Povidone-Iodine as a Pre-Procedural Mouthwash to Reduce the Salivary Viral Load of SARS-CoV-2: A Systematic Review of Randomized Controlled Trials

The use of pre-procedural rinses has been investigated to reduce the number of viral particles and bacteria in aerosols, potentially decreasing the risk of cross-infection from severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) during medical and dental procedures. This review aims to confirm whether there is evidence in the literature describing a reduction in salivary load of SARS-CoV-2 when povidone-iodine (PVP-I) is used as a pre-intervention mouthwash. An search of the MEDLINE, Embase, SCOPUS, and the Cochrane library databases was conducted. The criteria used followed the PRISMA® Statement guidelines. Randomized controlled trials investigating the reduction of salivary load of SARS-CoV-2 using PVP-I were included. Ultimately, four articles were included that met the established criteria. According to the current evidence, PVP-I is effective against SARS-CoV-2 in saliva and could be implemented as a rinse before interventions to decrease the risk of cross-infection in healthcare settings.


Introduction
Aerosols are defined as particles of liquid or solid in gas and are ≤5 µm in diameter [1].Due to their small size, they are inspirable and can remain suspended in the air for hours [2].These particles are generated by aerosol-generating procedures (AGPs) in medical settings, including airway suctioning, bronchoscopies, and high-flow oxygen therapy, among many others [3].Subsequently, aerosols in dental offices are generated by the frequent use of high-speed handpieces, ultrasonic devices, and 3-in-1 air-water syringes.For this reason, dentists are one of the collectives that have the highest risk of infection of COVID-19 due to the close proximity with the patients' oral cavities and the numerous AGPs performed routinely [4].Saliva and blood are main components for viral and bacterial spread; therefore, procedures that generate aerosols should be minimized.However, dental clinicians have a particularly limited range of options regarding treatments and armamentarium that do not generate aerosols [4].
The primary mode of transmission of severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) is through aerosols and respiratory droplets generated during daily activities, such as speaking or coughing [5].Several factors, including the immune response of the host, the pathogenicity of the virus, and the amount of infected particles, determine the susceptibility of being infected via an aerosol [6][7][8].Furthermore, it has been demonstrated that COVID-19-positive patients present high viral loads in saliva [9,10]; therefore, healthcare professionals performing AGPs have a greater risk of becoming infected with SARS-CoV-2 [11].
For this reason, there have been several investigations attempting to mitigate the negative effects of aerosols during AGPs.Pre-procedural rinses have been explored to reduce the salivary load of different microorganisms and the colony-forming units (CFUs) in aerosols, which could potentially decrease the risk of cross-infection during medical and dental procedures [9,[12][13][14][15].One of the most predominant mouthwash solutions studied is povidone-iodine (PVP-I).This molecule is an iodophor globally used due to its broad-spectrum antiseptic properties with a low number of contraindications, including allergy to iodine, thyroid disease, and pregnancy [16].Several in vitro studies and, more recently, randomized controlled trials (RCTs) using PVP-I as pre-procedural mouthwash have published their results.
Therefore, this study aimed to investigate the effectiveness of PVP-I used as a mouthwash to decrease the salivary viral load of SARS-CoV-2.

Protocol
This review was performed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA ® ) Statement [17,18].The protocol was registered at the International Prospective Register of Systematic Reviews (PROSPERO) under the registration number CRD42022303756.

Focused Question
A PICO (P, population/patient/problem; I, intervention; C, comparison; O, outcome) question was formulated based on the PRISMA ® guidelines: "In patients diagnosed with COVID-19 (P), does the use of PVP-I mouthwashes (I) compared to not prescribing them (C) reduce the viral load present in saliva (O)?"

Eligibility Criteria
Prior to the search, inclusion and exclusion criteria were defined:

Inclusion Criteria
Included studies were (a) RCTs; (b) studies in which the participants had a reversetranscription polymerase chain reaction (RT-PCR) examination positive for SARS-CoV-2; (c) studies that used PVP-I as a form of intervention; and (d) studies published in English.

