Efficacy of alternative or adjunctive measures to conventional non-surgical and surgical treatment of peri-implant mucositis and peri-implantitis: a systematic review and meta-analysis

Abstract

Purpose

To evaluate the efficacy of alternative or adjunctive measures to conventional non-surgical or surgical treatment of peri-implant mucositis and peri-implantitis.

Material and methods

Prospective randomized and nonrandomized controlled studies comparing alternative or adjunctive measures, and reporting on changes in bleeding scores (i.e., bleed0ing index (BI) or bleeding on probing (BOP)), probing depth (PD) values or suppuration (SUPP) were searched.

Results

Peri-implant mucositis: adjunctive use of local antiseptics lead to greater PD reduction (weighted mean difference (WMD) = − 0.23 mm; p = 0.03, respectively), whereas changes in BOP were comparable (WMD = − 5.30%; p = 0.29). Non-surgical treatment of peri-implantitis: alternative measures for biofilm removal and systemic antibiotics yielded higher BOP reduction (WMD = − 28.09%; p = 0.01 and WMD = − 17.35%; p = 0.01, respectively). Surgical non-reconstructive peri-implantitis treatment: WMD in PD amounted to − 1.11 mm favoring adjunctive implantoplasty (p = 0.02). Adjunctive reconstructive measures lead to significantly higher radiographic bone defect fill/reduction (WMD = 56.46%; p = 0.01 and WMD = − 1.47 mm; p = 0.01), PD (− 0.51 mm; p = 0.01) and lower soft-tissue recession (WMD = − 0.63 mm; p = 0.01), while changes in BOP were not significant (WMD = − 11.11%; p = 0.11).

Conclusions

Alternative and adjunctive measures provided no beneficial effect in resolving peri-implant mucositis, while alternative measures were superior in reducing BOP values following non-surgical treatment of peri-implantitis. Adjunctive reconstructive measures were beneficial regarding radiographic bone-defect fill/reduction, PD reduction and lower soft-tissue recession, although they did not improve the resolution of mucosal inflammation.

Introduction

Peri-implant diseases were defined during the 2017 World Workshop as biofilm‐associated pathological conditions affecting osseointegrated dental implants, and they were further classified into peri-implant mucositis and peri-implantitis [1,2,3]. Peri-implant mucositis is characterized by inflammation in the soft tissue compartment, whereas peri-implantitis also features loss of the implant-supporting bone [1,2,3]. It is assumed that untreated peri-implant mucositis is the precursor to peri-implantitis [4]. The onset of peri-implantitis was shown to occur early on, and its progression was characterized by a nonlinear, accelerating pattern that, in the absence of therapy, may ultimately lead to implant loss [5]. Numerous cross-sectional studies have recently reported on the high prevalence of peri-implant diseases, pointing to their common appraisal in daily clinical practice [6,7,8,9].

There is evidence from experimental clinical studies that peri-implant mucositis is a reversible condition if adequate bacterial plaque control is implemented [10, 11]. Non-surgical therapy in conjunction with oral hygiene reinforcement is considered a standard care treatment for managing peri-implant mucositis [1, 12]. At peri-implantitis sites, in contrast, non-surgical mechanical treatment alone or with adjunctive (i.e., local antibiotics, antimicrobial photodynamic therapy—aPDT) or alternative measures (e.g., air abrasive devices, erbium-doped yttrium aluminum garnet—Er:YAG laser monotherapy), has demonstrated only limited efficacy in obtaining disease resolution, indicating the necessity of surgical therapy in a majority of the cases [12, 13].

Recently, numerous surgical treatment protocols have been advocated for treatment of peri-implantitis using various surface decontamination approaches, along with resective measures (e.g., apical flap, osteoplasty, implantoplasty), reconstructive measures (e.g., bone fillers/autografts, guided bone regeneration), or a combination thereof (referred to as combined therapy) [13, 14]. Nonetheless, the reported efficacy of different surgical treatment approaches in arresting further disease progression varied considerably [15,16,17,18,19,20].

Currently, it remains unclear which interventions are most effective for the management of peri-implant diseases. Therefore, the aim of this systematic review and meta-analysis was to address the following focused question: In patients with peri-implant mucositis or peri-implantitis, what is the efficacy of non-surgical and surgical treatment with alternative or adjunctive measures on changing signs of inflammation compared to conventional non-surgical and surgical treatments alone?

Materials and methods

The review protocol was developed and structured according to the PRISMA (Preferred Re-porting Items for Systematic Review and Meta-Analyses) Statement [21]. The review was registered in PROSPERO, an international prospective register of systematic reviews (CRD42021247402).

Focused question

The focused question serving for literature search was structured according to the PICO format: “In patients with peri-implant mucositis and peri-implantitis, what is the efficacy of non-surgical (i.e., referring to peri-implant mucositis and peri-implantitis) and surgical (i.e., referring to peri-implantitis) treatments with alternative or adjunctive measures on changing signs of inflammation compared with conventional non-surgical and surgical treatments alone?”.

Population

Patients with peri-implant mucositis and peri-implantitis based on case definitions used in respective studies.

Intervention

Alternative (for biofilm removal) or adjunctive (local or systemic application of adjunctive antiseptic/antibiotic or reconstructive/resective therapy) measures to non-surgical and surgical treatments of peri-implant mucositis or peri-implantitis.

Comparison

Conventional measures for non-surgical and surgical treatments.

Outcome: primary outcomes

Changes in bleeding scores (i.e., bleeding index (BI), modified BI (mBI), sulcus bleeding index (SBI), or bleeding on probing (BOP), suppuration (SUPP), and probing depth (PD) values; secondary outcomes: changes in peri-implant mucosal level (ML) and radiographic marginal bone levels (RBL), radiographic defect fill (RDF).

Study design: Prospective randomized controlled (RCT), or nonrandomized controlled (CCT) studies (split-mouth or parallel group designs).

Study inclusion and exclusion criteria

Inclusion criteria:

• 1. Studies on peri-implant mucositis: Studies comparing alternative (i.e., for biofilm removal) or adjunctive measures (i.e., adjunctive antiseptic/antibiotic oral or systemic application) to conventional non-surgical (i.e., mechanical/ultrasonic debridement) treatment with at least 3 months of follow-up.

• 2. Studies on non-surgical treatment of peri-implantitis: Studies comparing alternative (i.e., for biofilm removal) or adjunctive measures (i.e., adjunctive antiseptic/antibiotic oral or systemic application) to conventional non-surgical (i.e., mechanical/ultrasonic debridement with or without chlorhexidine (CHX) irrigation) treatment with at least 6 months of follow-up.

• 3. Studies on surgical treatment of peri-implantitis: Studies comparing adjunctive measures (i.e., adjunctive measures for implant surface decontamination, resective therapy by means of implantoplasty or reconstructive approaches) to conventional surgical treatment (i.e., access flap surgery) with at least 6 months of follow-up.

