Interventions for managing halitosis

Summary of findings for the main comparison. Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis

Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mechanical tongue cleaning Comparison: no tongue cleaning
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with no tongue cleaning	Risk with mechanical tongue cleaning
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 1.804 units	MD 0.20 units lower (0.34 lower to 0.07 lower)	‐	46 (2 RCTs)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aAcar 2019; Wang 2017. ^bDowngraded for imprecision ‐ low sample size and event rate. ^cDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, selection bias, detection bias, and reporting bias. High risk of performance bias.

Summary of findings 2. 0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis

0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis
Patient or population: patients reporting halitosis Setting: University hospital Intervention: 0.6% eucalyptus chewing gum Comparison: placebo chewing gum
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo chewing gum	Risk with 0.6% eucalyptus chewing gum
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported organoleptic score was 1.60 units	MD 0.10 units lower (0.31 lower to 0.11 higher)	‐	65 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aTanaka 2010. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment. ^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

Summary of findings 3. 1000 mg champignon compared to placebo for managing halitosis

1000 mg champignon compared to placebo for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 1000 mg champignon Comparison: placebo
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo	Risk with 1000 mg champignon
Dentist‐reported OLT score assessed with dentist's perception	‐	‐	‐	‐	‐	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 100 Follow‐up: mean 2 weeks	The mean patient‐reported VAS was 63.47 units	MD 1.07 units lower (14.51 lower to 12.37 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNishihira 2017. ^bDowngraded for risk of bias ‐ unclear risk of performance and detection bias and high risk of bias in reporting bias. ^cDowngraded for imprecision ‐ wide confidence interval crossing the line of no effect, low sample size and event rate.

Summary of findings 4. Hinokitiol gel compared to placebo gel for managing halitosis

Hinokitiol gel compared to placebo gel for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: hinokitiol gel Comparison: placebo gel
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo gel	Risk with hinokitiol gel
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 2.10 units	MD 0.27 units lower (1.26 lower to 0.72 higher)	‐	18 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aIha 2013. ^bDowngraded for risk of bias ‐ high risk of performance and detection bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Summary of findings 5. 0.3% triclosan toothpaste compared to control toothpaste for managing halitosis

0.3% triclosan toothpaste compared to control toothpaste for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 0.3% triclosan toothpaste Comparison: control toothpaste
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with control toothpaste	Risk with 0.3% triclosan toothpaste
Dentist‐reported breath odour score assessed with dentist's perception Scale from: 1 to 9 Follow‐up: mean 1 week	The mean dentist‐reported breath odour score was 7.14 units	MD 3.48 units lower (3.77 lower to 3.19 lower)	‐	81 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aHu 2005. ^bDowngraded for risk of bias ‐ unclear risk of bias due to improper selection, lack of allocation concealment, performance, detection and reporting. ^cDowngraded for imprecision ‐ low sample size and event rate.

Summary of findings 6. Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis

Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing chlorhexidine and zinc acetate Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing chlorhexidine and zinc acetate
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 months	The mean dentist‐reported OLT score was 2.30 units	MD 0.20 units lower (0.58 lower to 0.18 higher)	‐	44 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aAdemovski 2017. ^bDowngraded for risk of bias ‐ unclear risk of selection bias, detection bias and other bias. ^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

See Characteristics of included studies; Characteristics of excluded studies; Characteristics of studies awaiting classification and Characteristics of ongoing studies.

Summary of findings 7. Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis

Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: brushing + cetylpyridium mouthwash Comparison: brushing
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with brushing	Risk with brushing + cetylpyridium mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT score was 1.37 units	MD 0.48 units lower (0.72 lower to 0.24 lower)	‐	70 (1 RCT)^a	⊕⊕⊝⊝ LOW^b	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aFeres 2015. ^bDowngraded for imprecision ‐ low sample size and event rate.

Background

The term halitosis is a general term used to describe any disagreeable odour of expired air, regardless of its origin. It is derived from the Latin word halitus meaning 'breath' or halare 'to breath' with a suffix from the Greek based noun osis (Harper 2016) which means pathologic alteration (Wu 2019). The lay term, bad breath, is the generally accepted term for foul smells emanating from the mouth but the term oral malodour is reserved for halitosis originating from the oral cavity (Tangerman 2002). Mouthwashes and tongue scrapers are popular ways of dealing with oral malodour.