Exclusion Criteria
Excluded studies were the following: (a) animal studies; (b) experimental laboratory studies; (c) studies whose study base focused on other areas besides the oral cavity and/or oropharynx; (d) studies that did not evaluate the reduction of viral load in saliva; (e) non-RCTs; (f) systematic reviews and meta-analyses; (g) literature review studies; (h) case reports; (i) letters to the editor; (j) abstracts or conference papers; (k) comments; and (l) unpublished articles.

Information Sources and Search Strategy
The search was conducted in four different electronic databases: MEDLINE (via PubMed), Embase, SCOPUS, and the Cochrane library database.
The search strategy was carried out by two researchers independently (A.G.-S.and A.-O.S.-P.).The search was not time-restricted and was updated to January 2022.MeSH (Medical Subjects Headings) terms, keywords, and other free terms were used with Boolean operators (OR, AND) to merge searches: ('povidone' OR 'povidone-iodine' OR 'polyvidone iodine' OR iodopovidone' OR 'PVP-I' OR 'iodine') AND ('COVID-19' OR 'SARS-CoV-2' OR 'SARS').These keywords were implemented in all databases according to the syntax rules of each database.

Study Records
The results were independently compared by two authors (A.G.-S.and A.-O.S.-P.) to guarantee completeness and removal of duplicates.Next, the title and abstract of the remaining articles were screened individually.Ultimately, full-text papers to be included in this study were selected following the criteria previously described.Disagreements over eligible articles were resolved by including a third author (J.-F.P.-C.) to reach a consensus.

Risk of Bias Assessment
The methodology of eligible studies was evaluated following the Joanna Briggs Institute (JBI) Critical Appraisal Tool [19] by two independent authors (A.G.-S.and A.-O.S.-P.).The studies were categorized as low-quality (0-7 domains) or high-quality assessment (8-13 domains).A third author (J.-F.P.-C) was included to resolve any disagreements between the two authors.

Study Selection
The search strategy resulted in 630 articles.There were 218 duplicates; therefore, 412 remained.Then, two authors (A.G.-S.and A.-O.S.-P.) independently examined the titles and abstracts and excluded 375 articles that were beyond the scope of this study.Therefore, we obtained 37 possible references.After reading the full text of those 37 papers, 33 were excluded because they investigated areas other than oral cavity saliva and/or oropharyngeal saliva (n = 6) or were experimental laboratory studies (n = 9), systematic reviews (n = 1), literature reviews (n = 2), letters to the editor (n = 12), and commentaries (n = 3).Therefore, four studies were included in our systematic review (Figure 1).

Study Characteristics
All the studies included were RCTs published in 2020 and 2021.There are some discrepancies in the sample size of the different articles (ranging from 36 to 84).Due to the low number of studies available, it was decided that there would not be exclusion criteria set for a minimum of participants.The total number of patients included within the studies was 221.All patients recruited had a positive RT-PCR examination for SARS-CoV-2.
In these studies, rinsing times ranged between 30 s and 1 min.In the placebo groups, distilled water [20][21][22] and saline [23] were used.In the test group, several concentrations of PVP-I were used: 0.50% [21,23], 1% [22], and 2% [20].All saliva samples were evaluated with RT-PCR.Baseline samples were collected immediately before the interventions.The number of saliva samples after interventions varies among the studies.One study collected one sample of saliva after intervention [22], one collected two samples [23], and two collected three samples [20,21].A summary of the findings of the included articles is described in Table 1.There were no differences in the reduction of salivary load in all intervention groups.When compared with the control group, PVP-I and CPC showed a significant decrease at 6 h.CPC also showed a significant reduction at 5 min.
The central findings of the resulting articles are described as follows: Chaudhary et al. [23] (2021) evaluated the effect of 0.50% PVP-I, 1% hydrogen peroxide (HP), 0.12% chlorhexidine (CHX), and normal saline.Forty patients were randomly allocated into groups, and saliva samples were collected and tested with RT-PCR.In the PVP-I group, there was a median reduction of 61% and 97% at 15 and 45 min, respectively.
Seneviratne et al. [21] (2020) recruited 36 patients that were randomly allocated to four different groups: distilled water (control), 0.50% PVP-I, 0.20% CHX, and 0.075% cetylpyridinium chloride (CPC).Samples were collected before rinsing, 5 min, and 3 and 6 h post-rinse.Cycle threshold (Ct) changes were estimated at each time-point value.The PVP-I group exhibited greater changes in Ct values after 5 min and 3 h; however, there was a significant increase in the virucidal activity only at 6 h when compared to distilled water.
Ferrer et al. [20] (2021) evaluated the differences in viral load of SARS-CoV-2 in 80 participants using 2% PVP-I, 1% HP, 0.07% CPC, 0.12% CHX, and distilled water (control).Samples were collected at baseline, 30, 60, and 120 min post-rinse.There was not a statistically significant reduction in salivary load when 2% PVP-I was used compared with the control group.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.[19] for RCTs.Reprinted with permission from JBI.Copyright 2020.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.[19] for RCTs.Reprinted with permission from JBI. Copyr 2020.signment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.Elzein e (20 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.
Table 2. JBI Critical Appraisal Tool [19] for RCTs.Reprinted with permission from JBI.Copyright 2020.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.
Table 2. JBI Critical Appraisal Tool [19] for RCTs.Reprinted with permission from JBI.Copyright 2020.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Seneviratne et al. [21] (2020)
Elzein e (20 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed descrip-tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.201.Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the rece recognition and cell membrane fusion process.The S protein mediates cell entry wh contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa salivary gland epithelium present a great amount of these receptors [25][26][27].In a stud Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE pressing ducts of salivary glands and in epithelial cells of the oral mucosa.They Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.