• 4. Studies reporting on clinical changes in bleeding scores (i.e., BI/BOP), SUPP and/or PDs, following non-surgical (referring to peri-implant mucositis and peri-implantitis) or surgical (referring to peri-implantitis) treatments in respective groups.

• 5. Studies providing case definitions of peri-implant mucositis and peri-implantitis.

• 6. Studies with a minimum of 10 patients (5 per treatment group).

The literature search was restricted to English language.

Exclusion criteria:

• 1. Inclusion of less than five patients per treatment group.

• 2. Lack of case definition.

• 3. Lack of clinical data on the changes in BOP/BI, PD or SUPP.

Information source and search

Two electronic databases (MEDLINE (via PubMed) and The Cochrane Library) were searched for relevant articles published until 1^st April 2021. The search filter ‘humans’ was applied. Electronic search was complemented by a hand search of the following journals:

Clinical Implant Dentistry and Related Research; Clinical Oral Implants Research; International Journal of Oral and Maxillofacial Implants; Journal of Clinical Periodontology; Journal of Periodontology.

The combination of the following key words (i.e., Medical Subject Headings MeSH) and free text terms included:

“treatment” OR “nonsurgical treatment” OR “non-surgical treatment” OR “surgical treatment” OR “regenerative treatment” OR “augmentative treatment” OR “respective treatment” OR “reconstructive treatment” OR “therapy” OR “nonsurgical therapy” OR “non surgical therapy” OR “surgical therapy” OR “regenerative therapy” OR “augmentative therapy” OR “resective therapy” OR “reconstructive therapy” OR “antiseptic treatment” OR “antibiotic treatment” OR “adjunctive treatment” OR “antiseptic therapy” OR “antibiotic therapy” OR “adjunctive therapy”

AND

“peri-implant disease” OR “periimplant disease” OR “peri-implant infection” OR “periimplant infection” OR “mucositis” (MeSH) OR “peri-implant mucositis” OR “periimplant mucositis” OR “Periimplantitis” (MeSH) OR “peri-implantitis”.

Study selection

During the first literature-selection stage, according to the defined inclusion criteria, the titles and abstracts of all identified studies were screened for eligibility by two independent reviewers (A.R. and F.S.). In the second stage, the full texts of potentially eligible articles were reviewed and evaluated according to the aforementioned exclusion criteria. Differences between reviewers were resolved by discussion. The level of inter-examiner agreement for the first- and second literature-selection stages was expressed by Cohen’s kappa-scores.

Risk of bias in individual studies

The Cochrane Collaboration’s tool for assessing risk of bias (RoB 2) was used in the case of randomized clinical trials, whereas for nonrandomized studies, the ROBINS-I tool was employed [22].

Data collection

A data extraction template was generated and based on the study design, patient- and implant-related information, case definition, follow-up period, interventions, comparisons, and primary and secondary outcomes, patient enrollment into supportive therapy following the treatment as well as the study quality.

Data analyses

Heterogeneity among the studies, meta-analysis (i.e., weighted mean differences (WMDs) and 95% confidence intervals, random effect model to account for potential methodological differences between studies) and forest plots were assessed using a commercially available software program (Comprehensive Meta-Analysis V3, Biostat, Englewood, NJ 07,631 USA). Statistical significance was defined as p < 0.05.

Results

Search and screening

The screening process yielded 16.586 articles, of which 106 were selected for full-text evaluation (Fig. 1; Cohen’s kappa = 0.723). Upon analysis of the full texts, 26 studies (28 publications) were excluded mainly due to a follow-up period < 6 months (n = 8 studies) (for the studies reporting on peri-implantitis treatment) or a lack of a control/comparative treatment group (n = 3 studies), or different diagnoses (i.e., peri-implantitis and peri-implant mucositis) being pooled into the analysis (n = 2 studies) (Additional file 1). Finally, 80 articles describing 62 studies were included in the review (Cohens kappa = 0.80). Of those studies, 18 reported on the treatment of peri-implant mucositis, 17 reported on non-surgical treatment of peri-implantitis, and the remaining 27 reported on the surgical treatment of peri-implantitis.

Fig. 1

PRISMA flowchart

Subdivision of selected studies

All selected studies were subdivided according to differences in the treatment protocol:

Non-surgical treatment of peri-implant mucositis:

• Alternative measures for biofilm removal (4 RCTs [23,24,25,26]);

• Adjunctive diode laser/antimicrobial photodynamic therapy (aPDT) (4 RCTs (5 publications) [27,28,29,30,31]);

• Adjunctive local antiseptics (4 RCTs [32,33,34,35]);

• Adjunctive systemic antibiotics (2 RCTs [29, 36]);

• Adjunctive probiotics (2 RCTs [37, 38]);

• Adjunctive antiseptic home care mouthrinse (3 RCTs [39,40,41]).

Non-surgical treatment of peri-implantitis:

• Alternative measures for biofilm removal (5 RCTs (6 publications) [42,43,44,45,46,47]);

• Adjunctive diode laser/aPDT (2 RCTs [48, 49])

• Adjunctive local antiseptics/antibiotics (6 RCTs (7 publications) [50,51,52,53,54,55,56]);

• Adjunctive systemic antibiotics (1 RCT and 1 CCT [57, 58]);

• Adjunctive probiotics (2 RCTs [59, 60]).

Surgical treatment of peri-implantitis:

• Adjunctive and alternative measures for implant surface decontamination following non-reconstructive therapy (7 RCTs (8 publications) [18, 61,62,63,64,65,66,67]);

• Adjunctive and alternative measures for implant surface decontamination following reconstructive therapy (1 RCT [68] and 1 CCT [69]);

• Alternative and adjunctive measures for implant surface decontamination following combined therapy (2 RCTs [19, 70]);

• Adjunctive implantoplasty following non-reconstructive therapy (2 RCTs (3 publications) [71,72,73]);

• Adjunctive local and systemic antibiotics following non-reconstructive therapy (3 RCTs (4 publications) [18, 61, 64, 74]);

• Reconstructive therapy versus non-reconstructive surgery (6 RCTs (7 publications) [75,76,77,78,79,80,81]);

Reconstruction of the defect with different bone fillers, with and without a membrane (4 RCTs (5 publications) [82,83,84,85,86] and 3 CCTs (5 publications) [87,88,89,90,91].

Non-surgical treatment of peri-implant mucositis

The details regarding peri-implant mucositis definitions, non-surgical treatment protocols, and supportive peri-implant therapy are presented in Table 1. The follow-up periods in the included studies were 3 months (9 studies), 4.5 to 8 months (6 studies), and 12 months (3 studies).