Description of the condition

Prevalence and aetiology

The reliability of relevant epidemiological data has been questioned, but the prevalence of halitosis has been reported to be as high as 50% to 65% of the world's population (Mookem 2014; Yaegaki 2000). Severe halitosis may involve less than 5% of the population (Rosing 2011). In a study in Japan, 24% of patients complained of oral malodour (Miyazaki 1995) while in France it was reported that between 50% and 60% of the population suffer from chronic halitosis (Meningaud 1999). In Belgium, a study evaluated the characteristics of 2000 patients who visited a halitosis clinic, and reported that 76% of the patients had a possible oral cause e.g. tongue coating 43%, gingivitis/periodontitis 11%, or a combination of the two 18% (Quirynen 2009). A review of the literature reported a wide variation in the prevalence of halitosis around the world, with a rate ranging from 22% to 50% of the population (Akaji 2014). A systematic review and meta‐regression analysis done by Silva 2018 reported the prevalence of halitosis to be 31.8% (95% confidence interval (CI) 24.6% to 39.0%).

Multiple factors contribute to the aetiology of halitosis, and these may be the combination of drugs, food, local, systemic and psychological causes (Singh 2015; Thoppay 2018). It is now fairly widely accepted that halitosis originates from the oral cavity (Ayers 1998; Delanghe 1997; De Geest 2016). Accumulation of bacteria and food residues at the posterior part and in the furrows of the tongue (Seeman 2014; van Steenberghe 1997) is considered to be the major cause (Scully 1997; Thoppay 2018). Interdental plaque and gingivitis may also play a contributory role, and although periodontal pockets may produce putrid odours, their contribution to oral malodour is still unclear (De Geest 2016; Morita 2001).

3% to 10% of halitosis cases are caused by ear, nose and throat related problems like tonsillitis, sinusitis and postnasal drip which are commonly known as extraoral or non‐oral halitosis or throat halitosis (Bollen 2012). Interventions for such halitosis are not covered under the scope of this review.

Halitosis‐causing bacteria are the primary sources of volatile sulphur compounds (VSC); the chief components of which are hydrogen sulphide and methyl mercaptans (Kleinberg 1990; Tonzetich 1977). VSC and other additional odours such as indole, skatole, putrescine and cadaverine (Kleinberg 1995) are produced through the bacterial metabolic degradation of food debris, desquamated cells, saliva proteins, dental plaque and microbial putrefaction (Ratcliff 1999). The periodontal pocket also provides an ideal environment for VSC production thus explaining why patients with periodontal disease often complain of oral malodour (Morita 2001). The intensity of clinical bad breath has been shown to be significantly associated with the intraoral VSC level and to be correlated directly with periodontal health status (Bosy 1994; Replogle 1996; Stamou 2005).

Classification of halitosis

Halitosis has been defined as an unpleasant odour exhaled through the mouth and upper airways, caused by biofilm accumulation on the dorsum of the tongue, the interdental spaces or due to periodontal disease, although the condition is multifactorial and may involve both oral and non‐oral conditions (Oliveira‐Neto 2013; van den Broek 2007).

Although this classification has not been universally accepted by all experts in the field there is general agreement that halitosis can be categorised as genuine halitosis, pseudo‐halitosis and halitophobia (Yaegaki 2000). Genuine halitosis has been further subclassified as physiological halitosis in which there is no readily apparent disease or pathological condition, or pathological halitosis which occurs as a result of an infective process of the oral tissues. Pseudo‐halitosis is a condition in which there is absence of halitosis but patients believe that they have oral malodour. Halitophobia can occur when there is no physical or social evidence to suggest that halitosis is present and which can persist after treatment for either genuine halitosis or as pseudo‐halitosis.