Seneviratne et al. [21] (2020)
Elzein e (20 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) in recognition and cell membrane fusion process.The S protein med contacts the angiotensin-converting enzyme 2 (ACE2) receptors [2 salivary gland epithelium present a great amount of these recepto Huang et al. [28], RNA molecules of SARS-CoV-2 were consiste pressing ducts of salivary glands and in epithelial cells of the 12.
Was appropriate statistical analysis used?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.
Table 2. JBI Critical Appraisal Tool [19] for RCTs.Reprinted with permission from JBI.Copyright 2020.[20] (2021) Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the rece recognition and cell membrane fusion process.The S protein mediates cell entry wh contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa salivary gland epithelium present a great amount of these receptors [25][26][27].In a stud Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE pressing ducts of salivary glands and in epithelial cells of the oral mucosa.They

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.Elzein e (20 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) in recognition and cell membrane fusion process.The S protein med contacts the angiotensin-converting enzyme 2 (ACE2) receptors [2 salivary gland epithelium present a great amount of these recepto Huang et al. [28], RNA molecules of SARS-CoV-2 were consiste pressing ducts of salivary glands and in epithelial cells of the 13.
Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.[20] (2021) 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none o articles presented a low-quality assessment (0-7 domains), and all of the articles inclu [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed des tion of the studies included.Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?

Critical Appraisal
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the rece recognition and cell membrane fusion process.The S protein mediates cell entry wh contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa salivary gland epithelium present a great amount of these receptors [25][26][27].In a stud Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE pressing ducts of salivary glands and in epithelial cells of the oral mucosa.They

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we deter articles presented a low-quality assessment (0-7 domains), and all [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 sh tion of the studies included.Elzein e (20 1. Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?
12. Was appropriate statistical analysis used?
13. Was the trial design appropriate and any deviations from the standard RCT design accounted for in the conduct and analysis of the trial?

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) in recognition and cell membrane fusion process.The S protein med contacts the angiotensin-converting enzyme 2 (ACE2) receptors [2 salivary gland epithelium present a great amount of these recepto Huang et al. [28], RNA molecules of SARS-CoV-2 were consiste pressing ducts of salivary glands and in epithelial cells of the PEER REVIEW 6 of 10

Bias Assessment
ing the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the presented a low-quality assessment (0-7 domains), and all of the articles included had a high-quality assessment (8-13 domains).Table 2 shows a detailed descriphe studies included.
JBI Critical Appraisal Tool [19] for RCTs.Reprinted with permission from JBI.Copyright ssion e surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor ion and cell membrane fusion process.The S protein mediates cell entry when it the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and gland epithelium present a great amount of these receptors [25][26][27].In a study by et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-exducts of salivary glands and in epithelial cells of the oral mucosa.They also = yes,

Risk Bias Assessment
Using the JBI Critical Appraisal Tool for RCTs [19], we determined that none of the articles presented a low-quality assessment (0-7 domains), and all of the articles included [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also = no, [20][21][22][23] had a high-quality assessment (8-13 domains).Table 2 shows a detailed description of the studies included.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also = uncertain.