Table 1 Included studies reporting on peri-implant mucositis treatment

Marked inconsistencies in case definitions for peri-implant mucositis appeared among the studies. Specifically, in all but 1 study [27], peri-implant mucositis diagnosis was based on the presence of BOP and/or SUPP, along with a radiographic MBL assessment. Regarding an MBL assessment, a peri-implant mucositis diagnosis was defined via an absence of bone loss compared to the baseline radiograph or via threshold values (i.e., ≤ 3 mm or ≤ 2 mm). In 9 studies, peri-implant mucositis diagnosis was supplemented by an assessment of PDs, with the large variations in the applied threshold values.

Three RCTs reported on patients’ enrollment into a supportive maintenance program [25, 30, 31, 41]. All treatments implemented for peri-implant mucositis resulted in improved clinical parameters. However, complete disease resolution (i.e., absence of BOP) rarely occurred throughout the short investigation periods (Table 1).

Efficacy of interventions

Alternative measures for biofilm removal

Alternative measures utilized to remove biofilm from contaminated implant surfaces (i.e., air-powder abrasive devices with glycine powder or chitosan brush) showed no beneficial clinical effect in terms of BI/BOP and PD values compared to the control treatment alone (i.e., mechanical debridement) [23,24,25,26].

Adjunctive diode laser/aPDT

In 4 RCTs (5 publications), either antimicrobial photodynamic therapy (aPDT) [27,28,29] or a diode laser [30, 31] was used in addition to mechanical debridement. Over a 3-month period, adjunctive use of aPDT led to similar treatment outcomes in terms of BOP [27,28,29] and PD changes [28, 29], while 1 study reported on a higher reduction in PD values for the sites treated with adjunctive aPDT [27]. Similarly, the additional application of a diode laser resulted in similar BOP and PD changes compared to the mechanical treatment alone over 3- and 12-month periods [30, 31].

Adjunctive local antiseptics

As an adjunct to mechanical debridement, included studies employed either applications of CHX (0.12%) gel [32], a full-mouth disinfection concept utilizing CHX gel and mouth rinse [33, 34], or applications of sodium hypochlorite [35]. Over a 3- to 6-month follow-up period, adjunctive use of the aforementioned local antimicrobials led to similar changes in BOP scores [32, 34, 35] and PD values [33,34,35] compared to control treatments (i.e., mechanical debridement alone), whereas one study reported on a greater PD reduction following the adjunctive use of local CHX (0.12%) applications [32].

Adjunctive systemic antibiotics

The potential beneficial effect of adjunctive systemic antibiotic use for peri-implant mucositis treatment was investigated in 2 RCTs [29, 36]. In particular, administration of systemic antibiotics (azithromycin) along with mechanical debridement [36] or in combined with subgingival debridement and aPDT therapy [29] failed to show any beneficial effect upon the changes of BOP and PD values over follow-up periods of 3- to 6-months.

Adjunctive probiotics

Two RCTs investigated the potential benefits of probiotics [37, 38]. Of those, 1 RCT in which probiotics were administered for 15 days following the mechanical treatment failed to detect additional beneficial effects of probiotics in BOP and PD changes compared to the controls [37]. Another RCT pointed to significantly higher BOP reduction following the adjunctive use of probiotics for 30 days compared to the controls, whereas changes in PD values were similar to those obtained in the control group [38].

Adjunctive antiseptic home care mouth rinse

Three RCTs investigated the possible beneficial effect of home care use of cetylpiridinum chloride (CPC) + CHX 0.03% mouth rinse [40], oral irrigator with or without 0.06% CHX [39], or CHX 0.2% mouth rinse compared to 0.2% delmopinol hydrochloride [41]. Although 2 of them found similar BOP and PD changes irrespective of the adjunctive use of home care antibacterial mouth rinse throughout a 3-month follow-up period [40, 41], the remaining RCT indicated significantly higher BOP reduction for the patients in the test group [39].

Synthesis of results

Alternative measures for biofilm removal

Based on the patient-level analysis, the WMD in PD values were − 0.33 mm [SE = 0.35; p = 0.34; 95% CI (− 1.02, 0.35)], not favoring the use of alternative measures (i.e., air powder abrasive device with glycine powder) for biofilm removal (p value for heterogeneity: 0.02, I² = 81.5% = substantial heterogeneity) [23, 24] (Fig. 2a). At the implant level, WMD in PD amounted to − 0.49 mm [SE = 0.17; p = 0.01; 95% CI (− 0.82, − 0.15)], thus pointing to no favorable effect of alternative measures (i.e., air abrasive device with glycine powder and chitosan brush) for biofilm removal compared to mechanical debridement (p value for heterogeneity: 0.00, I² = 0.0% = low heterogeneity) [23, 26] (Fig. 2b).

Fig. 2

Forest plots indicating weighted mean difference (95% CI) in the changes of the assessed treatment outcomes following non-surgical treatment of peri-implant mucositis. a Alternative measures for biofilm removal (patient-level analysis)—PD. b Alternative measures for biofilm removal (implant-level analysis)—PD. c Adjunctive aPDT (patient-level analysis)—BOP. d Adjunctive aPDT (patient-level analysis)—PD. e Adjunctive local antiseptic therapy (implant-level analysis)—BOP. f Adjunctive local antiseptic therapy (implant-level analysis)—PD. g Adjunctive probiotics (implant-level analysis)—PD. h Adjunctive home care mouthrinse (implant-level analysis)—PD

Adjunctive aPDT

The WMD in BOP and PD values were − 0.85% [SE = 0.56; p = 0.13; 95% CI (− 1.96, 0.26)] and − 0.22 mm [SE = 0.26; p = 0.39; 95% CI (− 0.72, 0.28); unit of analysis: patient], respectively, thus not favoring the adjunctive use of aPDT compared to mechanical debridement alone (p value for heterogeneity: 0.013, I² = 77% = substantial heterogeneity and p = 0.747, I² = 0.0% = low heterogeneity, repsectively) (Fig. 2c and d) [27,28,29].

Adjunctive local antiseptics

The WMD in BOP amounted to − 5.30% [SE = 5.04; p = 0.29; 95% CI (− 15.06, 4.57); unit of analysis: implant], thus not supporting the superiority adjunctive use of local antiseptics (i.e., CHX) along with mechanical debridement (2 RCTs; p value for heterogeneity: 0.828, I² = 0.0% = low heterogeneity; Fig. 2e) [33, 34]. Based on 4 RCTs, the WMD in PD values was − 0.23 mm [SE = 0.10; p = 0.03; 95% CI (− 0.43, − 0.03); unit of analysis: implant], favoring the adjunctive use of local antiseptics (i.e., CHX and sodium hypochlorite; p value for heterogeneity: 0.929, I² = 0.0% = low heterogeneity; Fig. 2f) [32,33,34,35].