Organoleptic test (OLT) measurement by trained breath judges is considered to be the gold standard and the most reliable way of evaluating malodour (Rosenberg 1992; Rosenberg 1995), but this has been contested by studies showing that measurements with the halimeter appear to be more reproducible albeit possibly less reliable than OLT methods (Silwood 2001). Methods of assessment of levels of malodour include those which are very simple, highly subjective and others which are complex, time consuming and involve the use of sophisticated equipment:

OLT score (Rosenberg 1992): 0: no detectable odour; 1: hardly detectable odour; 2: light odour; 3: moderate odour; 4: strong odour; and 5: extremely strong odour
portable VSC monitor, the halimeter (Rosenberg 1991): normal: 80 to 160 parts per billion (ppb); weak: 160 to 250 ppb; and strong: > 250 ppb (Baharvand 2008)
gas chromatography coupled with flame‐photometric detection (Solis‐Gaffar 1975)
culture of plaque and periodontal pocket exudates (Loesche 1995) and
multisensor approach, BIONOTE (Marchetti 2015).

Measurement of VSC levels can be carried out by a variety of methods: OLT which are considered subjective by some investigators but are the ones most relevant to patients (Tsunoda 1981), and the more complex gas chromatography techniques (Solis‐Gaffar 1975). Portable computerized VSC monitors or halimeters are available, they are compact, easy to use and relatively inexpensive (Pedrazzi 2004) but have their limitations in that they have a high sensitivity for hydrogen sulphide, but low sensitivity for one of the other sources of malodour, methyl mercaptan (Rosenberg 1991). Silwood 2001 have shown good reproducibility of VSCs in their study.

A correlation rate has been reported between the self‐estimation of bad breath and the presence of oral malodour as determined by OLT examination by odour‐judge assessment in patients with slight or moderate oral halitosis (Romano 2010).

Description of the intervention

At present there are no standard and accepted protocols for the treatment of oral malodour (Morita 2001) which could be because of its multiple aetiology. Halitosis, by itself, is not a disease, but a sign/symptom of a disease. Patients who are conscious that they have halitosis may attempt to mask it through compulsive brushing or with a range of over‐the‐counter methods such as chewing gum, mints, scented liquid drops, and the use of mouthrinses (Borden 2002). Most of these merely provide a competing and temporary smell that is capable of masking the unfavourable malodour. Some mouthrinses contain certain components that can neutralise the malodour or the bacteria which produce it. The most common of these include alcohol, zinc, phenol, chlorhexidine and folic acid. Reduction of the causative bacteria can also be accomplished through improving oral hygiene (Tonzetich 1978) in addition to cleaning of the tongue (Rosenberg 1996). This can be achieved by brushing or scraping the dorsum of the tongue to dislodge trapped food, cells, and bacteria from between the filiform papillae. Methods for treating or masking halitosis include:

mechanical methods: tongue cleaners which are more commonly made of plastic, resin, rubber or metal. These may contain nylon bristles and grooves or corrugations but they must be smooth. Toothbrushes can be used but these normally have soft bristles (or extra soft bristles) only (Pedrazzi 2004)
chemical methods: these include a range of mouthwashes containing antimicrobials such as chlorhexidine (0.2% to 0.12%), cetylpyridinium chloride (0.05% to 0.07%), hydrogen peroxide and essential oils to combat proteolytic odoriferous bacteria, producing VSCs, and those that mask odours, without interfering with microbial viability (zinc chloride or lactate, chlorine dioxide – 0.3%). Combinations of antimicrobial agents in one mouthwash, such as zinc salts and essential oils, or zinc salts and chlorhexidine or cetylpyridinium chloride are also available (van den Broek 2008).

Some combinations of mechanical and chemical methods have also been explored, with the combination of brushes and toothpastes containing zinc salts or even toothbrushes and chlorhexidine or other antimicrobial agents (Slot 2015; Slots 2012).

How the intervention might work

A range of mechanical and chemical hygiene (mouthrinses or mouthwashes) methods have been advocated (Oliveira‐Neto 2013), however the effectiveness of any intervention may be influenced by the nature of the mouthrinse formulation (Fedorowicz 2008), or by the type of mechanical device (dental floss, toothbrush, toothpaste) used to reduce VSCs (Oliveira‐Neto 2013).