Discussion
The surface of SARS-CoV-2 presents a spike protein (S) involved in the receptor recognition and cell membrane fusion process.The S protein mediates cell entry when it contacts the angiotensin-converting enzyme 2 (ACE2) receptors [24], and oral mucosa and salivary gland epithelium present a great amount of these receptors [25][26][27].In a study by Huang et al. [28], RNA molecules of SARS-CoV-2 were consistently found in ACE2-expressing ducts of salivary glands and in epithelial cells of the oral mucosa.They also proposed that the virus replicating in infected glands and the shedding of the infected oral mucosa are the sources of SARS-CoV-2 in saliva [28].
Patients undergoing medical and dental procedures were thoroughly screened for COVID-19 signs and symptoms as a means to prevent the risk of infection of healthcare providers.However, SARS-CoV-2 mRNA was detected in saliva of asymptomatic/presymptomatic patients [28].For that reason, it might be beneficial to use mouthwashes, such as PVP-I, to decrease the risk of cross-infection in healthcare settings where AGPs are performed in both confirmed COVID-19 and asymptomatic patients.
PVP-I efficacy in reducing the salivary viral load was compared to other solutions in the articles included in this review.Chaudhary et al. [23] found that reduction in viral load at 15 and 45 min did not differ among 1% HP, 0.12% CHX, and 0.50% PVP-I.Similarly, Elzein et al. [22] did not find any significant difference in the reduction of salivary load between 0.20% CHX and 1% PVP-I, and both were significantly effective compared to distilled water.In the RCT by Seneviratne et al. [21], salivary Ct values within all groups at the different time periods did not demonstrate any significant differences.Nonetheless, compared to distilled water, CPC was significantly more effective at 5 min and 6 h, while PVP-I was only significantly more effective at the 6-h mark.Ferrer et al. [20] found no statistically significant changes in salivary load of SARS-CoV-2 in any of the mouthwashes evaluated.However, comparing the loads at baseline and after intervention, PVP-I and CPC groups showed mean reductions of 30%, with the highest activity 2 h after intervention.None of the articles included showed any complications after oral PVP-I use at different concentrations.
Various in vitro studies have also assessed the virucidal activity of PVP-I.Many studies used the logarithmic reductions scales of viral load in their results.As a reference, a 3log 10 reduction equals a 99.90% reduction, and a 4log 10 reduction equals a 99.99% reduction in viral load.Xu et al. [29] found a virucidal activity of >3log 10 with a contact time of 30 min.Hassandarvish et al. [30] evaluated the reduction in salivary viral load using PVP-I at concentrations of 1% and 0.50%, which resulted in virucidal activities of >5log 10 (> 99.99%) at 15 and 30 s, respectively.Similar studies found virucidal activities of >4log 10 at 15 [31], 30 [32], and 60 s [33,34].
When evaluating PVP-I as a nasal rinse, Frank et al. [35] showed a complete inactivation of the virus using 0.50% PVP-I with a contact time of 15 s.Furthermore, a RCT study by Guenezan et al. [14] evaluated the reductions of viral titers in the nasopharynx using a 1% PVP-I rinse followed by 1% PVP-I nasal spray and an application of a 10% PVP-I balm over nasal mucosa during 7 days.The mean reductions in salivary load were 75% at day 1 compared with a reduction of 32% in the placebo group.However, there was no difference in the reduction of viral load over 7 days.
The results of this systematic review show that PVP-I is a potentially effective preprocedural mouthwash to decrease the salivary viral load of symptomatic and asymptomatic COVID-19-positive patients.The prevention of the asymptomatic transmission of SARS-CoV-2 is still one of the biggest challenges [36], and the implementation of protocols to reduce the salivary load of SARS-CoV-2 before AGPs could play a significant role in decreasing the risk of cross-infection in healthcare settings.

Strengths and Limitations
This systematic review presents several strengths, including an unrestricted search in the literature, the search protocol, data retrieval, and risk assessment analysis performed in duplicate.
However, COVID-19 is a disease that is continuously being investigated, and multiple RCTs are evaluating the use of different mouthwashes in progress at this moment.In addition, this systematic review only included four RCTs; therefore, our results must be interpreted with caution, and further investigations must be carried out soon.