Adjunctive probiotics

According to 2 RCTs, the WMD in PD values amounted to − 0.22 mm [SE = 0.15; p = 0.14; 95% CI (− 0.52, 0.08); unit of analysis: implant], suggesting no superiority of probiotics in terms of PD reduction (p value for heterogeneity: 0.749, I² = 0.0% = low heterogeneity; Fig. 2g) [37, 38].

Adjunctive antiseptic home care mouthrinse

Based on 2 RCTs, the estimated WMD in PD amounted to − 0.11 mm [SE = 0.12; p = 0.37; 95% CI (− 0.33, 0.12); unit of analysis: implant), not favoring the use of adjunctive antiseptic home care mouthrinse as an adjunct to mechanical debridement (p value for heterogeneity: 0.8, I² = 0.0% = low heterogeneity; Fig. 2h) [40, 41].

Non-surgical treatment of peri-implantitis

Peri-implantitis definitions, non-surgical treatment protocols, and supportive peri-implant therapies are addressed in Table 2. The follow-up periods in the included studies were either 6 months (10 studies) or 12 months (7 studies).

Table 2 Included studies reporting on non-surgical treatment of peri-implantitis

Definitions of peri-implantitis varied widely among the included studies. All studies defined peri-implantitis as the presence of BOP and/or SUPP and radiographic MBL. The reference points (i.e., baseline radiographs) and threshold values used to identify MBL were either not specified [42, 43, 48, 54] or exhibited large variations [44,45,46,47, 49,50,51,52,53, 55,56,57,58,59,60].

In four RCT’s patients were enrolled into a regular maintenance program following the treatment [44,45,46,47, 58]. Although the investigated clinical parameters tended to improve significantly 6 to 12 months after the implemented non-surgical interventions, the treated sites were frequently associated with residual BI and/or BOP scores.

Efficacy of interventions

Alternative measures for biofilm removal

As an alternative to mechanical debridement, Er:YAG laser [42, 43], ultrasonic devices [44], and air-powder abrasive devices with glycine powder [45,46,47] were utilized to remove biofilm from contaminated implant surfaces. While the use of Er:YAG laser [42, 43] and an air-powered abrasive device with glycine powder [46, 47] led to significant improvements in BOP scores compared to mechanical debridement, the aforementioned alternative measures had no beneficial effect upon the changes in PD values. The use of an ultrasonic device failed to improve clinical treatment outcomes in terms of changes in BOP and PD when compared to mechanical debridement alone [44].

Adjunctive diode laser/aPDT

As an adjunct to mechanical therapy, the use of a diode laser resulted in comparable outcomes (i.e., BOP and PD changes) to the control group [49], whereas adjunctive aPDT therapy led to significantly higher PD and SBI reduction over a 6-month period compared to the control treatment (i.e., mechanical debridement) [48].

Adjunctive local antiseptics/antibiotics

In addition to mechanical debridement, application of local antibiotics (i.e., single [50, 52, 53] or repeated applications of minocycline microspheres [51]), CHX 1.0% gel (single [50] or repeated [51]), repeated application of CHX-containing chips [54, 55], or single subgingival placement of desiccant material [56] were investigated. Single application of minocycline microspheres in initial peri-implantitis cases (i.e., bone loss ≤ 3 mm) led to significantly higher PD reduction and comparable BOP changes [50], while repeated applications, on a contrary, yielded significantly greater BOP reduction, but similar PD changes [51] compared to the control sites (i.e., sites treated with mechanical debridement along with CHX 1.0% gel applications). Two RCTs reported similar changes in BOP values, but significantly higher PD improvements at implant sites treated with repeated CHX chips or single desiccant material application compared to placebo over 6 months [55, 56]. One study, however, failed to demonstrate any clinical beneficial effect in terms of BOP and PD changes of CHX chips over a 6-month period compared to the placebo group [54].

Adjunctive systemic antibiotics

Two RCTs investigated the potential benefits of the administration of systemic antibiotics along with mechanical debridement [57, 58]. Based on one RCT, prescribed systemic antibiotics (azithromycin 500 mg 3 day prior to treatment) along with mechanical debridement resulted in significant BOP and PD reduction (peri-implantitis definition: BO p + PD > 5 mm + bone loss > 2 mm) [57], whereas another RCT observed no beneficial effects of a combination of metronidazole 400 mg and amoxicillin 500 mg for BOP and PD changes in severe cases of peri-implantitis (i.e., BO p + PD > 5 mm + bone loss > 4 mm) [58].

Adjunctive probiotics

Contradictory findings were reported by 2 RCTs that evaluated the effects of the adjunctive use of probiotics for 6 months [59, 60]. In particular, one analysis failed to reveal any benefits of the adjunctive use of probiotic tablets and single local applications of probiotic drops upon the BOP and PD changes [60], whereas another RCT found similar BOP changes, but significant improvements in PD values following mechanical debridement along with systemic antibiotics in patients who also took probiotics for 6 months [59].

Synthesis of results

Alternative measures for biofilm removal

According to 3 RCTs, the WMD in BOP was − 28.09% [SE = 3.74; p = 0.01; 95% CI (− 35.43, − 20.76); unit of analysis: patient] in favor of alternative measures for biofilm removal (i.e., Er: YAG laser, air-powder abrasive device with glycine powder; p value for heterogeneity: 0.95, I² = 0.0% = low heterogeneity) [42, 43, 47](Fig. 3a). The WMD in PD values was − 0.27 mm [SE = 0.21; p = 0.19; 95% CI (− 0.68, 0.13)]; unit of analysis: patient), thus not favoring the alternative measures used for biofilm removal (i.e., Er: YAG laser, air-powder abrasive device with glycine powder, ultrasonic device) as an adjunct to mechanical debridement (p value for heterogeneity: 0.938, I² = 0.0% = low heterogeneity) (5 RCTs) [42,43,44, 47, 56](Fig. 3b).

Fig. 3

Forest plot indicating weighted mean difference (95% CI) in the reduction of assessed treatment outcomes following non-surgical treatment of peri-implantitis. a Alternative measures for biofilm removal (patient-level analysis)—BOP. b Alternative measures for biofilm removal (patient-level analysis)—PD. c Alternative measures for biofilm removal (patient-level analysis)—ML. d Adjunctive local antiseptic/antibiotic therapy (patient-level analysis)—BOP. e Adjunctive local antiseptic/antibiotic therapy (patient-level analysis)—PD. f Adjunctive local antiseptic therapy (patient-level analysis)—ML. g Adjunctive systemic antibiotics (patient-level analysis)—BOP. h Adjunctive systemic antibiotics (patient-level analysis)—PD. i Adjunctive probiotics (patient-level analysis)—PD

Based on 2 RCTs, the WMD in ML was − 0.21 mm [SE = 0.34; p = 0.55; 95% CI (− 0.87, 0.46); unit of analysis: patient], suggesting no superiority of alternative measures for biofilm removal (p value for heterogeneity: 0.026, I² = 80% = substantial heterogeneity) [42, 47] (Fig. 3c).