The intervention needs to be able to reduce, eliminate or mask the production of VSCs, i.e. actions aimed at minimising the food available for odoriferous bacteria, reduce the total number exists of these bacteria, or make any environment where VSC‐producing bacteria live, less hospitable. The success of any halitosis intervention appears to hinge on the reduction of VSC levels and other foul volatiles and consequently the majority focus on mechanical and chemical options.
Mechanical interventions (i.e. brushing, flossing and tongue scraping) aim to reduce the numbers of VSC‐producing bacteria, residual food matter and cellular debris from the gingivae and tongue. In an earlier version of a systematic review of the effectiveness of tongue scraping for treating halitosis, the review authors found that mechanical tongue cleaning with tongue scrapers appeared to have very limited and short acting benefits in controlling halitosis (Outhouse 2006).
The limitations of mechanical methods to effectively reach and remove VSC‐producing bacteria from all oral ecological sites are acknowledged. The possibility that mouthrinses may be more effective in reaching the less accessible parts of the oral cavity, their greater social acceptance and ease of use has led to the development of a large number and range of over‐the‐counter mouthrinses (Ayers 1998; Richter 1996).

A number of mouthrinses contain antibacterial agents in addition to flavouring agents and these have been generally categorised into those that neutralise and those that mask the odour. Components which neutralise can further be divided into those that affect the bacteria directly or the chemical compounds they produce, and include chlorhexidine, phenol, triclosan, chlorine dioxide, alcohol and metal ions, the most common of which is zinc (Carvalho 2004; Farrell 2006). Some of the odour‐masking agents, consist of essential oils, which can also provide a competing and purely temporary smell that is capable of disguising the unfavourable malodour.

Tongue cleaning has been claimed to reduce oral malodour by decreasing VSC concentration by 20% to 70% (Tonzetich 1977). Oliveira‐Neto 2013 compared both mechanical and mouthrinses for treatment of morning breath and concluded that chlorhexidine and mechanical oral hygiene reduced bad breath for longer periods than tongue cleaning alone.

Continuous usage of mouthrinses can lead to adverse effects such as oral mucosa and dental‐crown staining, mucosal lesions, taste modifications, or abnormal oral sensation (Tartaglia 2019).

Why it is important to do this review

Halitosis can be serious enough to cause personal embarrassment, reduce self‐esteem and adversely affect personal relationships. It may also be a barrier to certain types of employment. There is existing uncertainty as to which is the most effective method of oral malodour control. The most popular method used involves mouthwashes containing chemicals which destroy odour‐forming bacteria and include other odour‐masking constituent which can disguise the smell. The simplicity in use and social acceptance of mouthrinses appear to support their popularity over mechanical means.

This new Cochrane Review incorporates the previous Cochrane Reviews on tongue scraping (Outhouse 2006) and mouthrinses (Fedorowicz 2008) for halitosis and aims to assess the effects of interventions used to control halitosis due to oral diseases.

Objectives

Methods

Criteria for considering studies for this review

Types of studies

We included randomised controlled trials only.

Types of participants

We included studies that recruited adolescents and adult participants over the age of 16 who presented with a clinical or self‐assessed diagnosis of halitosis, with no significant comorbidity or health condition that might lead to increased halitosis (e.g. diabetes). We excluded studies which had been conducted on participants with induced halitosis either by stopping or altering the oral hygiene habits, physiological halitosis such as morning breath, non‐oral halitosis, refractory and severe chronic periodontal diseases. Subjects with clinical attachment level (CAL) ≥ 5 mm were considered to have severe chronic periodontitis (Kinane 2015).

Types of interventions

Any intervention for the management of halitosis compared to another or placebo, or no intervention. The active interventions or controls were administered over a minimum of one week and with no upper time limit.

Studies which included single use mouthwashes were not considered for this review as the aim was to evaluate therapeutic effect rather than masking effect (Dadamio 2013).

Types of outcome measures

We did not consider these prespecified outcomes as criteria for including studies in this review, but they are a representative list of the outcomes of interest within whichever studies were included. See Section 5.1.2 in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011).

Primary outcomes

For the primary outcomes in this review we considered self‐expressed (perceived) (Greenman 2004) and organoleptic test (OLT) (human nose) assessments of halitosis using any validated malodour intensity scale.

Secondary outcomes

Quality of life.
Assessment of halitosis as measured by any of the validated methods (halimeter, portable sulphide monitor or gas chromatography coupled with flame‐photometric detection).
Peak and steady‐state volatile sulphur compound levels using a sulphide monitor, prior to and at several time points after any intervention.
Adverse events.