Recommendations for Further Research
This study systematically reviewed the first RCTs investigating PVP-I as a pre-procedural rinse.Further in vitro studies evaluating potential new mouthwash solutions and additional RCTs are needed to demonstrate the safety and efficacy of different mouthwashes.

Conclusions
Within the limitations of this study, PVP-I presents a significant virucidal activity against SARS-CoV-2 in saliva with concentrations ranging from 0.5% to 1%.On the other hand, concentrations of 2% did not show statistically significant changes in salivary load in one of the included studies.In clinical practice, a 30-or 60-s pre-procedural rinse of 0.50/1% PVP-I could be beneficial to reduce the risk of cross-infection in healthcare settings performing AGPs in diagnosed, suspected, or asymptomatic COVID-19-positive patients.However, these results should be taken with caution, as this review included a low number of studies, and additional RCTs are essential to confirm the validity of these findings.

Figure 1 .
Figure 1.PRISMA ® flow diagram of the search processes and results.

1 .
Was true randomization used for the as-

1 .
Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way for treatment groups?11.Were outcomes measured in a reliable way?Int.J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW 6 zation used for the asipants to treatment treatment groups conoups similar at the baseblind to treatment asring treatment blind to ent? essors blind to treatment oups treated identically rvention of interest?plete and if not, were en groups in terms of equately described and Int.J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW 6 of 10

1 .
Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and Int.J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW 6

1 .
Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4. Were participants blind to treatment assignment? 5. Were those delivering treatment blind to treatment assignment? 6. Were outcome assessors blind to treatment assignment? 7. Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and Int.J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW ization used for the asipants to treatment treatment groups conoups similar at the baseblind to treatment asring treatment blind to ent? essors blind to treatment oups treated identically rvention of interest?plete and if not, were en groups in terms of equately described and analyzed in the groups e randomized?easured in the same way ps? easured in a reliable tatistical analysis used? n appropriate and any e standard RCT design he conduct and analysis = yes, = no, = uncertain.

Table 2 .
JBI Critical Appraisal Tool[19] for RCTs.Reprinted with permission from JBI.Copyright 2020.ization used for the asipants to treatment treatment groups conoups similar at the baseblind to treatment asring treatment blind to ent? essors blind to treatment oups treated identically rvention of interest?plete and if not, were en groups in terms of equately described and analyzed in the groups e randomized?easured in the same way ps? easured in a reliable tatistical analysis used? n appropriate and any e standard RCT design he conduct and analysis = yes, = no, = uncertain.

1 .
Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4.
ization used for the asipants to treatment treatment groups conoups similar at the baseblind to treatment asring treatment blind to ent? essors blind to treatment oups treated identically rvention of interest?plete and if not, were en groups in terms of equately described and analyzed in the groups e randomized?easured in the same way ps? easured in a reliable tatistical analysis used? n appropriate and any e standard RCT design he conduct and analysis = yes, = no, = uncertain.

1 .
Was true randomization used for the assignment of participants to treatment groups? 2. Was allocation to treatment groups concealed 3. Were treatment groups similar at the baseline?4.
Int. J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW domization used for the asarticipants to treatment n to treatment groups connt groups similar at the baseants blind to treatment aselivering treatment blind to ignment? e assessors blind to treatment nt groups treated identically e intervention of interest?p complete and if not, were etween groups in terms of p adequately described and ants analyzed in the groups were randomized?es measured in the same way groups?es measured in a reliable iate statistical analysis used? design appropriate and any om the standard RCT design r in the conduct and analysis = yes, = no, = uncertain.

Table 1 .
Results of the included RCTs.

Table 2 .
JBI Critical Appraisal Tool

Table 2 .
JBI Critical Appraisal Tool

Appraisal Questions Chaudhary et al. [23] (2021) Seneviratne et al. [21] (2020) Elzein et al. [22] (2021) Ferrer e [20] (20
Were treatment groups treated identically other than the intervention of interest?8. Was follow up complete and if not, were differences between groups in terms of their follow-up adequately described and analyzed?9. Were participants analyzed in the groups to which they were randomized?10.Were outcomes measured in the same way Int. J. Environ.Res.Public Health 2022, 19, x FOR PEER REVIEW