removal. The undesirable effects of SPIC have not been described in the included studies.

Overall certainty of the evidence

Currently, there is no high-quality evidence available to answer the PICOS of the SR. Based on the available literature, a meta-analysis was not possible. The overall evidence on the effect of SPIC on the secondary prevention of peri-implantitis is based on one RCT, seven prospective and five retrospective clinical trials. Provision of SPIC following peri-implantitis therapy may prevent disease recurrence or progression. Insufficient evidence is available to identify (i) a specific supportive care protocol for secondary prevention of peri-implantitis, (ii) the effect of adjunctive local antiseptic agents in the secondary prevention of peri-implantitis and (iii) the impact of frequency of supportive care provision. Future prospective randomized controlled studies designed to evaluate supportive care protocols are needed.

Acceptability

In most of the identified studies, the number of drop-outs was few and the study participants seemed to be compliant. Based on the findings of the SR (Stiesch et al., 2023), it may be assumed that the provision of SPIC with a frequency between 3 and 6 months over a time span of 3 years is acceptable for patients following peri-implantitis treatment.

Feasibility

There were no perceived barriers.

Ethical considerations

As an example, in Germany, neither implant therapy nor SPIC is part of the statutory health insurance. Patients only receive access to SPIC through private health insurance or self-payment.

Economic considerations

As SPIC may prevent peri-implantitis recurrence, it is an important tool to support overall oral health and well-being of patients with implants. The loss of an implant may be associated with bone loss, psychological distress, pain, and costly and time-demanding retreatments, which may require specialist management.

Legal considerations

There were no legal constraints.

Risk of bias

Not applicable.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

Not applicable.

Overall certainty of the evidence

Not applicable.

Acceptability

Self-performed OH measures are generally well accepted by individuals.

Feasibility

Non-surgical mechanical/physical treatment of peri-implant mucositis can be performed by dental hygienists, general dentists as well as specialist.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

The SR (Gennai et al., 2023) included two RCTs evaluating the effect of oral irrigators used by the patient adjunctively to PMPR compared with PMPR, demonstrating significant BOP reduction at 3 months in patients with peri-implant mucositis.

Risk of bias

The overall risk of bias of the included studies was judged as ‘moderate’ (RoB 2 tool), with one study with a low risk of bias and one with a moderate risk of bias.

Effect sizes and their clinical relevance

These two RCTs show imprecision in the effect estimates, the results are not consistent and publication bias could not be assessed.

Consistency

The reported results are not consistent.

Balance of benefit and harm

It could not be assessed.

Overall certainty of the evidence

Two RCTs have evaluated the adjunctive self-use of oral irrigators, one using 0.06% chlorhexidine (CHX) as the irrigating fluid and the other water, one study was at low and the other at moderate risk of bias. Furthermore, the imprecision of the effect estimates, the lack of consistency of the results and the potential risk of publication advises downgrading the quality of the evidence.

Acceptability

Oral irrigators are usually well accepted by patients.

Feasibility

There are no perceived barriers.

Ethical considerations

There are no perceived ethical considerations.

Economic considerations

There is an additional cost on buying the irrigator.

Legal considerations

There are no perceived legal considerations.

Risk of bias

Study quality assessment identified some concerns of risk of bias in one study and high risk of bias in the other.

Effect sizes and their clinical relevance

One study reported disease resolution/treatment success in 8.3%–16.7% at 6 months, and BOP severity of 0.70–0.74. In this study, OH instruction was performed before the baseline examination. Another study reported BOP extent at 12.1%–18.6% at 12 months.

Consistency

Evidence was consistent in the two studies with limited reduction in BOP. The only patient-reported outcome showed no difference in pain during treatment when titanium curettes were compared with chitosan brush.

Balance of benefit and harm

An overall consideration of the benefit versus harm of professionally administered non-surgical mechanical/physical therapy supports the recommendation.

Overall certainty of the evidence

Low.

Acceptability

Patients usually accept and understand the need for treatment.

Feasibility

Non-surgical mechanical/physical treatment of peri-implant mucositis can be performed by dental hygienists, general dentists as well as specialist.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

Three RCTs addressed the PICOS question (n = 313 patients). Two RCTs analysed the effect of laser therapy adjunctive to ultrasonics and curettes (n = 289), and one RCT analysed the effect of air-polishing adjunctive to ultrasonics (n = 24), all with a 3-month follow-up. One study compared ultrasonics with carbon fibre tip plus glycine powder air polishing versus ultrasonics alone (n = 24). The results on mean BOP severity were 1.1 and 1.0, respectively. The second study (n = 220) compared ultrasonics with carbon fibre tip and titanium-coated curettes with and without diode laser (980 nm) application. Results were 34.5% and 30.9% disease resolution, respectively. BOP extent at 3 months was 23.2% and 26.8%, respectively. The third study (n = 69) compares ultrasonic with plastic tips and plastic curettes with and without diode laser (810 nm). The reported BOP extent was 0.26 and 0.57 respectively at 3 months, this difference being statistically significant.

Risk of bias

Study quality assessment identified some concerns of risk of bias in two studies, and a third had a high risk of bias.

Effect sizes and their clinical relevance

One RCT reported disease resolution/treatment success in 30.9%–34.6% and 23.2%–26.8% BOP extent at 3 months. Another RCT reported BOP extent of 0.26 and 0.57 in favour of adjunctive laser at 3 months, which was statistically significant. The third RCT reported BOP severity of 1.0 and 1.1 at 3 months.

Consistency

Evidence was consistent in the studies with a reduction in BOP, but statistically significant only in one of the RCTs with laser therapy adjunctive to ultrasonics and curettes. No patient-reported outcomes were reported.

Balance of benefit and harm

An overall consideration of the benefit versus harm of professionally administered non-surgical mechanical/physical therapy supports the recommendation.

Overall certainty of the evidence

Moderate.

Acceptability

Patients usually accept and understand the need for treatment.

Feasibility

Non-surgical mechanical/physical treatment of peri-implant mucositis can be performed by dental hygienists, general dentists as well as specialist.

Ethical considerations

Not applicable.

Economic considerations

Additional costs associated with adjunctive laser therapy may not be justified.

Legal considerations

Not applicable.

Risk of bias

Not applicable.

Effect sizes and their clinical relevance

No RCTs were available, but in one of the included trials (Riben-Grundstrom et al., 2015), results were reported at multiple time points after providing repeated mechanical instrumentation. After an initial reduction of 20.9%–28.6% in BOP extent, the effect of further repetitions was limited (1.9%–6.3%, and 0.0–11.3%, respectively).

Consistency

Not applicable.

Balance of benefit and harm

Not applicable.

Overall certainty of the evidence

Not applicable.

Acceptability

Not applicable.

Feasibility

Non-surgical mechanical/physical treatment of peri-implant mucositis can be performed by dental hygienists, general dentists as well as specialist.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Risk of bias

Low risk of bias.

Effect sizes and their clinical relevance

Results at 6 months demonstrated reductions in the modified bleeding index of 1.14 and 0.50 for test and control groups, respectively; these differences were statistically significant; and, at 6 months, disease resolution was 66.6% and 9.6%, respectively.

Consistency

Not applicable.

Balance of benefit and harm

There is a clear benefit and minimal harm in the prosthesis modification to improve access for biofilm control.

Overall certainty of the evidence

Limited due to the scarcity of the available evidence.

Acceptability

Well accepted intervention, although patients may complain for a short time of food entrapment.

Feasibility

Prosthesis modification should be implemented by general dentists as well as specialist.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

No study could be identified when considering the inclusion criteria outlined in the SR (Dommisch et al., 2023). However, one RCT (n = 32) evaluated the adjunctive effect of minocycline microspheres in the treatment of peri-implant mucositis/incipient peri-implantitis (bone loss less or equal to three threads). Results showed a significant added effect in reducing BOP and PD at 6 months. However, BOP relapsed after 9 months (Renvert et al., 2006).

