A Systematic Review of the Effect of Oral Rinsing with H2O2 on Clinical and Microbiological Parameters Related to Plaque, Gingivitis, and Microbes

Background Hydrogen peroxide (H2O2) has been used for more than a century clinically to control plaque and gingival inflammation, with unclear supporting evidence. Aim The aim of the present systematic review of the literature is to assess the effect of mouth rinses with H2O2 on dental plaque, gingival inflammation, and oral microorganisms. Methods Five databases (PubMed, Scopus, Embase, Cochrane Library, and Web of Science) were searched with the following focused question: what is the effect of hydrogen peroxide, in comparison to chlorhexidine or to a placebo solution, in oral microbiota control, dental plaque, and gingival inflammatory outcomes? Two independent examiners retrieved the articles and evaluated the evidence. Results The majority of included studies were performed with 1.5% H2O2. Results related to plaque accumulation generally demonstrate a slightly better effect of H2O2 as compared to placebo mouth rinses, however with a lower performance as compared to chlorhexidine. In terms of gingival inflammation, H2O2 performs better than placebo and more clearly demonstrates an anti-inflammation effect. No studies evaluated the effect of H2O2 against viruses or fungi. In terms of bacteria, H2O2 demonstrates an antibacterial effect. Conclusion Rinsing with H2O2 has the potential to affect plaque, gingivitis, and oral bacteria, as compared to placebo. However, the antibacterial results are not comparable to the performance of chlorhexidine.


Introduction
Hydrogen peroxide (H 2 O 2 ) mouthwashes have been used for a long time [1]. ey have been utilized in an attempt to complement mechanical plaque control methods as well as to prevent/control oral infections [2]. However, the evidence supporting its use is not unequivocal even though it is still used by a number of professionals.
In 2011, Hossainian et al. [3] published a systematic review to evaluate the effect of H 2 O 2 mouthwashes on the prevention of plaque and gingival inflammation. e focused question of such review was as follows: "what are the effects of oxygenating mouthwashes on plaque accumulation and gingival inflammation parameters in adults, when compared with positive or negative controls mouthwashes or no oral hygiene, when used as a monotherapy or as an adjunct in daily oral hygiene?" Surprisingly, the number of included studies was relatively low (n � 12, in which only 5 were specifically formulated with H 2 O 2 ). e other 7 studies were related to other oxygenating agents. e results of the review demonstrated that mouthwashes containing H 2 O 2 do not consistently prevent plaque accumulation when used as a short-term monotherapy. mouth rinses are being recommended as a preprocedural rinse, as well as a regular rinsing solution with the aim of diminishing contamination possibilities by the new coronavirus. A number of associations, including the American Dental Association, are recommending the use of H 2 O 2 mouth rinses as prerinses prior to procedures [4][5][6]. However, the evidence for the use of H 2 O 2 for oral antiviral purposes is virtually nonexistent. Most of the evidence only demonstrates the potential of H 2 O 2 to disinfect surfaces [7].
New studies have been conducted and published after the systematic review of Hossainian et al. [3]. Hence, the existing review [3] could be broadened to consider these additional studies, especially in this particular moment. erefore, the aim of this study is to systematically review the literature, assessing the effects of H 2 O 2 mouth rinses in controlling dental plaque, gingival inflammation, and oral microbiota.