Search methods for identification of studies

Electronic searches

Cochrane Oral Health's Information Specialist conducted systematic searches in the following databases for randomised controlled trials and controlled clinical trials without language or publication status restrictions:

Cochrane Oral Health's Trials Register (searched 8 April 2019) (Appendix 1);
Cochrane Central Register of Controlled Trials (CENTRAL; 2019, Issue 3) in the Cochrane Library (searched 8 April 2019) (Appendix 2);
MEDLINE Ovid (1946 to 8 April 2019) (Appendix 3);
Embase Ovid (1 November 2016 to 8 April 2019) (Appendix 4).

Subject strategies were modelled on the search strategy designed for MEDLINE Ovid. Where appropriate, they were combined with subject strategy adaptations of the highly sensitive search strategy designed by Cochrane for identifying randomised controlled trials and controlled clinical trials as described in the Cochrane Handbook for Systematic Reviews of Interventions Chapter 6 (Lefebvre 2011).

Due to the Cochrane Centralised Search Project to identify all clinical trials in the database and add them to CENTRAL, only the most recent months of the Embase database were searched. See the searching page on the Cochrane Oral Health website for more information. No other restrictions were placed on the date of publication when searching the electronic databases.

We also conducted additional searches in the following databases:

LILACS BIREME Virtual Health Library (Latin American and Caribbean Health Science Information database; from 1982 to 19 April 2019) (Appendix 5);
CINAHL EBSCO (Cumulative Index to Nursing and Allied Health Literature; 1937 to 19 April 2019) (Appendix 6);
the National Database of Indian Medical Journals (IndMed, indmed.nic.in/) (1985 to 19 April 2019) (Appendix 7);
OpenGrey (1992 to 19 April 2019) (Appendix 8).

Searching other resources

Cochrane Oral Health's Information Specialist searched the following trials registers/databases for ongoing trials on 8 April 2019:

the US National Institutes of Health Ongoing Trials Register ClinicalTrials.gov (Appendix 9);
the World Health Organization International Clinical Trials Registry Platform (www.who.int/trialsearch) (Appendix 10).

We also conducted additional searches in the following trials registries on 19 April 2019:

ISRCTN registry (www.isrctn.com) (Appendix 11);
Clinical Trials Registry ‐ India (ctri.nic.in/Clinicaltrials/login.php)(Appendix 12).

Prashanti Eachempati (PE) examined the bibliographies of the included and excluded studies and systematic reviews published in the year 2019 and 2018 for further references to potentially eligible randomised controlled trials based on the assumption that these reviews could have included previously published trials.

Sumanth Kumbargere Nagraj (SKN), Vijendra Pal Singh (VPS) and Eswara Uma (EU) contacted trial investigators and asked them to provide missing data or clarify study details.

We did not conduct a separate search for adverse effects of interventions for halitosis. However, we examined data on adverse effects from the included studies that were identified.

We checked that none of the included studies in this review were retracted due to error or fraud.

Data collection and analysis

Selection of studies

Two pairs of review authors (PE and VPS; EU and Eby Varghese (EV); Noorliza Mastura Ismail (NMI) and SKN) independently assessed the abstracts of studies resulting from the searches. The search was designed to be sensitive and include controlled clinical trials, these were filtered out early in the selection process if they were not randomised. We obtained full copies of all relevant and potentially relevant studies, those appearing to have met the inclusion criteria, or for which there was insufficient information in the title and abstract to make a clear decision on eligibility. We assessed the full‐text papers independently and resolved any disagreement on the eligibility of included studies through discussion and consensus. We excluded those records that did not meet the inclusion criteria, and we noted the reasons for their exclusion in the 'Characteristics of excluded studies' section of the review.

Data extraction and management

Two pairs of review authors (PE and VPS; EU and SKN; NMI and EV) independently collected study details and outcome data using a predetermined form designed for this purpose. We entered study details into the 'Characteristics of included studies' table in Review Manager (RevMan) (Review Manager 2014). The authors included data if there was an independently reached consensus.

We extracted the following details from the eligible trials.

Trial methods: method of sequence generation and concealment of allocation sequence; masking of participants, trialists and outcome assessors; exclusion of participants after randomisation; proportion of and reasons for losses to follow‐up.
Participants: country and study setting; sample size; age; ethnicity; inclusion and exclusion criteria.
Intervention: type; concentration, dose, and frequency; route of administration; duration of intervention and follow‐up.
Control: type; duration of intervention and follow‐up.
Outcomes: primary and secondary outcomes as specified in the 'Types of outcome measures' section.