Risk of bias

Not applicable.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

Harm versus benefit considerations on the use of antibiotics need to be undertaken. The use of antibiotics should always meet the antibiotic stewardship guideline.

Overall certainty of the evidence

Not applicable.

Acceptability

Not applicable.

Feasibility

Not applicable.

Ethical considerations

The use of antibiotics should always meet the antibiotic stewardship guideline.

Economic considerations

High economic costs and limited availability of products in European countries need to be considered.

Legal considerations

Not applicable.

Number and design of included studies

Two RCTs were selected by the SR (Dommisch et al., 2023). One of them (n = 37 patients) assessed the professional administration of 0.12% CHX in 119 implants over the time periods of 1, 3 and 6 months in non-smokers. In the control group, CHX was not professionally applied. Outcome measures compared PD, BOP and visible plaque index (PlI). Disease resolution, SOP and PROMs were not reported. In both the control and test groups, significant reductions in PD, BOP and visible PlI were observed when comparing values at baseline with values at the 3- and 6-month follow-up. The inter-group comparison revealed no differences when comparing test and control groups (Menezes et al., 2016). The second one (n = 46 patients) tested the professional administration of 0.95% NaOCl in 68 implants over a time period of 1, 3 and 6 months. In the control group, NaOCl was not applied. Outcome measures compared reduction in BOP, PD and a modified PlI. In addition, disease resolution was evaluated. Significant reductions in BOP, PD and the modified PlI for oral implants in both the test and control groups were observed at the 6-month follow-up. The inter-group comparison did not show differences among groups regarding BOP, disease resolution, PD or the modified PlI (Iorio-Siciliano et al., 2020). Changes in SOP as well as PROMs were not reported. Thus, the main finding of the RCTs identified in the SR was that 0.12% CHX or 0.95% NaOCl did not additionally improve clinical outcomes.

Apart from the evidence included in the SR, three additional RCTs were considered, evaluating the adjunctive effect of an antiseptic (CHX chip), a ‘postbiotic’ (‘Lactobacillus Ferment’) and a desiccant gel/liquid (concentrated aqueous mixture of hydroxybenzenesulphonic and hydroxymethoxybenzene acids, together with sulphuric acid) as adjunctives to mechanical therapy, compared with mechanical therapy plus application of 1% CHX gel (Butera et al., 2022; Lombardo et al., 2019; Sahrmann et al., 2019).

Risk of bias

For CHX and NaOCl, study quality assessment using the RoB 2 tool identified a low risk of bias for both studies included (Iorio-Siciliano et al., 2020; Menezes et al., 2016). For the other three RCTs, risk of bias was not specifically evaluated, since they were not included in the SR, but their overall quality of evidence was considered as low.

Effect sizes and their clinical relevance

For CHX, based on one RCT (n = 37 patients), no additional effect of 0.12% CHX was demonstrated regarding reductions in BOP, PD and PlI. For NaOCl, based on one RCT (n = 46 patients), no additional effect of 0.95% NaOCl was identified regarding reductions in BOP, PD and PlI. In the case of the individual RCTs, the magnitude of the effect (comparison vs. a negative control or a placebo) could not be determined as the control groups were 1% CHX gel. For the ‘postbiotic’ gel, based on one RCT (n = 20 patients), no additional effect was demonstrated (Butera et al., 2022). For the desiccant solution, based on one RCT (n = 23 patients), significant differences between groups were only observed for plaque indices (Lombardo et al., 2019). For the CHX chip, based on one RCT (n = 32 patients), significant additional benefits in BOP were observed in the test group, but the statistically significant differences observed at baseline precluded a strong conclusion on the adjunctive effect (Sahrmann et al., 2019). The effect of the 1% CHX was heterogeneous, which is beneficial in two studies (Butera et al., 2022; Lombardo et al., 2019), but not in the third (Sahrmann et al., 2019).

Consistency

Not applicable.

Balance of benefit and harm

In both identified studies, the adjunctive professional administration of 0.12% CHX or 0.95% NaOCl did not cause unintentional side effects that suggest harm to the patient (Iorio-Siciliano et al., 2020; Menezes et al., 2016). Thus, formulations of both CHX and NaOCl may be considered as a professional treatment adjunctive to non-surgical mechanical/physical therapy in the treatment of peri-implant mucositis. Future studies are needed to further investigate the efficacy of the given and other concentrations of CHX and NaOCl. For CHX, several adverse effects such as taste alteration, mouth numbness, xerostomia, and tooth discoloration have been reported (Poppolo Deus & Ouanounou, 2022). For NaOCl, the occurrence of potential adverse effects is uncertain for various concentrations. Potential adverse side effects must be considered to balance benefits and harms. For the ‘postbiotic gel’, potential unintentional side effects were not reported (Butera et al., 2022); based on the composition of the postbiotic gel, potential side effects, such as allergic reactions, cannot be excluded. For the CHX chip, numerous unintentional side effects are listed in the product information, but they are reported to be not frequent and usually mild. For the desiccant, no unintentional side effect was reported (Lombardo et al., 2019); however, potential side effects of sulphuric acid are listed by the company, and thus, the application is not recommended in patients if allergic to sulphur in any form and in the case of pre-existing skin disorders.

Overall certainty of the evidence

The certainty is weak lack of studies.

Acceptability

In general, the application of antiseptics is well accepted by patients when understanding the pathogenesis of peri-implant mucositis.

Feasibility

CHX gels, CHX chips, desiccant materials, ‘postbiotics’ and NaOCl formulations can be professionally applied by the general dentist or specialist. Their adjunctive use is not clinically demanding or time-consuming. For the NaOCl formulation (PeriSolv®, RLS Global AB, Mölndal, Sweden); CHX chip (PerioChip, Karr Dental, Wollerau, Switzerland), the ‘postbiotic’ (Biorepair Parodontgel Intensive, Coswell SPA, Funo di Argelato, BO, Italy), and the desiccant liquid (HybenX® Oral Tissue decontaminant™, EPIEN Medical Inc., Saint Paul, MN, USA), specific brands were tested and the information provided may only be valid for those products, which may not be available in all markets.

Ethical considerations

Based on the available evidence, no evaluation of ethical aspects could be performed.

Economic considerations

CHX gels, CHX chips, desiccant materials, “postbiotics” and NaOCl formulations are associated with additional costs to the patient as well as to the dental professional team. The application of any antiseptic treatment adjunctive to non-surgical mechanical/physical therapy may lead to additional costs for the patients depending on individual health insurance plans in the individual countries. As examples, the additional costs associated with the use of the desiccant material, in Germany, are approximately €100 for two syringes of 1 mL each, and for the use of CHX chips is approximately €300 for 20 applications. No information on cost-effectiveness could be retrieved from the RCTs (Butera et al., 2022; Iorio-Siciliano et al., 2020; Lombardo et al., 2019; Menezes et al., 2016; Sahrmann et al., 2019).

Legal considerations

The NaOCl formulation (PeriSolv®) is approved as Class I medical device in the European Union, and the desiccant material (HybenX®) has also been approved as Class I medical device in the European Union and Canada. The implications of the use in other geographical locations or the use for indications besides the ones approved are unclear.