Materials and Methods
e focused question of the present study was as follows: "what is the effect of hydrogen peroxide, in comparison to chlorhexidine or to a placebo solution, in oral microbiota control, dental plaque, and gingival inflammatory outcomes?" In order to be included, the study must fulfill all of the following inclusion criteria: (i) Clinical trials with humans of any age. (ii) Test group: individuals that used, at least one time per day, hydrogen peroxide mouthwash. Any concentration of hydrogen peroxide was accepted. (iii) Control group: individuals that used, at least one time per day, a placebo or chlorhexidine mouthwash. Any concentration of chlorhexidine was accepted. (iv) Outcomes: any oral microbiological, plaque index, or gingival index analysis.
No restriction to language or date of publication was imposed. Studies that used both chlorhexidine and hydrogen peroxide in the same group were excluded. Studies that involved outcomes assessed in dental implants were also excluded.
A search strategy was performed, up to April 23, 2020 An adaptation of the abovementioned search strategy was performed in the other databases. Two researchers independently selected the studies and extracted the data in a spreadsheet specifically developed for this study (CKR and FWMGM). Regarding study selection, both screening (title and abstract analyses) and eligibility (full-text analysis) phases were performed independently. All discrepancies were solved between the researchers by discussion.
In addition, hand search was performed in the list of references of the included studies and in previously published literature reviews [3]. Studies from the last 30 years were hand searched in the following journals: Journal of Clinical Periodontology, Journal of Periodontology, Journal of Periodontal Research, and Journal of Dental Research. Searches for grey literature were also performed in the Clinical Trials (clinicaltrials.gov) and Google Scholar databases, using an adaptation of the abovementioned search strategy. All the corresponding authors of the included studies were contacted by e-mail in order to detect other potentially relevant clinical trials. In addition, manufactures were contacted to supply information about other published, unpublished, or ongoing research studies using H 2 O 2 .
e risk of bias of all randomized clinical trials was assessed by the RoB2 tool, as recommended by Cochrane [8]. Randomization process, deviations from the intended interventions, missing outcome data, measurement of the outcome, selection of the reported results, and overall risk of bias were assessed by two reviewers (FWMGM and CKR). In case of disagreements, a consensus was made between reviewers. A positive sign was given for an item when sufficient information was available, indicating low risk of bias, and a negative mark was used, for high risk of bias, when information was lacking. When risk of bias could not be assessed, the item was classified as unclear. For the nonrandomized trials, the ROBINS-I tool was used [9]. Several sources of bias were assessed, such as confounding, selection of participants, classification of interventions, deviations from intended interventions, missing data, measurement of outcomes, selection of the reported result, and overall bias.

Main Characteristics of the Included Studies.
e search strategy and flowchart of articles retrieval is demonstrated in Figure 1. e reasons for exclusion of the identified studies are reported in Table S1. It should be noted that five databases were searched. e additional search strategies did not add any study to the present review, except for one additional study identified in the Google Scholar database [10]. e retrieved studies were very distinct in all aspects, preventing the possibility of a meta-analysis. erefore, descriptive information will be given, according to design and outcome (experimental gingivitis or not; plaque, gingival inflammation, or microbiological parameters). Figure 2 demonstrates the analysis of risk bias of the randomized clinical trials included in this review according to the RoB2 instrument. It may be detected that only one study presented low risk of bias in all criteria analyzed [11]. Four other studies presented an overall high risk of bias [10,[12][13][14]. e criteria randomization process showed unclear risk of bias in almost all included studies. e other sources of bias comprised mainly lack of information of reproducibility. e risk of bias for the nonrandomized trials included in the present review is demonstrated in Table 1. Bias due to confounding was critical in all studies. e other analyses mostly demonstrate moderate to low risk of bias.
e plaque index measurements are demonstrated in Table 2. It should be highlighted that 10 studies evaluated plaque parameters, of which six were performed as clinical trials allowing mechanical plaque control [11,12,15,[19][20][21]. In one, it was not possible to determine if mechanical plaque control was possible [10]. All studies that used H 2 O 2 as adjunct to mechanical oral hygiene were performed with the concentration of 1.5% . ose studies demonstrate a higher antiplaque efficacy of chlorhexidine in comparison to H 2 O 2 , except two, in which H 2 O 2 presented similar efficacy to chlorhexidine [10,21]. Generally, very little differences from negative controls were detected.
(2) Experimental Gingivitis Studies. Among the studies that evaluated plaque parameters, 3 used the experimental gingivitis model [16,18,22]. e information coming from these studies gives an idea of efficacy of the mouth rinses in undisturbed dental biofilms.
is enhances the proofs of principle of the antiplaque effect, which would give useful information, e.g., for areas where mechanical plaque control is not effective.