Adjunctive local antiseptics/antibiotics

Based on 3 studies, the estimated WMD in BOP values was − 10.65% [SE = 5.63; p = 0.06; 95% CI (− 21.69, 0.38)] (unit of analysis: patient), pointing to no beneficial effect of the local use of adjunctive antibiotics (i.e., minocycline microspheres) and local antiseptic (i.e., CHX) compared with mechanical debridement alone (p value for heterogeneity: 0.962, I² = 0% = low heterogeneity; Fig. 3d) [50, 51, 54]. Based on 4 RCTs, the WMD in PD amounted to –0.25 mm [SE = 0.18; p = 0.16; 95% CI (− 0.60, 0.10)]; unit of analysis: patient), with adjunctive local antiseptic/antibiotic therapy not yielding higher PD reduction (p value for heterogeneity: 0.988, I² = 0.0% = low heterogeneity; Fig. 3e) [50, 51, 54, 56]. The estimated WMD in ML was − 0.11 mm [SE = 0.09; p = 0.22; 95% CI (− 0.29, 0.07)]; unit of analysis: patient], thus indicating that the adjunctive local application of antiseptics did not lead to superior soft-tissue levels compared to mechanical debridement alone (p value for heterogeneity: 0.988, I² = 0% = low heterogeneity; Fig. 3f) [55, 56].

Adjunctive systemic antibiotics

Based on 2 RCTs with 12 months of follow-up, the WMD in BOP and PD amounted to − 17.35% [SE = 2.56; p = 0.01; 95% CI (− 22.37, − 12.32)]; unit of analysis: patient) and − 1.46 mm [SE = 0.35; p = 0.01; 95% CI (− 2.15, − 0.77)]; unit of analysis: patient), thus supporting the favorable effect of adjunctive systemic antibiotics following mechanical debridement (p value for heterogeneity: 0.474, I² = 0.0% and p = 0.562, I² = 0.0% = low heterogeneity, Fig. 3g and h) [57, 58].

Adjunctive probiotics

The WMD in PD values was − 0.15 mm [SE = 0.16; p = 0.35; 95% CI (− 0.47, 0.17)]; unit of analysis: patient), not favoring adjunctive probiotics compared to mechanical debridement alone (2 RCTs) (p value for heterogeneity: 0.719, I² = 0.0% = low heterogeneity, Fig. 3i) [59, 60].

Surgical treatment of peri-implantitis

Ten RCTs (12 publications) reported on the non-reconstructive surgical treatment of peri-implantitis [18, 61,62,63,64,65,66,67, 71,72,73,74] and 9 studies (13 publications) reported on the surgical treatment of peri-implantitis employing adjunctive reconstructive measures (4 CCTs [69, 87,88,89,90,91], 5 RCTs [68, 82,83,84,85,86, 92]). The remaining 6 RCTs (7 publications) compared reconstructive peri-implantitis treatment over non-reconstructive approach [75,76,77,78,79,80,81, 93] and 2 RCTs (5 publications) reported on combined peri-implantitis therapy (i.e., implantoplasty + reconstructive therapy) (2 RCTs (5 publications) [19, 70, 94,95,96]) (Table 3).

Table 3 Included studies reporting on surgical peri-implantitis treatment

Follow-up periods among the included studies varied from 6 months (6 studies), 1 year (12 studies), 3 to 4 years (4 studies), to 5 and 7 years (5 studies). Peri-implantitis was commonly defined by the presence of BOP/SUPP and a presence of radiographic bone loss, with the threshold values of ≥ 2 mm or > 3 mm being most frequently used. In fact, the majority of the studies (13 studies) reporting on reconstructive and combined peri-implantitis therapy indicated the presence of intrabony peri-implant defect configuration (Table 3). Twenty studies (29 publications) reported on patient engagement into a regular supportive therapy following the surgery [18, 19, 61, 63,64,65,66,67,68, 71, 74, 76,77,78, 80,81,82,83,84,85,86,87, 89,90,91,92, 94,95,96].

Efficacy of interventions

Adjunctive and alternative measures for implant surface decontamination following non-reconstructive treatment

Over the 6-month follow-up period, alternative measures for implant surface decontamination, including a titanium brush and an air-powder abrasive with glycine powder, were more effective in reducing signs of inflammation, as shown by a higher reduction in BOP and PD values over the implant sites treated with the conventional decontamination method (i.e., plastic curettes) [67]. Furthermore, sites treated with a titanium brush revealed significant improvements in marginal bone levels compared to the implants treated with either an air powder abrasive device or plastic curettes (i.e., control group). Nonetheless, as addressed by the authors, treatment success (i.e., PD ≤ 5 mm, no BOP, no bone loss ≥ 5 mm) was rarely obtained irrespective of the decontamination protocol (i.e., plastic curettes: 22% of implants; air-powder abrasive: 33% of implants; titanium brush: 33% of implants) [67].

Based on 2 RCTs, the adjunctive use of either a PDT or diode laser failed to reveal any beneficial clinical effect with respect to BOP and PD changes throughout the 6-month period [62, 63]. Three RCTs investigated the additional use of 0.2% CHX solution for implant surface decontamination [18, 61], and adjunctive decontamination using 0.12% CHX + 0.05% cetylpiridinium chloride (CPC) versus placebo [65], or 0.12% CHX + 0.05% CPC versus 2.0% CHX [66]. Over 1- to 3-year follow-up periods, the adjunctive use of the aforementioned antimicrobials as a part of implant surface decontamination protocol did not lead to improved clinical (i.e., BOP and PD) or radiographic outcomes compared with the respective controls [18, 61, 65, 66].

Adjunctive implantoplasty following non-reconstructive treatment

Two RCTs (3 publications) assessed the clinical efficacy of implantoplasty used as an adjunct to non-reconstructive therapy [71,72,73]. In particular, data from a 6-month RCT pointed to no differences in clinical (i.e., BOP and PD changes) and radiographic parameters between implant sites treated with either implantoplasty or air polishing with glycine powder [71]. A 3-year RCT, contrarily, indicated that adjunctive implantoplasty enhanced implant survival rates, significantly reduced PDs, SUPP, and BI, and was associated with stable marginal bone levels compared to the control sites, where bone loss amounted to 1.45–1.54 mm [72]. However, sites treated with implantoplasty resulted in significantly more soft tissue recession (test group: 2.3 [1.45] mm, control group: 1.64 [1.29] mm) [72, 73] (Table 3).