Number and design of included studies

For the application of aPDT adjunctive to sub-marginal instrumentation, five RCTs (in total, n = 204 patients) analysed an estimated number of 231 implants over a time period of 3 months (Dommisch et al., 2023). Of these five studies on adjunctive application of aPDT, four included patients with habitual tobacco intake (cigarette smokers, smoke-less tobacco chewers and vaping individuals). In the control groups, aPDT was not applied. In the test groups, the intervention varied in terms of a range in the applied wavelength between 660 and 670 nm, power density between 100 milliwatts (mW) and 150 mW. One study did not report on treatment modalities. Outcome measures compared BOP, PD and PlI. Disease resolution and PROMs were not reported. In the synthesis of data, three studies were evaluated for changes in BOP and PD and four studies for changes in PlI, comparing test and control groups. For BOP and PD, no difference was identified between test and control groups, whereas for PI, a significant difference was shown in favour of aPDT adjunctive to sub-marginal instrumentation (Dommisch et al., 2023). High heterogeneity and a high level of asymmetry were evident (Dommisch et al., 2023). Two RCTs were excluded due to the lack of reporting mean and standard deviation or assessing a modified bleeding index instead of BOP. Changes in SOP as well as PROMs were not reported. The main findings were that aPDT did not additionally improve clinical outcomes for changes in BOP, PD or PlI.

Risk of bias

For aPDT, study quality assessment using the RoB 2 tool identified a low risk of bias for one study, whereas some concerns indicated a risk of bias in four studies on aPDT.

Effect sizes and their clinical relevance

For aPDT, based on three RCTs (204 patients) included in the meta-analysis, no additional effect of the adjunct application of aPDT was demonstrated regarding reduction in BOP and PD (Dommisch et al., 2023). A significant reduction of PlI was identified in the meta-analysis; however, clinically, this reduction was not related to the reduction of surrogate parameters for disease resolution (reduction or absence of BOP, reduction in PD) (Dommisch et al., 2023).

Consistency

For aPDT, the identified RCTs included male patients only, and from these five RCTs four focused on patients with habitual tobacco intake (cigarette smokers, smoke-less tobacco chewers and vaping individuals). The analysis of data revealed high heterogeneity among the studies (Dommisch et al., 2023). This inconsistency among the studies may be explained by the heterogeneity of reported outcome parameters as well as regarding the variation of tobacco intake habits, even though only male patients were evaluated. In addition, the intervention varied in terms of a range in the applied wavelength between 660 and 670 nm, power density between 100 mW and 150 mW, and choice of photosensitizer (phenothiazine chloride, methylene blue) in the respective test groups.

Balance of benefit and harm

For the additional application of aPDT adjunctive to sub-marginal instrumentation, no benefit was identified in the meta-analysis (Dommisch et al., 2023). Potential harm of aPDT adjunctive to sub-marginal instrumentation has not been studied to date. However, potential adverse effects cannot be entirely ruled out due to various wavelength, power density and photosensitizer available on the market.

Overall certainty of the evidence

The overall certainty regarding the additional effect of aPDT is weak. The quality of evidence is low.

Acceptability

The adjunct application of aPDT is accepted by patients when understanding the pathogenesis of peri-implant mucositis.

Feasibility

The application of aPDT can only be performed by a trained operator and appropriate eye protection must be used by the dental professional team and the patient.

Ethical considerations

Not applicable.

Economic considerations

The application of aPDT causes comparatively high costs for the dental team with regard to the acquisition and maintenance of the corresponding equipment. For the patient, aPDT adjunctive to sub-marginal debridement may lead to additional costs depending on individual health insurance plans in the individual countries. No information on cost-effectiveness could be retrieved from the five selected RCTs. Additional costs associated with adjunctive laser therapy may not be justified.

Legal considerations

Not applicable.

Number and design of included studies

The SR (Gennai et al., 2023) included five RCTs evaluating the effect of antiseptics used by the patient adjunctively to sub-marginal instrumentation compared with sub-marginal instrumentation alone or combined with a negative control or a placebo in terms on BOP reduction at 3 months in patients with peri-implant mucositis.

In these five RCTs, self-administered antiseptics as adjuvant to PMPR were used in the format of gels (0.5% CHX) or mouth rinses. In this latter delivery format (mouth rinses), the following active agents have been tested: CHX at different concentrations (0.03%, 0.12% or 0.2%) alone or combined with cetylpyridinium chloride (CPC) (0.05%); herbal-based mouth rinses; delmopinol (0.2%).

Risk of bias

The overall risk of bias of the included studies was judged as ‘low’ (RoB 2 tool), with all five studies with a low risk of bias.

Effect sizes and their clinical relevance

For CHX gel: 29 patients received PMPR at the implant sites, they were instructed to brush around the implant twice daily using a chlorhexidine gel (0.5%) (n = 15) or a placebo gel (n = 14) for a period of 4 weeks, and there were significant reductions in the mean number of sites with BOP from baseline to 1 month for both test and control groups (p < .05), with little apparent change between 1 and 3 months (p > .1); there was no statistically significant difference in the changes in BOP between the test and control groups at 1 month or at 3 months (p > .1).

For CHX mouth rinses, four RCTs with 166 patients compared the efficacy of self-administered CHX mouth rinses versus distilled water/saline or placebo, for 2 weeks, 1 month or 1 year, and the results showed significant reductions over time of BOP, with conflicting results in terms of superiority versus control. Statistically significant differences in BOP or in modified gingival index (MGI) were noted after 3 months, while no statistically significant differences in terms of BOP were reported at 1 month or with the usage of 0.05% CHX plus 0.05% CPC at 1 year (Pulcini et al., 2019).

For herbal mouth rinses, 2 RCTs with 62 patients were managed with self-administered herbal-based mouth rinses for 2 weeks or NaCl/distilled water. At 3 months, statistically significant differences in BOP and MGI, between test and control groups, were reported, with better performance in the herbal mouth rinse groups.

For delmopinol, one RCT analysed the efficacy of 1-month self-performed delmopinol mouth rinse versus placebo, with 59 patients. Both treatments showed reduction on BOP with no differences among test and control groups.

Consistency

Conflicting results were reported when using CHX.

Balance of benefit and harm

In the included studies, some antiseptics have been associated with undesirable side effects, such as transient anaesthetic sensation in the oral mucosa (delmopinol) or higher levels of staining on the teeth or tongue (CHX). Moreover, other rarer side effects cannot be excluded.

Overall certainty of the evidence

Low.

Acceptability

Antiseptics are widely accepted by the population.

Feasibility

There are no perceived barriers.

Ethical considerations

The issue has not been addressed. There are no perceived ethical considerations.

Economic considerations

For dentifrices, it may not be relevant since it is always combined with mechanical tooth brushing. For mouth rinses use, the extra cost should be taken into consideration.

Legal considerations

It should also be noted that the evidence base contains studies using products that may no longer be available.

Number and design of included studies

The SR (Gennai et al., 2023) included six RCTs evaluating the effect of systemic probiotic used by the patient, adjunctively to sub-marginal instrumentation, compared with sub-marginal instrumentation alone or combined with a negative control or a placebo, in terms of BOP reduction at 3 months in patients with peri-implant mucositis.

Risk of bias

The overall risk of bias of the included studies was judged as ‘low’ (RoB 2 tool), with three studies with a low risk of bias and three with a moderate risk of bias.

Effect sizes and their clinical relevance

At 6 months, no statistically significant differences were found when comparing probiotics versus control groups for any study outcome. No adverse events were reported due to the adjunctive use of L. reuteri tablets.

Consistency

All studies reported the same tendency.

Balance of benefit and harm

No adverse events have been reported. Clear benefits observed at 3 months, although they were not sustained at 6 months.

Overall certainty of the evidence

Moderate.

Acceptability

Systemic probiotics are still not widely accepted by the population.

Feasibility

There are no perceived barriers.

Ethical considerations

There are no perceived ethical considerations.

Economic considerations

There are no perceived economic considerations, although an extra economic cost is derived from the prescription of the probiotics.

Legal considerations

There are no perceived legal considerations.