Gingival Inflammation
(1) Nonexperimental Gingivitis Studies. e results related to gingival inflammatory parameters are presented in Table 3. Nine studies were included in this outcome [10-12, 15, 16, 18-21], six allowed mechanical plaque control [11,12,15,[19][20][21]. In one study, it was not clear if mechanical plaque control was allowed [10]. In these parameters, H 2 O 2 mouth rinse performs better than negative controls, however less than chlorhexidine. A possibility of a decrease in inflammation could be raised since it seems that H 2 O 2 performs better in terms of gingivitis than it does in relation to plaque.
(2) Experimental Gingivitis Studies. Among the studies that evaluated gingival inflammation, two were based on the experimental gingivitis model [16,18]. e experimental gingivitis model provides information on the effect of the mouth rinse in areas in which plaque control is not adequate. Also, it rules out the eventual confounding effect of the adjunct plaque control in the study of chemical substances.
(2) Experimental Gingivitis Studies. Two studies performed microbiological analysis using an experimental gingivitis design [16,18]. e information coming from such studies supports the quality/quantity of different germs when plaque is accumulating overtime. It also rules out the effect of the uncontrolled mechanical plaque removal.

Qualitative
Results-Safety. Among the 13 included studies, only five of them assessed for side effects. All of these five studies reported no side effects in individuals that used H 2 O 2 mouthwashes [11,[14][15][16]19]. Additionally, no side effects were reported in those that used chlorhexidine [14,16]. Conversely, an increased tendency for desquamation of the mucosal lining was reported in individuals that used a placebo solution [16]. e other studies that used a negative control group reported no side effect in this group [11,14,15,19].   Excluded articles Hydrogen peroxide was not used = 8 Hydrogen peroxide and others substance in the same group = 7 Absesnce of control group = 2 In vitro study = 4 Mouthwash was not used = 2 Outcomes were not assessed = 4 Study with dental implants = 1 Review = 2 Not found = 11 Study protocol (data not provided or not available) = 2    International Journal of Dentistry    International Journal of Dentistry

Studies with experimental gingivitis
International Journal of Dentistry       which also performed better than distilled sterile water International Journal of Dentistry 13  Regarding the risk of bias of both randomized and nonrandomized clinical trials, it is important to highlight that most of the included studies presented an unclear or high risk of bias. Only one study demonstrated an overall low risk of bias [11]. is randomized clinical trial demonstrated a superior antigingivitis efficacy of H 2 O 2 mouthwash in comparison to a placebo solution. However, no significant difference was observed for the antiplaque efficacy.
e overall high risk of bias must be put into perspective when interpreting the results of the present study. is means that the use in clinical practice should be indicated with caution and not performed routinely since the support is not robust. e strengths of the present systematic review were based on the importance of the topic, especially because the mouth is a very contaminated cavity and mouth rinses are used to reduce different degrees of contamination. In addition, with the COVID-19 pandemic, the use of mouth rinses has been considered an additional way for reducing all sorts of contamination. e limitations are related to the quality of the evidence. erefore, the information contained herein should be cautiously interpreted. Also, in an attempt to decrease the time for publication of this information, no registration was performed and it was not possible to make a post hoc registration.