Adjunctive local and systemic antibiotics following non-reconstructive treatment

Based on 1 RCT, the repeated local applications of antibiotics (i.e., minocycline oinment 1, 3 and 6 months postoperatively) lead to significant benefits in terms of greater mean PD reduction and radiographic marginal bone levels compared to the control implant sites (i.e., mechanical debridement and air-powder polishing), while changes in BOP/SUPP were comparable between test and control groups [64].

Two RCTs investigated the potential beneficial effect of systemic antibiotics following non-reconstructive peri-implantitis treatment [18, 61, 74]. Specifically, over a 1-year period, the adjunctive administration of postoperative systemic antibiotics lead to similar clinical (i.e., changes in BOP and PD), radiographic (i.e., RBL) or microbiological treatment outcomes compared to the control group [74]. Based on the results of another RCT, a positive effect of systemic antibiotics on the success of treatment (i.e., PD ≤ 5 mm, no BOP/SUPP, bone loss ≤ 0.5 mm) during a 1-year period was observed only for implants with modified surface characteristics [61]. The benefits of the systemic antibiotic regimen, however, did not last through the 3-year follow-up, leading to similar changes in BOP, SUPP, PD and RBL values [18].

Adjunctive and alternative measures for implant surface decontamination following reconstructive therapy

Adjunctive use of ozone therapy for implants as part of implant surface decontamination protocol along with reconstructive peri-implantitis treatment over a 1-year period resulted in significantly greater peri-implant bone defect fill compared to decontamination with sterile saline solution (2.32 mm vs. 1.17 mm, respectively), whereas clinical outcomes (i.e., changes in BOP and PD) were comparable between test and control groups [68]. After 5 years of follow-up period, adjunctive application of CO₂ laser provided similar clinical (i.e., changes in BOP and PD) and radiographic treatment outcomes to the conventional decontamination approach (i.e., air polishing) [69].

Adjunctive and alternative measures for implant surface decontamination following combined therapy

Use of a titanium brush as an adjunct treatment to surface decontamination protocol (i.e., debridement with ultrasonic scaler + rinsing with H₂O₂ 3%) after 1 year resulted in significantly greater PD reduction compared to control implant sites, while BOP changes were similar in both treatment groups [70]. After 7 years of follow-up, implant surface decontamination by means of Er:YAG monotherapy following combined peri-implantitis therapy led to similar BOP and PD changes as to implant sites where conventional decontamination protocols were used (i.e., mechanical debridement + saline-soaked cotton gauze) [19] (Table 3).

Reconstruction of peri-implant bone defects with different bone fillers

After 12 months of healing, significantly higher RDF and mean PD reduction were obtained at peri-implantitis defects filled with xenogenic bone filler particles in comparison with autogenous bone, whereas BOP changes were similar for both reconstruction approaches [82]. In comparison with synthetic bone filler (i.e., nanocrystalline hydroxyapatite particles), the use of a bovine-derived xenograft after 4 years led to significantly greater BOP and PD improvements [83]. Increased RDF and higher BOP reduction were detected at implants treated with porous titanium granules compared with xenograft, whereas PD reduction and clinical attachment changes did not differ between the treatment groups [88]. The comparison of the 2 xenograft materials over a 12-month period led to similar treatment outcomes as depicted by similar changes in the BOP, PD, and RDF values, as well as the treatment success (defined as PD ≤ 5 mm + no BOP/SUPP + no further bone loss) [86].

Reconstruction of peri-implant bone defects with and without a membrane

One 3-year CCT reported significantly lower PD reduction and less RDF at implant sites treated with autogenous bone along with non-resorbable membrane compared with those treated with either autogenous bone alone or in combination with resorbable membrane [87]. Peri-implantitis defects reconstructed using bovine bone along with a collagen membrane after 4 years showed significantly lower BOP and PD values compared with the implant sites treated with synthetic bone filler (i.e., nanocrystalline hydroxyapatite particles) [83]. Another 5-year CCT indicated no beneficial effect of the adjunctive use of a synthetic resorbable membrane along with xenogenic bone substitute particles, as the changes in PD, ML, and RDF were comparable between the treatment groups [90]. Furthermore, the comparison of the 2 membranes (i.e., concentrated growth factor membrane and collagen membrane) applied over the xenogenic bone filler after 1 year resulted in similar BOP changes and comparable RDF, whereas a greater PD reduction was registered at sites treated with the adjunctive collagen membrane [85].

Reconstructive therapy versus non-reconstructive surgery

Six RCTs (7 studies) assessed the clinical efficacy of reconstructive therapy over access flap surgery [75,76,77,78,79,80,81, 93] (Table 4 b). One to 7 years following the treatment, a significantly higher RDF was observed at the implant sites treated with either titanium granules or xenograft bone filler, as compared with the control sites (i.e., access flap surgery) [75, 76, 78, 80, 93]. On a contrary, as noted in 2 RCTs with 6-month and 5-year follow-up periods, the adjunctive use of either enamel matrix protein (EMD) or platelet-rich fibrin (PRF) had no beneficial effect upon RDF changes [77, 79, 81]. In terms clinical outcomes, after 1- to 7-years of follow-up, the PD and BOP changes did not differ between the implant sites treated with either titanium porous granules or xenogenic bone filler particles and those obtained at the control sites [75, 76, 80, 93]. Two studies, in contrast, reported greater PD reduction after 1 year at implants treated with either adjunctive xenogenic bone substitute or PRF, while changes on BOP values were similar between the test and control groups [78, 79]. Regarding changes to soft-tissue levels, the use of xenogenic bone filler particles did not lead to superior ML outcomes after 1 year [78, 80], whereas implant sites treated with adjunctive PRF after 6 months showed significantly lower ML values as compared to the controls (test: 0.14 mm, control: 1.04 mm) [79].

Synthesis of results

Adjunctive implantoplasty following non-reconstructive treatment

A meta-analysis based on 2 RCTs indicated the WMD in PD of − 1.11 [SE = 0.48; p = 0.02; 95% CI (− 2.05, − 0.18)] (unit of analysis: implant); p value for heterogeneity: 0.429, I² = 0% = low heterogeneity), thus suggesting higher PD reduction at implant sites treated with implantoplasty [71,72,73]. The WMD in ML amounted to − 0.02 [SE = 0.28; p = 0.95; 95% CI (− 0.56, 0.53); unit of analysis: implant], pointing to no significant difference between test and control groups in terms of soft-tissue level changes (p value for heterogeneity: 0.99, I² = 0% = low heterogeneity) [71, 72] (Fig. 4a and b).