Number and design of included studies

The SR (Gennai et al., 2023) included three RCTs evaluating the effect of systemic antibiotics prescribed as oral administration adjunctively to sub-marginal instrumentation. In one study, amoxicillin was compared with sub-marginal instrumentation combined with probiotics. In another study, the adjunctive administration of azithromycin was compared with instrumentation alone. In the third study, the adjunctive use of azithromycin plus a 0.12% CHX mouth rinse was compared with instrumentation plus a 0.12% CHX mouth rinse. Outcomes evaluated in these three studies were the percentage of BOP, PlI and PD.

Risk of bias

The overall risk of bias in the included studies was judged as ‘moderate’ (RoB 2 tool), with all the three studies with a moderate risk of bias.

Effect sizes and their clinical relevance

No studies reported a longer follow-up than 6 months.

Consistency

All studies reported the same tendency.

Balance of benefit and harm

In one study that collected side effects, no adverse events were observed after antibiotic intake. No specific concerns can be raised for antibiotics as adjunctive use for treating peri-implant mucositis.

Overall certainty of the evidence

Moderate.

Acceptability

The population widely accepts antibiotics. Nevertheless, there is an issue related to the need of diminishing the usage of antibiotics due to the potential risks associated with antibiotic resistance.

Feasibility

There are no perceived barriers.

Ethical considerations

The issue has not been addressed. There are no perceived major ethical considerations. Yet it must be reiterated the need of containing prescription of antibiotics for the population at large.

Economic considerations

The specific economic considerations can be stated.

Legal considerations

No specific legal consideration can be stated.

Number and design of included studies

In the SRs prepared for the present project (Cosgarea et al., 2023; de Waal et al., 2023; Liñares et al., 2023), 15 RCTs with at least 6-month follow-up were considered as valid for developing recommendations. For the present recommendation, outcomes from both test and control groups are considered.

Risk of bias

Ten presented with low risk of bias, three with some concerns and two with high risk.

Effect sizes and their clinical relevance

The observed improvements after treatment are significant in magnitude and consistent across the considered RCTs. Taken together, the evidence is unlikely to arise from the placebo or the Hawthorne effect. Still, it is not possible to assess the relative contribution of the different components that have been tested.

Consistency

Not applicable.

Balance of benefit and harm

Benefits were observed in both the test and control groups. Of 17 test groups, statistically significant benefits were observed in 11 for PD reduction and 9 for BOP. Of 17 control groups, statistically significant benefits were observed in 11 for PD reduction and 7 for BOP. The percentage of disease resolution was provided by seven test groups (ranging 0%–65%) and seven control groups (ranging 14%–55%). Limited evidence of harm was presented.

Overall certainty of the evidence

Moderate.

Acceptability

The interventions for treating peri-implantitis seem to be acceptable for patients, health providers and health authorities, although no direct evidence is available.

Feasibility

The interventions for treating peri-implantitis are feasible, although some of them may need specific training.

Ethical considerations

The interventions for treating peri-implantitis may negatively impact equity, if public services are not covering the cost, and those will need to be directly covered by patients.

Economic considerations

Limited evidence is available, see Section 1.

Legal considerations

Not applicable.

Number and design of included studies

In the SRs prepared for the present project (Cosgarea et al., 2023; de Waal et al., 2023; Liñares et al., 2023), 15 RCTs with at least 6 months of follow-up were considered as valid for developing recommendations. For the present recommendation, both test and control groups are considered.

Risk of bias

Ten presented with a low risk of bias, three with some concerns and two with high risk.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

Benefits were observed in both test and control groups (see background text of previous recommendation). Limited evidence of harm was presented.

Overall certainty of the evidence

Low.

Acceptability

The interventions for treating peri-implantitis seem to be acceptable for patients, health providers and health authorities, although no direct evidence is available.

Feasibility

The interventions for treating peri-implantitis are feasible, although some of them may need specific training.

Ethical considerations

The interventions for treating peri-implantitis may negatively impact equity if public services are not covering the cost, as in these situations they will need to be directly covered by patients.

Economic considerations

Limited evidence is available, see Section 1.

Legal considerations

Not applicable.

Number and design of included studies

In the SRs prepared for the present project (Cosgarea et al., 2023; de Waal et al., 2023; Liñares et al., 2023), 15 RCTs with at least 6 months of follow-up were considered as valid for developing recommendations.

Risk of bias

Ten presented with low risk of bias, three with some concerns and two with high risk.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

Not applicable.

Overall certainty of the evidence

Low.

Acceptability

The evaluation of the outcomes after the non-surgical step of peri-implantitis treatment seems to be acceptable for patients, health providers and health authorities, although no direct evidence is available.

Feasibility

The evaluation of the outcomes after the non-surgical step of peri-implantitis treatment seems to be feasible.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

No study was found answering this question.

Risk of bias

Not applicable

Effect sizes and their clinical relevance

Due to the lack of studies, indirect evidence was used, analysing the clinical impact in control groups in the 15 RCTs identified in the three SRs (Cosgarea et al., 2023; de Waal et al., 2023; Liñares et al., 2023). Of 17 control groups, statistically significant benefits were observed in 11 for PD reduction and in seven for BOP. The percentage of disease resolution was provided for seven control groups, and it ranged 14%–55%. Limited evidence of harm was presented.

Consistency

Most control groups found a statistically significant impact of the treatment; this was similar to that reported in test groups.

Balance of benefit and harm

No proper evaluation of PROMs was carried out.

Overall certainty of the evidence

Moderate.

Acceptability

There is no evidence so far for clinicians' or patients' acceptability.

Feasibility

Implementation of therapy may be negatively influenced by the lack of retrievability and/or shape of the prosthetic suprastructure.

Ethical considerations

No data are available to address ethical considerations.

Economic considerations

Cost-effectiveness has not been evaluated in these studies.

Legal considerations

So far, if the manufacturer's indications are respected, there are no legal considerations.

Risk of bias

Two studies were at high risk, and three studies at low risk of bias.

Effect sizes and their clinical relevance

Due to heterogeneity in the treatment protocol, no meta-analysis was carried out. All studies showed improvements in both test and control groups in PD and BOP, at 3 and/or 6 months compared with baseline. In general, studies showed no additional benefit from the application of lasers at 6 months, in terms of either PD or BOP reductions. Only in one study did the adjunctive application of a Er,Cr:YSGG laser show statistically significantly larger PD reductions at 6 months, compared with sub-marginal instrumentation alone (Alpaslan Yayli et al., 2022). An Er:YAG laser as monotherapy (Schwarz et al., 2005, 2006) led to statistically significant differences in BOP. Their magnitude, however, was small.

Consistency

Positive results for the primary outcomes were observed in all five RCTs, for both control and test groups.

Balance of benefit and harm

No proper evaluation of PROMs was carried out in the studies.

Overall certainty of the evidence

Low.

Acceptability

None of the included studies provides evidence of superior patients' acceptance of laser application as compared with mechanical instrumentation with curettes. There is no evidence so far for clinicians' acceptability.

Feasibility

Implementation of therapy may be negatively influenced by the lack of retrievability and/or shape of the prosthetic suprastructure.

Ethical considerations

No data are available to address ethical considerations.

Economic considerations

Cost-effectiveness has not been evaluated in these studies.

Legal considerations

So far, the manufacturer's indications are respected, there are no legal considerations.

Number and design of included studies

Two RCTs (n_patients = 64, n_implants = 75) assessed the sub-marginal peri-implant instrumentation with air-polishing (Merli et al., 2020; Sahm et al., 2011). One used air-polishing as monotherapy (Sahm et al., 2011), while the other combined ultrasonics and air-polishing (Merli et al., 2020).

Risk of bias

Both studies had low risk of bias.