Quality of Evidence and Strength of Recommendation.
Initially, the focused question included both a negative and a positive control group. e negative control could be either placebo, water, or no solution, whereas the control group should include the gold standard in terms of oral rinsechlorhexidine. e results of this systematic review should be put into the perspective that H 2 O 2 is widely used in oral care despite the lack of a large number of studies, especially in some of the aforementioned indications. We looked at the systematic review published by Hossainian et al. [3] that critically appraised the evidence until the beginning of this decade. Such work led to the conclusion that H 2 O 2 does not consistently prevent plaque accumulation in short-term periods. erefore, we expanded the search criteria, not restricting age, including microbiological parameters, updating the publication year to 2020, and including five databases instead of the two previously searched databases. Due to the higher usage of H 2 O 2 , we restricted the search to only include this substance and not any other oxygenating agent.
H 2 O 2 has been used clinically for more than a century, and recently, H 2 O 2 containing mouth rinse are being recommended, especially due to a possible antiviral effect and the pandemic of COVID-19. To the best of the authors' knowledge and making a systematic search in the same databases, no studies have observed any antiviral effect of H 2 O 2 in the mouth. However, associations are supporting its use [5,6]. e present systematic review used the most strict quality criteria for retrieving the studies. However, the interpretation will be contextualized in the moment that the world is facing a pandemic in which any kind of effort should be at least collated to make the sense of any preventive guideline.
In terms of plaque, one study [15] was performed in adolescents and the others in adults. One of them also included handicapped individuals [20]. Four of the six studies that allowed oral hygiene compared 1.5% H 2 O 2 with a negative control [11,15,19,20] and 2 of them with chlorhexidine [12,21], and one of them was also compared to a negative control [12]. Among the studies that used the experimental gingivitis model [16,18,22], two were compared with a negative control [18,22] and the other included a positive control [16]. In one study, the effect of H 2 O 2 was compared to chlorhexidine, but it was not possible to determine if mechanical plaque control was allowed [10]. In these studies, different concentrations of H 2 O 2 were used.
It is clear from the encountered results that 1.5% H 2 O 2 is the most studied concentration in the formula of a mouth rinse.
is result is in accordance with the previously published review [3]. For the publications evaluating the effect of H 2 O 2 on plaque, only one study (which evaluated the antiplaque effect over an 18-month time period) demonstrated improved results when compared to a placebo [15]. e other studies, which evaluated the effect over shorter periods, did not find statistically significant differences. Also, in the studies that used the experimental gingivitis model, only one study demonstrated the superiority of H 2 O 2 in comparison to placebo [18]. e same cited publications that evaluated plaque also evaluated the effect of H 2 O 2 on gingival inflammation. Although only a single study demonstrated the antiplaque benefit of H 2 O 2 , more studies clearly point to a better antigingivitis effect of H 2 O 2 mouth rinses as compared to placebo [11,15,20]. In fact, for one of the studies, no difference was observed between the H 2 O 2 mouth rinse and the positive control [12]. Because the participants of these studies were allowed for routine mechanical oral hygiene, an effect on clinical inflammation alone (without having the associated plaque reduction benefit) should be highlighted. ese results suggest that H 2 O 2 might perform differently in terms of plaque and gingivitis, which is of great clinical interest.
Also, it is of high importance to evaluate the effect of mouth rinses on the oral microbiome. is includes not only bacteria but also other germs, such as viruses and fungi. However, despite completing a broad search of the literature, no studies were identified that evaluated the effect of H 2 O 2 oral microorganisms other than bacteria. e comparisons of the effect of rinses on oral bacteria with H 2 O 2 and with the positive control generally demonstrate a better effect of the latter. However, the differences in terms of the composition of the oral microbiome when H 2 O 2 is compared to placebo are clear in a variety of bacterial species. e present study evaluated risk of bias both for the nonrandomized and randomized trials. As expected, the randomized clinical trials presented a higher quality, with decreased risk of bias. e nonrandomized studies in general present a higher risk of bias. is is inherent to the chosen design. Randomized studies tend to present a lower risk of bias. 16 International Journal of Dentistry A systematic review was recently published by Marui et al. [23] describing the effect of preprocedural rinses with different substances on dental office-generated aerosols. ey demonstrated that rinses with chlorhexidine, essential oils, and cetylpyridinium chloride are effective. No studies with H 2 O 2 were included.

Implications for Further
Research. Meanwhile, taking into consideration the precautionary principle [24], even without the qualified evidence, due to the high levels of morbimortality, it is of interest to see other potentials of the use of H 2 O 2 . In such conditions, the use of "collateral evidence" is recommended, always with a surveillance look. erefore, in the present moment, further studies including oral rinses with H 2 O 2 and other substances are warranted. Studies with the antiviral effect of H 2 O 2 are also needed.

Conclusions
In conclusion, rinsing with 1.5% H 2 O 2 has demonstrated an antigingivitis effect as compared to placebo, with also greater reductions in oral bacteria. Chlorhexidine has demonstrated, up to now, the best antiplaque and antigingivitis effect on the oral microbiome.

Data Availability
e data supporting the current study are available from the corresponding author upon request.