Fig. 4

Forest plot indicating weighted mean difference (95% CI) in the changes of clinical outcomes following non-reconstructive surgical treatment of peri-implantitis. a Adjunctive implantoplasty (implant-level analysis)—PD. b Adjunctive implantoplasty (implant-level analysis)—ML. c Adjunctive systemic antibiotics (implant-level analysis)—PD

Adjunctive systemic antibiotics following non-reconstructive treatment

Based on 2 RCTs with 1 year of follow-up, WMD in PD amounted to − 0.95 [SE = 0.83; p = 0.26; 95% CI (− 2.58, 0.69)]; unit of analysis: implant), thus not favoring administration of adjunctive systemic antibiotics following non-reconstructive peri-implantitis treatment (p value for heterogeneity: 0.009, I² = 85.39% = substantial heterogeneity; Fig. 4c).

Reconstructive therapy versus non-reconstructive surgery

The WMD in BOP reduction was − 11.11% [SE = 5.97; p = 0.11; 95% CI (− 24.77, 2.55)] (unit of analysis: implant), indicating no differences between reconstructive and non-reconstructive treatment approaches (p value for heterogeneity: 0.983, I² = 0% = low heterogeneity) [79, 97] (Fig. 5a). The WMD in PD revealed a significant difference between the test and control groups (WMD = − 0.51 mm [SE = 0.15; p = 0.01; 95% CI (− 0.81, − 0.20)] (unit of analysis: implant) that favored adjunctive reconstructive approaches (p value for heterogeneity: 0.28, I² = 21% = low heterogeneity) [78,79,80, 93] (Fig. 5b). The WMD in RDF amounted to − 56.46% [SE = 8.65; p = 0.01; 95% CI (− 73.42, − 39.50)] (unit of analysis: implant), pointing to a higher defect fill in the test group (p value for heterogeneity: 0.487, I² = 0% = low heterogeneity) [75, 76] (Fig. 5c). Based on data from 4 RCTs, the WMD in reduction of radiographic defects was − 1.47 mm [SE = 0.45; p = 0.01; 95% CI (− 2.36, − 0.59)] (unit of analysis: implant), suggesting significantly higher reduction in the test group (p value for heterogeneity: 0.389, I² = 0% = low heterogeneity) (Fig. 5d). The WMD in ML was − 0.63 mm [SE = 0.21; p = 0.01; 95% CI (− 1.05, − 0.21)] (unit of analysis: implant), favoring reconstructive measures (p value for heterogeneity: 0.579, I² = 0 = low heterogeneity) [79, 80] (Fig. 5e).

Fig. 5

Forest plots depicting weighted mean differences (95% CI) in the changes of primary and secondary outcomes between reconstructive and non-reconstructive peri-implantitis surgical treatment. a BOP reduction (implant-level analysis). b PD (mm; implant-level analysis). c RDF (%; implant-level analysis). d Radiographic defect reduction (mm; implant-level analysis). e ML (implant-level analysis)

Risk of bias in individual studies

Of the included 55 RCTs, 23 appeared to have an overall unclear risk of bias, 18 were judged to have a high risk of bias, and the remaining 14 had a low risk of bias (Additional file 2a).

Four of the included CCTs had an overall serious risk of bias, and the remaining 3 had an overall critical risk of bias (Additional file 2b).

Discussion

The present systematic review aimed to evaluate the efficacy of alternative and adjunctive measures compared to conventional treatment of peri-implant mucositis and peri-implantitis. In total, 55 RCTs and 7 CCTs were included in the analysis. Of those, 18 reported on non-surgical treatments of peri-implant mucositis, and 17 and 27 reported on non-surgical and surgical peri-implantitis treatments, respectively.

The primary goal of peri-implant mucositis treatment has been established as the resolution of inflammation as evidenced by the absence of BOP [98]. Based on the current data synthesis, the investigated alternative measures for biofilm removal (i.e., glycine powder air polishing and chitosan brushes) and adjunctive measures (i.e., diode laser, aPDT, local antiseptic therapy, probiotics, home care mouth rinse) failed to improve BOP scores over mechanical debridement alone. In terms of PD values, while the adjunctive use of local antiseptics (i.e., CHX and sodium hypochlorite) along with mechanical debridement led to significantly greater PD reduction (WMD = − 0.23 mm, p = 0.03, respectively), similar PD improvements were noted regardless of the implementation of the aforementioned adjunctive measures for biofilm removal, aPDT, probiotics or home care mouthrinse. The present findings partially align with the results of previous systematic reviews and meta-analyses according to which adjunctive measures for treating peri-implant mucositis (i.e., antiseptics, local and systemic antibiotics, air-abrasive devices) failed to improve the efficacy of professionally administered plaque removal in reducing clinical signs of inflammation, as shown by comparable changes in BOP and PD values [13, 99]. However, the calculations in those analyses were based on pooled data from clinical studies that employed both local and systemic adjunctive measures (i.e., local and systemic antibiotics), which in turn might at least partially explain the aforementioned discrepancies [13, 99]. Taken together, the use of investigated adjunctive and alternative measures were not found to be superior in resolving peri-implant mucositis, thus supporting recent consensus statements suggesting that non-surgical mechanical instrumentation in conjunction with oral hygiene reinforcement is a standard-of-care intervention for the management of peri-implant mucositis [4, 12, 100].

According to recent recommendations, results of peri-implantitis treatment should be assessed following a healing period of at least 6 months and should be based on a composite outcome, including parameters such as bone fill, peri‐implant soft tissue recession, PD, BOP, and SUPP [97, 98]. The present analysis included clinical studies reporting on peri-implantitis treatment outcomes with an observation period of at least 6 months [97]. In contrast to peri-implant mucositis, non-surgical treatment of peri-implantitis including alternative measures for biofilm removal (i.e., glycine powder air polishing, Er:YAG laser) yielded higher BOP reduction compared to conventional measures (i.e., mechanical debridement with or without CHX; WMD = − 28.09%; p = 0.01), whereas these improvements were not observed in PD scores (WMD = − 0.27 mm; p = 0.19). Additionally, BOP and PD reductions were not improved by the adjunctive use of local antiseptics/antibiotics (BOP WMD = − 10.65%, p = 0.06; PD WMD = − 0.25 mm, p = 0.16), nor by the use of probiotics (PD WMD = − 0.15, p = 0.35). Furthermore, soft-tissue level changes following treatment were similar regardless of whether alternative biofilm removal measures (WMD = − 0.21, p = 0.55) or local antiseptics (WMD = − 0.11, p = 0.22) were employed. The aforementioned findings corroborate the results of one former meta-analysis, which reported significantly greater BOP reduction at implant sites treated with either adjunctive local antibiotic therapy (i.e., minocycline microspheres) or alternative plaque removal measures (i.e., Er:YAG laser or glycine powder air polishing) over respective control treatments [13]. Further analysis revealed a significantly higher reduction in BOP and PD values throughout the 12-month period with administration of systemic antibiotics along with the mechanical debridement (WMD = − 17.35%; p = 0.01 and WMD =  − 1.46 mm; p = 0.01, respectively). However, this estimation is based on only 2 RCTs, one of which included only severe cases of peri-implantitis (case definition: BOP + PD > 5 mm + bone loss > 4 mm) and found no beneficial effect of systemic antibiotics (amoxicillin + metronidazole) following non-surgical peri-implantitis treatment [58]. Likewise, one recent RCT reported no clinical and microbiological benefits of systemic antibiotics (amoxicillin + metronidazole) along with non-surgical treatment of peri-implantitis (case definition: bone loss ≥ 2 mm + BOP/SUPP + PD ≥ 5 mm) compared to mechanical debridement and local CHX irrigation after 3 months, thus concluding that the administration of systemic antibiotics should not be routinely recommended [101]. Notably, the majority of the included studies reported on residual BOP/BI scores following non-surgical peri-implantitis treatment, and disease resolution (i.e., absence of BOP and further bone loss) was obtained in 14% to 47% of the cases 6 to 12 months after the treatment [45, 52, 56]. Therefore, in line with earlier findings, non-surgical treatment of peri-implantitis seems to have limited efficacy in predictably resolving inflammation, thus supporting the necessity of surgical treatment in the majority of patients diagnosed with peri-implantitis [12, 102]. Nonetheless, according to the recent recommendations, non-surgical therapy should always precede surgical intervention in treating peri-implantitis [102].