Effect sizes and their clinical relevance

Due to the heterogeneity of the treatment protocols, no meta-analysis was carried out. Both studies on air-abrasive decontamination showed PD and BOP reductions but no statistically significant differences. Inter-group differences for BOP were observed with air-polishing as monotherapy (Sahm et al., 2011).

Consistency

Not feasible to be assessed.

Balance of benefit and harm

One study reported higher levels of pain values during treatment and after 1 week for the glycine powder group as compared with mechanical instrumentation with ultrasonics (Merli et al., 2020). Cases of subcutaneous emphysema have been reported after the use of air-polishing devices (Alonso et al., 2017; Bassetti et al., 2014; Bruckmann et al., 2022). Among members of the expert panel, three groups had experienced such adverse events.

Overall certainty of the evidence

Very low.

Acceptability

Patient perception and acceptance were assessed in one study, showing no statistically significant differences (Merli et al., 2020).

Feasibility

Implementation of therapy may be negatively influenced by the lack of retrievability and/or shape of the prosthetic suprastructure. Sometimes sub-marginal delivery may not be possible due to the size of the nozzle.

Ethical considerations

Consider that the additional clinical benefit, if present, is small; that there is a potential risk of harm (subcutaneous emphysema); and that no clear benefit in terms of patient acceptability has been demonstrated.

Economic considerations

Cost-effectiveness has not been evaluated in these studies.

Legal considerations

So far, the manufacturer's indications are respected, and there are no legal considerations.

Number and design of included studies

Two RCTs assessing aPDT as adjunct to sub-marginal instrumentation, using either toluidine blue (66/66 patients) (Wang et al., 2019) or methylene blue (25/26 patients and 30/33 implants) (Alasqah, 2022), with appropriate wavelengths for the photosensitizers (635 nm for toluidine blue, 670 nm for methylene blue). As expected, no studies were found assessing aPDT as monotherapy, since aPDT cannot remove biofilm.

Risk of bias

One study was considered at low risk of bias, and the other had some concerns in terms of bias.

Effect sizes and their clinical relevance

Although both studies reported some favourable results in terms of PD reduction for aPDT as adjunct to sub-marginal instrumentation, over sub-marginal instrumentation alone, results were inconsistent and/or showed no differences for other outcome variables (BOP, MBL and/or CAL). No meta-analysis could be performed due to the limited number of studies identified and their heterogeneity.

Consistency

Substantial heterogeneity was observed in study design, interventions (laser type, photosensitizer and pre-treatment), populations studied and reported results of the studies.

Balance of benefit and harm

No adverse effects were reported.

Overall certainty of the evidence

Due to the heterogeneity in study design, interventions, populations studied and reported outcomes, the certainty of evidence is very low.

Acceptability

There are insufficient data to support or refute the use of aPDT as adjunct to sub-marginal instrumentation in the non-surgical treatment of peri-implantitis.

Feasibility

The adjunctive use of aPDT following sub-marginal instrumentation is not clinically demanding or time-consuming but requires the availability of a laser.

Ethical considerations

There is no evidence for ethical considerations. The studied photosensitizers are generally considered as safe.

Economic considerations

The additional cost associated with aPTD may not be justified.

Legal considerations

There are no obvious legal considerations.

Number and design of included studies

One factorial design RCT with two control and two test groups (16 of 16 patients and 16 of 16 implants) assessed the adjunctive desiccant antiseptic gel and the method of sub-marginal instrumentation (Merli et al., 2020). No studies were found testing efficacy as monotherapy.

Risk of bias

The study was considered at low risk of bias.

Effect sizes and their clinical relevance

PD and CAL reduction were greater in patients treated with the desiccant material, regardless of the sub-marginal instrumentation method (ultrasonic scaler alone or combined with glycine powder air-polishing). The magnitude of the additional improvements in PD was 0.5 mm. There were no significant differences for any of the other outcomes reported.

Consistency

Not applicable.

Balance of benefit and harm

No adverse effects were reported. However, since the product is an acid, a negative impact on the surrounding tissues may happen (caustic effect on the soft tissues).

Overall certainty of the evidence

Due to the limited number of studies, the certainty of the evidence is very low.

Acceptability

There are insufficient data to support the use of desiccant material as an adjunct to sub-marginal instrumentation in the non-surgical treatment of peri-implantitis.

Feasibility

The adjunctive use of desiccant material following sub-marginal instrumentation is not clinically demanding or time-consuming. Currently, there is only one brand name/manufacturer for this material (HybenX®, EPIEN Medical Inc., Saint Paul, MN, USA).

Ethical considerations

There is no evidence for ethical considerations.

Economic considerations

There are additional costs associated with the use of the desiccant material (e.g., in Germany the cost are ca. €100 for two syringes of 1 mL each).

Legal considerations

The product has been approved as Class I medical device in the European Union and Canada. The implications of the use in other geographical locations or the use for indications besides the ones approved are unclear.

Number and design of included studies

Two placebo-controlled RCTs with 6-month follow-up assessed the adjunctive effect of locally applied chlorhexidine ‘chips’ to the non-surgical sub-marginal instrumentation (Machtei et al., 2012, 2021). These studies used an intense regime with multiple, repeated applications during the observation period. In addition, although they were not included in the SR, two RCTs evaluating locally applied minocycline microspheres were considered in the discussions (Renvert et al., 2006, 2008).

Risk of bias

Two RCTs with low risk of bias.

Effect sizes and their clinical relevance

Results of two studies evaluating multiple applications of a biodegradable matrix containing chlorhexidine were pooled for meta-analyses, showing a statistically significant improved PD reduction (WMD = 0.2 mm; 95% CI [0.0; 0.5]; p = .031; I² = 0.0%; p = .570). No or very limited information was available for BOP or disease resolution.

Consistency

Not feasible due to the limited information available.

Balance of benefit and harm

No increase in adverse effects was observed. PROMs were not reported. Harm versus benefit considerations on the use of locally delivered antibiotics need to be considered.

Overall certainty of the evidence

Low.

Acceptability

No specific information is available; however, local antimicrobials are normally easy to use by the practitioners. Conversely, some patients/clinicians may not be willing to use antimicrobial products.

Feasibility

Some of the evaluated products may not be commercially available in some countries. For chlorhexidine ‘chips’, only one brand/manufacturer is available (PerioChip®, Dexcel Pharma, Or Akiva, Israel). For minocycline microspheres, the brand tested in the considered studies was Arestin® (OraPharma, Bridgewater, NJ, USA).

Ethical considerations

No applicable.

Economic considerations

Economic costs and cost-effectiveness should be considered before their use. Economic cost may be relatively high (for chlorhexidine ‘chips’, one chip may cost around €30, while for minocycline microspheres, one cartridge costs around $100, especially if multiple applications are needed). Some additional information is presented in Section 1.

Legal considerations

Some of the evaluated products have not been registered for use in some countries, and/or may not have been approved for this specific indication.

Number and design of included studies

RCTs (n = 2) with a double-blind, placebo-controlled, parallel design with follow-up up to 12 months (Blanco et al., 2022; Shibli et al., 2019). One evaluated amoxicillin plus metronidazole (n = 40 patients/40 implants) (Shibli et al., 2019), and the other, metronidazole alone (n = 32 patients/62 implants) (Blanco et al., 2022).

Risk of bias

Risk of bias was low for one study, while the other study presented some concerns.

Effect sizes and their clinical relevance

Systemic antimicrobials showed a greater PD reduction when compared with mechanical debridement alone at 6 months and up to the 12 months follow-up (≈1.5 mm). These results were more pronounced when the deepest site of each implant was considered for the analysis.

Consistency

Not applicable.