If available, we collected data on sources of funding of the included studies, country, set‐up and number of centres.

Assessment of risk of bias in included studies

Two review authors (SKN and PE) assessed the risk of bias of the selected studies independently using Cochrane's tool for assessing risk of bias as described in Chapter 8, Section 8.5, in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We compared the evaluations and discussed and resolved any inconsistencies between the review authors.

We assessed the following domains as at 'low', 'unclear', or 'high' risk of bias:

sequence generation;
allocation concealment;
blinding of participants and personnel;
blinding of outcomes assessment;
incomplete outcome data;
selective outcome reporting; and
other bias.

We reported these assessments for each individual study in the 'Risk of bias' tables.

We categorised and reported the overall risk of bias of each of the included studies according to the following:

low risk of bias (plausible bias unlikely to seriously alter the results) if all criteria were met;
unclear risk of bias (plausible bias that raises some doubt about the results) if one or more criteria were assessed as unclear; or
high risk of bias (plausible bias that seriously weakens confidence in the results) if one or more criteria were not met.

Measures of treatment effect

We presented continuous outcomes on the original scale as reported in each individual study. If similar outcomes were reported using different scales, we intended to convert these to standardised mean differences (SMD). However, we did not find any studies using different scales to use SMD. We presented measures of treatment effect as mean differences (MD) with their 95% confidence intervals (CIs).

We intended to present the dichotomous outcomes as risk ratios (RR) and 95% CIs, if found significant, we intended to convert them to either: the number of patients needed to treat to find one additional beneficial outcome (NNTB); or the number needed to treat to find one additional harmful outcome (NNTH). However, none of the review outcomes were reported as dichotomous outcomes.

Unit of analysis issues

Cross‐over trials

Unit of analysis issues can arise in studies where participants have been randomised to multiple treatments in multiple periods or where there has been an inadequate wash‐out period. We analysed these data based on the advice provided in Section 16.4.4 in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We intended to assess the carry‐over and period effects descriptively, and if there was evidence of minimal impact and there were adequate data, we planned to carry out a paired analysis. However, we did not carry out paired analysis.

Studies with multiple treatment groups

Studies that are reported with multiple treatment groups have the potential for participant data to contribute to multiple comparisons. We planned to assess the treatments and determine which were relevant to our review then allocate the non‐intervention participants as the 'shared' group. We intended to split the 'shared' group equally into the number of comparisons made, as discussed in Section 16.5.4 in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). However, we did not encounter such studies in this review.

Dealing with missing data

If we encountered data missing from trials that are less than 10 years old, we would have tried wherever possible to contact the investigators or sponsors of these studies. We planned to re‐analyse data according to the intention‐to‐treat (ITT) principle whenever possible. However, we did not encounter such studies in the review.

Assessment of heterogeneity

We assessed clinical heterogeneity by examining the characteristics of the studies and the similarity between the types of participants and the interventions. We assessed the degree of heterogeneity between the studies using the I² statistic. We reported heterogeneity as important and at least moderate to substantial if the I² statistic > 60% (Higgins 2011). If this was explained by clinical reasoning and a coherent argument could be made for combining the studies, we entered these into a meta‐analysis. In cases where the heterogeneity could not be adequately explained, we intended to pool the data but would account for any heterogeneity and downgrade the certainty of the body of evidence according to GRADE methods. However, we did not find such cases in the review.

Assessment of reporting biases

We planned to follow reporting bias assessment as recommended by Egger 1997, through testing for funnel plot asymmetry as described in Section 10.4.3.1 in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We intended to perform these for primary and secondary outcomes for meta‐analysis if we included a minimum number of studies, to allow a reasonable estimate of the effect of intervention (nominally nine studies). However, none of our analyses included nine or more studies and hence we did not assess reporting bias as planned.

Data synthesis

Two review authors (SKN and PE) analysed the data in RevMan (Review Manager 2014) and reported them in accordance with the advice in Chapter 9 in the Cochrane Handbook for Systematic Reviews of Interventions (Higgins 2011). We carried out a random‐effects meta‐analysis and planned to report data that exhibited not more than moderate heterogeneity (Treadwell 2006). However, the majority of the analyses included not more than two studies and none of the analyses showed heterogeneity.