Due to heterogeneity in reporting, no quantitative analysis was feasible for the impact of implantoplasty on the resolution of peri-implant tissue inflammation (i.e., BOP/SUPP changes) following surgical non-reconstructive peri-implantitis treatment. Nonetheless, based on the present findings, though implant sites treated with or without implantoplasty resulted in similar postoperative changes in soft-tissue levels (WMD = − 0.02 mm, p = 0.95), significantly higher PD reduction was found at sites treated with adjunctive implantoplasty (WMD = − 1.11 mm, p = 0.02). With respect to the rationale for administration of systemic antibiotics following non-reconstructive peri-implantitis treatment, no differences in PD improvements were found between the test and control groups throughout the 12-month period (WMD = − 0.95 mm, p = 0.26). This latter finding supports the results of a 3-year RCT, which after 1 year observed positive effects of systemic antibiotics on the non-reconstructive peri-implantitis treatment success (i.e., PD ≤ 5 mm, no BOP/SUPP, bone loss ≤ 0.5 mm) at implants with a modified surface [61]. However, those benefits were not sustained over a 3-year period, thus not supporting the benefits of the systemic antibiotic regimen [18].

Six RCTs evaluated the potential beneficial effect of reconstructive peri-implantitis treatment over control approaches (i.e., access flap). In particular, meta-analyses identified a significantly higehr RDF (WMD = − 56.46%, p = 0.01), radiographic defect resolution (WMD = − 1.47 mm; p = 0.01) and greater PD reduction at the implant sites treated with adjunctive reconstructive measures compared to the controls (− 0.51 mm, p = 0.01). However, in terms of resolution of mucosal inflammation (i.e., BOP changes), no differences could be detected between the test and control groups (WMD = − 11.11%; p = 0.11). Those findings slightly contradict the results of previous meta-analyses that reported on radiographic bone-level gains and RDF for reconstructive treatment approaches over access flap surgery, whereas similar values were reported for PD and BOP changes [103, 104]. Nonetheless, noteworthy are the discrepancies among the studies included in the present meta-analysis with respect to grafting materials with different radiopacities and osteoconduction properties, which might have influenced the obtained outcomes. Upon further data analysis, implant sites treated with adjunctive reconstructive measures yielded lower postoperative changes soft-tissue recession compared to sites treated via access flap surgery (WMD = − 0.63 mm; p = 0.01). This latter outcome corroborates the results of one recent meta-analysis, according to which use of adjunctive reconstructive measures lead to significantly lower increase in mucosal recession when compared to non-reconstructive peri-implantitis treatment (WMD = − 1.35 mm, p = 0.038) [104].

Along these lines, it is worthwhile to note that the treatment outcomes of peri-implant mucositis and peri-implantitis might be influenced by the surface characteristics of the abutment and/or implant. In fact, clinical data have reported greater BOP reduction following the treatment of experimentally induced peri-implant mucositis lesions at implants with machined abutments, as compared to the modified surfaced abutments [105]. As documented by the previous analyses, significantly better outcomes were obtained after surgical non-reconstructive therapy of peri-implantitis at implants with non-modified surfaces compared to modified surfaces, as shown by the superior BOP, PD reductions and superior bone-level preservation at non-modified surfaced implants [15, 18]. Additionally, more favorable clinical and radiographic outcomes of surgical reconstructive peri-implantitis therapy were documented for moderately rough surfaced implants compared to rough surfaced implants [106]. The results of a majority of the studies included in the present analysis were based on implants with modified surfaces. Thus, due to the limited data availability, subanalyses to validate the extent to which implant/abutment surface properties might have influenced the treatment outcomes of peri-implant mucositis and peri-implantitis were not feasible.

Several limitations of the present systematic review must be addressed. First, a majority of the included studies lacked true control groups and therefore could not be included in the quantitative analysis. Second, most studies included in meta-analysis had follow-up periods that were limited to 12 months, thus the present findings are valid only for the short-term outcomes. Further, the present analysis pooled clinical studies that applied different case definitions for peri-implant mucositis and peri-implantitis. In fact, depending on the individual protocols used, factors such as peri-implant bone defect morphology and severity of the disease have previously been found to be influencing factors for the outcomes following surgical treatment of peri-implantitis [105,106,107,108]. Finally, peri-implant soft-tissue conditions (i.e., presence or lack of keratinized mucosa), patients` adherence to supportive therapy following peri-implant mucositis and peri-implantitis treatment as well as patient-related factors, such as smoking habits, systemic conditions (i.e., diabetes) and intake of different medications may also be important factors contributing to the outcomes of therapy. However, in the present analysis, due to inconsistencies in reporting among the studies, potential effects of these factors on treatment outcomes of peri-implant diseases could not be investigated.

Conclusions

Alternative and adjunctive measures provided no beneficial effect in resolving peri-implant mucositis, while alternative measures were superior in reducing BOP values following non-surgical peri-implantitis treatment. Adjunctive reconstructive measures along with surgical peri-implantitis treatment were beneficial regarding radiographic bone-defect fill/reduction, PD reduction and lower soft-tissue recession, although they did not improve the resolution of mucosal inflammation. Systemic antibiotics added no benefits to surgical non-reconstructive peri-implantitis treatment outcomes. The potential benefits of resective measures upon inflammation resolution need to be further investigated.