Balance of benefit and harm

One study assessed the potential side effects of systemic antibiotics, with six subjects (38%) in the test group (systemic metronidazole) and five (31%) in the control group (placebo) reporting either gastrointestinal disorders, headaches or metallic taste, without significant differences among groups. Global concerns regarding the overuse of antibiotics and the development of antibiotic resistance must be considered. Benefit versus harm analysis includes considerations on the overall use of antibiotics for the individual patient and public health. Systemic antibiotic regimens have shown long-lasting impacts on the faecal microbiome, including an increase in genes associated with antimicrobial resistance.

Overall certainty of the evidence

Limited evidence is available.

Acceptability

Due to concerns for patient's health and the impact of systemic antibiotic use on public health, its routine use as an adjunct to sub-marginal peri-implant instrumentation in patients with peri-implantitis is not recommended.

Feasibility

Adjunct systemic antimicrobials to non-surgical peri-implant therapy are a feasible procedure since these antimicrobials may be prescribed in most countries. Moreover, the procedure does not demand high clinical skills.

Ethical considerations

Important concerns are related to patient's health and the impact of systemic antibiotic use to public health.

Economic considerations

Although economic considerations have not been analysed in the included studies, some indications can be given. The cost of systemic antimicrobials is low, particularly in comparison to other potential adjuncts (e.g., local antimicrobials or probiotics). Although there is not enough evidence to provide any strong recommendation, the prescription of systemic antimicrobials in specific cases may reduce the need for additional treatment, including surgical procedures, reducing added costs and morbidity.

Legal considerations

There are no specific legal considerations.

Number and design of included studies

One placebo-controlled RCT assessed the adjunctive effect of probiotics to non-surgical sub-marginal instrumentation (Laleman et al., 2020), with a preparation containing L. reuteri, to be applied both locally and systemically.

Risk of bias

Some concerns.

Effect sizes and their clinical relevance

No adjunctive effect of the use of probiotics was observed on PD or BOP.

Consistency

Not applicable.

Balance of benefit and harm

No proper evaluation of PROMs was carried out, although the extrapolation from the periodontal field suggests that this formulation is safe, and patients do not frequently report adverse effects.

Overall certainty of the evidence

Very low.

Acceptability

No specific information is available. However, probiotics are normally easy to use by the practitioners. Conversely, some patients/clinicians may not be willing to use these products.

Feasibility

Adjunctive probiotics to non-surgical peri-implant therapy are a feasible approach since these products can be prescribed in many countries. Moreover, the procedure does not demand high clinical skills.

Ethical considerations

Not applicable.

Economic considerations

There is an additional cost associated with the use of probiotics that is borne by the patient.

Legal considerations

There are no specific legal considerations.

Number and design of included studies

Data from 13 prospectively collected studies (649 patients) with a follow-up ranging from 1 to 5 years addressed access flap and resective surgery. Seven RCTs assessed the efficacy of reconstructive surgery (194 patients) compared with access flap surgery. The respective datasets were evaluated in two SRs (Donos et al., 2023; Karlsson et al., 2023). All studies reported on reduction of PD and BOP. Clinically relevant end points (e.g., PD < 6 mm), PROMs, health economic parameters and adverse events were not consistently reported.

Risk of bias

The 13 studies on access flap and resective surgery were generally found to be at low RoB, while multiple studies evaluating reconstructive measures were judged to show high RoB.

Effect sizes and their clinical relevance

For access flap and resective surgery, the estimated reduction of PD was 2.2 mm (95% CI [1.8; 2.7]). Reconstructive surgery resulted in similar PD reduction (additional effect relative to access flap alone: −0.39 95% CI [−1.16; 0.24]) at 12 months. For access flap and resective surgery, reduction of standardized mean %BOP was estimated at 27.0 (95% CI [19.8; 34.2]) and an overall bone gain of 0.2 mm (95% CI [0.0; 0.5]) was noted. Reconstructive surgery resulted in an additional bone gain of 0.75 mm (95% CI [−1.39; −0.11]) over access flap alone at 12 months (CI is presented with negative values, since in the original analyses positive values indicated more gain for access flap and negative for reconstructive procedures). Over 5-year observation periods, disease recurrence/progression was observed at 32%–44% of treated implants. Corresponding implant loss was low in the short term but after 5 years ranged from 14% to 21%.

Consistency

Results were consistent across studies for changes of PD and MBL. Reduction of BOP was heterogenous across studies. Data were generated in various clinical settings, including university centres and private clinics.

Balance of benefit and harm

In general, considerable improvements in clinical and radiographic parameters were noted. However, disease recurrence and implant loss were not uncommon events after 5 years. Data on PROMs and adverse events were rarely reported.

Overall certainty of the evidence

The certainty of evidence is graded as moderate based on the lack of direct comparisons between surgical and non-surgical therapy of peri-implantitis.

Acceptability

PROMs were rarely reported. Limited data suggest a high degree of patient satisfaction at 1 year following surgical therapy. Adverse events reported were mostly related to the use of systemic antibiotics.

Feasibility

Related procedures are clinically demanding.

Ethical considerations

Some decontamination procedures and grafting materials evaluated in the studies included have not been tested for safety.

Economic considerations

Health economic parameters were not evaluated in the identified studies. In general, surgical therapy of peri-implantitis is a costly procedure. Some decontamination procedures and grafting materials may generate additional costs in the absence of documented benefit.

Legal considerations

Some decontamination procedures and grafting materials evaluated in the studies included have not been tested for safety and are considered off-label.

Number and design of included studies

Thirteen studies (n = 649 patients), with a follow-up range from 1 to 5 years (only two studies with a 5-year follow-up), were included (Karlsson et al., 2023). One study was an RCT comparing surgical therapy to non-surgical intervention. All datasets were prospective and a total of 10 originated from control arms within RCTs, while the remaining two were case series. All studies reported on reduction of PD and BOP. Clinically relevant end points (e.g., PD < 6 mm), PROMs, health economic parameters and adverse events were not consistently reported.

Risk of bias

The 13 studies were generally found to be at low RoB. In the two evaluations covering longer follow-ups (≥5 years; Carcuac et al., 2020; Heitz-Mayfield et al., 2018), loss to follow-up exceeded 20% and the overall rating was downgraded to ‘fair’.

Effect sizes and their clinical relevance

Based on 18 studies (n = 661 implants), the estimated reduction of PD was 2.2 mm (95% CI [1.8; 2.7]). Based on 8 studies (n = 477), reduction of standardized mean BOP% was estimated at 27.0 (95% CI [19.8; 34.2]). Based on 12 studies (n = 637), a standardized mean bone gain of 0.2 mm (95% CI [0.0; 0.5]) was estimated. Over 5-year observation periods, disease recurrence/progression was observed at 32%–44% of treated implants. Corresponding implant loss was low in the short term but after 5 years ranged from 14% to 21%.

Consistency

Results were consistent across studies in regard to changes in PD and MBL. Reduction of BOP was heterogenous across studies. Data were generated in various clinical settings, including university centres and private clinics.

Balance of benefit and harm

In general, considerable improvements in clinical and radiographic parameters were noted. However, disease recurrence and implant loss were not uncommon events after 5 years. Data on PROMs (two studies) and adverse events (three studies) were rarely reported.

Overall certainty of the evidence

The certainty of evidence is graded as moderate based on the lack of direct comparisons between surgical and non-surgical therapy of peri-implantitis.

Acceptability

PROMs were reported in two studies, only. Limited data suggest a high degree of patient satisfaction at 1 year after surgical therapy. Adverse events reported in three studies were mostly related to the use of systemic antibiotics.

Feasibility

Related procedures are clinically demanding.

Ethical considerations

Some decontamination procedures evaluated in the studies included have not been tested for safety.

Economic considerations

Health economic parameters were not evaluated in the identified studies. In general, surgical therapy of peri-implantitis is a costly procedure. Some decontamination procedures may generate additional costs in the absence of documented benefit.

Legal considerations

Some decontamination procedures evaluated in the studies included have not been tested for safety and are considered off-label.

Number and design of included studies

Seven RCTs assessed the efficacy of reconstructive surgery (total of 200 implants in 194 patients) compared with access flap surgery (total of 188 implants in 184 patients) (Donos et al., 2023). Different types of reconstructive surgeries were documented, including the use of titanium granules, amelogenin, deproteinized bovine bone mineral (DBBM or DBBM graft with 10% collagen) alone or combined with a native bilayer collagen membrane, or a beta-tricalcium phosphate graft formulated with prolonged release of local doxycycline.

Risk of bias

Based on RoB 2, there was concern for four studies in one domain (predominantly due to bias in measurement of the outcome), while three studies were considered at high risk of bias, mainly due to the combination of missing outcomes and bias in selection of the reported results.

Effect sizes and their clinical relevance

Meta-analysis (4 studies; 262 patients and 272 implants) showed an estimated MD in PD changes between access flap surgery and reconstructive surgery of −0.39 (95% CI [−1.16; 0.24]; p = .325, I² = 66.4%) at 12 months. No evidence of small-study effects was detected. Among the five studies that reported on BOP changes at 12 months, one study showed a statistically significant improvement for reconstructive therapy as compared with access flap surgery. No differences were indicated in relation to the change in SOP. At 12 months, implant survival was similar between the two treatment procedures, ranging from 85.7% to 100% for access flap and from 95% to 100% for reconstructive therapy. Meta-analysis for changes in radiographic mean bone levels (4 studies; 262 patients and 272 implants) showed a statistically significant benefit of reconstructive compared with access flap surgery of −0.75 mm (95% CI [−1.39; −0.11]; p = .022; I² = 83.4%). The CI is presented with negative values, since in the original analyses positive values indicated more gain for access flap and negative for reconstructive procedures. Irrespective of the surgical approach and biomaterial employed, resolution of peri-implantitis is unpredictable and a significant difference between the two treatment approaches was not consistently shown.

Consistency

Overall, inconsistency in the direction of effect was noticed for the included studies, as only one showed a significant improvement in PD change and one in BOP change, when reconstructive procedures were employed.

Balance of benefit and harm

A similar number of adverse events and complications was associated with reconstructive and access flap surgeries. In the long-term, a number of implants are expected to develop disease recurrence, which may require additional surgical procedures or could lead to implant loss.

Overall certainty of the evidence

The certainty of evidence is low based on the quality of the studies (RoB) and inconsistency of outcomes.

Acceptability

Only two studies considered PROMs, with no significant differences in terms of pain scores, number of tablets taken and satisfaction.

Feasibility

Related procedures are clinically demanding.

Ethical considerations

Some decontamination procedures applied in the studies have not been tested for safety.

Economic considerations

No study addressed health economic outcomes on this topic (Donos et al., 2023). Reconstructive surgery represents an additional financial burden for the patient, which should be discussed with the patient.

Legal considerations

Not applicable.

Number and design of included studies

None of the identified studies in the SR was designed to investigate the site prerequisites for a reconstructive surgery (Donos et al., 2023). Nevertheless, the five RCTs of the network meta-analysis included ≥3 mm, angular peri-implant bone defects, which showed significant improvements in clinical and radiographic parameters from baseline to 12 months post-reconstructive therapy. Deeper defects are more likely to result in radiographic defect fill and 3- and 4-wall defects result in higher reduction in PD and BOP.

Risk of bias

Based on RoB 2, the risk of bias varied from low to high in the relevant studies.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Despite the three identified studies showed consistency on the impact of defect morphology on the treatment outcome, none of these studies was designed to answer this question.

Balance of benefit and harm

Not applicable.

Overall certainty of the evidence

Low.

Acceptability

Not applicable.

Feasibility

Not applicable.

Ethical considerations

Not applicable.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

Five RCTs and six prospective case series assessed the efficacy of reconstructive peri-implantitis therapy (Donos et al., 2023).

Risk of bias

Based on RoB 2, two of the five included RCTs were at high risk of bias, some concerns were raised for two studies and one was at low risk of bias. Based on ROBINSI, one CCT was at serious risk of bias, three prospective cohort studies were considered at serious risk of bias and two prospective cohort studies were at critical risk of bias (Donos et al., 2023).

Effect sizes and their clinical relevance

Regardless of the biomaterials applied, reconstructive therapy led to a mean PD reduction ranging from 2.0 to 4.5 mm and to a mean reduction in BOP ranging from 44.8% to 86% at 12 months after therapy. Studies reporting on SOP showed a significant reduction at 12 months and 5 years post-surgery. Based on one study (45 patients and 75 implants), included in the network meta-analysis (4 studies; 160 patients and 190 implants) (Donos et al., 2023), an improved PD reduction was shown when a xenogeneic rather than an autologous graft was applied in combination with a collagen membrane. Implant survival at 12 months ranged from 92% to 100%, but when considering composite outcomes for peri-implantitis resolution the range reported by the included studies was considerably wider (0%–91% at 12 months).

Consistency

All reconstructive procedures improved clinical and radiographic outcomes as compared with baseline regardless of the biomaterials employed.

Balance of benefit and harm

None of the different reconstructive approaches was associated with early side effects or adverse events beyond what would be expected for this type of surgical procedure. Notably, the use of a combination of membrane and bone graft was associated with an increased risk for flap dehiscence in two studies.

Overall certainty of the evidence

Low.

Acceptability

Based on one study, the use of a graft alone was associated with significantly less pain at 2 weeks as compared with the combined use of a graft and collagen membrane.

Feasibility

Not applicable.

Ethical considerations

Not applicable.

Economic considerations

No study addressed health economic outcomes on this topic. However, it should be noted that reconstructive surgery represents an additional financial burden for the patient.

Legal considerations

Not applicable.

Number and design of included studies

No focused question in the current SR (Donos et al., 2023) was formulated to address this topic. Nevertheless, none of the included studies compared submerged with unsubmerged healing protocol.

Risk of bias

Not applicable.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

The main advantage of submerged healing would be to achieve primary wound closure and to promote an aseptic healing environment, which are crucial factors for stabilizing the blood clot, improving graft stability and maximizing the regenerative potential of the intra-bony compartment. On the other hand, unsubmerged healing eliminates the need for prosthesis removal, reduces treatment time, costs and possibly the overall complexity of treatment.

Overall certainty of the evidence

Very low.

Acceptability

It should be noted that a submerged healing protocol may result in the need of temporary tooth replacement.

Feasibility

Not applicable.

Ethical considerations

Not applicable.

Economic considerations

No study addressed health economic outcomes on this topic. It should be noted that unsubmerged healing eliminates the need of prosthesis removal, thus reducing treatment time and possibly costs.

Legal considerations

Not applicable.

Number and design of included studies

In total, five RCTs (4 two-armed and 1 three-armed; 183 patients/242 implants) with follow-up periods ranging from 6 to 24 months were included (Ramanauskaite et al., 2023). One RCT investigated the alternative use of air polishing with erythritol powder in conjunction with non-reconstructive surgical peri-implantitis therapy compared with standard instrumentation (Hentenaar et al., 2022); another RCT, with three arms, assessed the efficacy of titanium brushes (test 1) and air polishing with glycine powder (test 2) as alternative decontamination measures for implant surface decontamination compared with standard instrumentation in conjunction with non-reconstructive surgical peri-implantitis therapy (control) (Toma et al., 2019); two RCTs investigated the efficacy of Er:YAG laser compared with either standard instrumentation (Schwarz et al., 2012) or debridement with piezoelectric scaler and stainless-steal scaler (Wang et al., 2021) during reconstructive therapy and as an adjunct to implantoplasty; and one RCT evaluated the added value of a titanium brush, on top of ultrasonic decontamination and hydrogen peroxide in regenerative surgery (de Tapia, Valles, et al., 2019).

Risk of bias

Based on RoB 2, two RCTs were judged to have an overall low risk of bias, two RCTs had an overall high risk of bias and one RCT had an unclear risk of bias.

Effect sizes and their clinical relevance

Based on two RCTs with 6- to 12-month follow-ups, the adjunctive/alternative use of an air-abrasive device with glycine or erythritol powders did not result in improved BOP reductions compared with the control during surgical therapy of peri-implantitis (Hentenaar et al., 2022; Toma et al., 2019). One RCT indicated a significantly higher PD reduction following the alternative use of air polishing with glycine powder and titanium brushes compared with the standard decontamination (Toma et al., 2019). Based on one RCT, after 6 months, alternative use of titanium brush resulted in significantly higher BOP reduction compared with either air polishing or the standard instrumentation (i.e., curettes to remove hard deposits plus gauze soaked in saline/saline irrigation) (Toma et al., 2019).

During reconstructive therapy, a titanium brush resulted in significantly greater reduction of the deepest PD values compared with the control group (i.e., mechanical and chemical implant surface decontamination) (de Tapia, Valles, et al., 2019). An Er:YAG laser resulted in significantly higher PD reductions after 6 months in one RCT, but was not associated with improved BOP reductions over respective control measures (i.e., implantoplasty and standard instrumentation or debridement with piezoelectric scaler and stainless-steal scaler) as shown in two RCTs (Schwarz et al., 2012; Wang et al., 2021).

Consistency

Two RCTs reported on no benefit of air polishing either with erythritol or glycine powder on the reduction of BOP values (Hentenaar et al., 2022; Toma et al., 2019). A beneficial effect of the use of a titanium brush was reported in two RCTs in terms of BOP (Toma et al., 2019) and PD reductions (de Tapia, Valles, et al., 2019). Two RCTs consistently reported on no benefits of Er:YAG laser on changing BOP values after 6 and 12 months in conjunction with reconstructive therapy and as an adjunct to implantoplasty (Schwarz et al., 2012; Wang et al., 2021). Inconsistencies were found between the studies with respect to the PD changes following Er:YAG laser application. In fact, significantly higher PD reduction following the use of Er:YAG laser was reported after 6 months in one RCT (Wang et al., 2021), whereas after 24 months another RCT indicated no benefits of ER:YAG laser in reducing PD values (Schwarz et al., 2012).

Balance of benefit and harm

Harms have not been explicitly reported and evaluated in two RCTs. A slight pigmentation of peri-implant soft tissues was observed in one out of 30 patients treated with implantoplasty. One RCT reported on adverse events observed in one out of 16 patients associated with persistence of suppuration and swelling following air polishing. Another RCT reported on membrane exposure during the healing, following reconstructive therapy of peri-implantitis, however, without providing the number of implants/patients experiencing this complication.

Overall certainty of the evidence

The evidence was graded as low due to a low number of studies with a considerable heterogeneity.

Acceptability

None of the studies investigated PROMs.

Feasibility

Certain decontamination protocols may be considered as technically demanding.

Ethical considerations

Certain decontamination protocols have not been tested for safety.

Economic considerations

Economic aspects could not be assessed due to the lack of reporting.

Legal considerations

Not applicable.

Number and design of included studies

Evidence was available for PDT from two RCTs (n = 43) and for CHX from two RCTS (n = 130) (Wilensky et al., 2023). Both with a follow-up of ≥6 and up to 12 months. Only RCTs reporting mean PD changes and BOP changes were included.

Risk of bias

For PDT, the risk of bias was low to unclear, and for CHX it was unclear to high risk of bias.

Effect sizes and their clinical relevance

No benefits were observed with the adjunctive application of CHX; no improvement was observed for PDT in terms of PD reduction, and only minor reductions in BOP (MD = 7.4).

Consistency

For PDT, heterogeneity was low, and for CHX it was medium to high.

Balance of benefit and harm

One study did not report on adverse events, while three studies reported no to minor adverse effects. One study reported gastrointestinal problems in five patients who were taking systemic antibiotics. One study reported no adverse effects, and another study reported two patients with one complication.

Overall certainty of the evidence

The GRADE analysis showed a very low certainty of evidence for both adjunctive treatments in all the tested parameters.

Acceptability

None of the studies reported patient-reported outcomes and there is no evidence supporting one approach over the other, including the standard therapy.

Feasibility

While CHX solution is affordable and easily available, PDT results in additional costs without any documented clinical benefit.

Ethical considerations

The lack of efficacy together with possible side effects, such as hypersensitivity, suggests that these treatments are not justified.

Economic considerations

The additional costs associated with adjunctive PDT therapy are not justified.

Legal considerations

PDT is an off-label use during surgery, with no clear benefits.

Number and design of included studies

Two RCTs including 49 patients (25 test and 24 control) and 39 patients (20 test and 19 control) and followed for 1 year showed inconsistent results in terms of PD, BOP and bone level changes: one assessed the systemic application of amoxicillin, 750 mg, twice per day for 10 days, and starting 3 days prior to surgery (Carcuac et al., 2016); the other evaluated the systemic application of azithromycin, 500 mg at the day of surgery, and 250 mg, once per day, during 4 additional days (Hallstrom et al., 2017).

Risk of bias

Some concerns (Carcuac et al., 2016) and high risk (Hallstrom et al., 2017), as evaluated with RoB 2.

Effect sizes and their clinical relevance

Disease resolution (based on <5 mm PDs, no BOP and no additional bone loss >5 mm) was consistent between studies and favoured systemic antibiotics: 56% test versus 29.2% control (Carcuac et al., 2016); 46.7% test versus 25% control group (Hallstrom et al., 2017). Two implant losses occurred in the control group of the first study (Carcuac et al., 2016).

Consistency

See previous section.

Balance of benefit and harm

The potential benefit of the use of systemic antibiotics needs to be balanced with the overall risks, which include adverse events (e.g., allergic reactions) and antibiotic resistance.

Overall certainty of the evidence

Low.

Acceptability

Due to concerns about patients' health and the impact of systemic antibiotic use on public health and inconsistent evidence, its use as adjunct to surgical therapy of peri-implantitis is not recommended.

Feasibility

Not applicable.

Ethical considerations

Harms related to the intake of systemic antibiotics must be balanced with potential benefits.

Economic considerations

Not applicable.

Legal considerations

Not applicable.

Number and design of included studies

Two RCTs were identified: one assessing local minocycline application at the time of surgery in 50 patients (25 test and 25 control), and repeated at 1, 3 and 6 months, with all patients also receiving systemic amoxicillin thrice per day, 500 mg, for 3 days (Cha et al., 2019); and another evaluating local doxycycline application in 27 patients (14 test and 13 control), formulated in a bone graft, at the time of surgery (Emanuel et al., 2020).

Risk of bias

High risk of bias for both RCTs.

Effect sizes and their clinical relevance

Not applicable.

Consistency

Not applicable.

Balance of benefit and harm

The potential benefit of the use of local antibiotics needs to be balanced with the overall risks, which include adverse events (e.g., allergic reactions) and antibiotic resistance.

Overall certainty of the evidence

Very low.

Acceptability

Not applicable.

Feasibility

Related products may not be available in all European countries.

Ethical considerations

Harms related to the intake of local antibiotics must be balanced with potential benefits.

Economic considerations

Additional costs related to the medical product must be considered.

Legal considerations

Not applicable.