Subgroup analysis and investigation of heterogeneity

We planned to conduct the following subgroup analyses subject to availability of a reasonable number of studies (n ≥ 3) reporting data:

OLT level of halitosis ≥ 3 at baseline;
evaluation method: OLT or halimeter;
duration of treatment and the time of assessments.

However, because of a less number of studies in the analyses, we did not conduct any subgroup analysis.

Sensitivity analysis

We planned to carry out sensitivity analyses to assess the robustness of the results of this review. This intended to include repeating the analyses with the following adjustment: exclusion of studies at high risk of bias and reporting of any comparative difference between the results of these analyses. However, we did not have multiple similar studies included to carry out sensitivity analysis.

Presentation of main results

We produced 'Summary of findings' tables using GRADEpro GDT 2015 for the most important comparisons and the following outcomes:

dentist‐reported OLT change from baseline in halitosis;
patient‐reported OLT change from baseline in halitosis; and
adverse events.

We assessed the level of certainty in the findings with reference to the risk of bias assessments, the directness of the evidence, the inconsistency of the results, the precision of the estimates, and the risk of publication bias. The level of certainty for each of the comparisons was categorised as high, moderate, low, or very low.

Results

Description of studies

Results of the search

We included 44 trials (55 reports) in the review. (If the same study (one population) was separated into multiple reports we included the primary study and considered the rest as reports as per Higgins 2011.) See Figure 1 for the selection process of search results.

Figure 1

Study flow diagram.

Included studies

Characteristics of trial settings and investigators

Publication status

Out of 44 reports, 43 were published and one was an unpublished report (NCT02628938).

Language

We had five studies in foreign languages. Two in Mandarin (An 2011; Wang 2017), one in Portuguese (Garcia 2014), one in Spanish (López Jornet 2003), and one in Arabic (Talebian 2009). The other 39 trials were in the English language.

Sample size

The minimum sample size was seven (Talebian 2009) and the maximum sample size was 190 (Navada 2008).

Characteristics of participants

Age

The minimum age of the participants in the included trials was 17 years (Asokan 2011; Patil 2017; Rassameemasmaung 2007) and the maximum age was 77 years (Ademovski 2017).

Gender

32 trials included both the genders (Acar 2019; Ademovski 2012; Ademovski 2017; An 2011; Barak 2012; Borden 2002; Dadamio 2013; Feres 2015; Hu 2005; Hu 2018; Iha 2013; Kara 2008; Kozlovsky 1996; Lee 2018; Lomax 2017; López Jornet 2003; Marchetti 2015; Nakano 2017; Nishihira 2017; Nogueira‐Filho 2002; Patil 2017; Payne 2011; Rassameemasmaung 2007; Rassameemasmaung 2012; Satthanakul 2014; Suzuki 2014; Tanaka 2010; Wang 2017; Watanabe 2018; Wigger‐Alberti 2010; Wilhelm 2012; Winkel 2003). Four trials included only males (Aung 2015; Iwamura 2016; Nohno 2012; Talebian 2009), one trial included only females (NCT02628938), and the other seven trials did not mention the gender details of the participants (Asokan 2011; Caygur 2017; Garcia 2014; Mamgain 2016; Navada 2008; Niles 1999; Wirthlin 2011).

Characteristics of interventions

1. Mechanical debridement

We included four studies and the following comparisons were identified.

1a. Scaling and root planing (SRP) with air polishing versus SRP: we included one study (Caygur 2017) in this comparison. This study had two arms, comparing SRP plus glycerine powder air polishing with SRP alone in patients with halitosis with follow‐ups after 7, 14 and 30 days. The outcome measure used was volatile sulphur compound (VSC) measured using a halimeter. In this review, we have used data after 30 days follow‐up only.

1b. SRP + laser versus SRP: we included one study (Kara 2008) for this comparison. The study had three arms, SRP (group I), subgingival laser irradiation combined with povidone‐iodine application (group II), and SRP and subgingival laser irradiation (group III). They followed‐up for one week and four weeks after the intervention. The outcome measures they used were organoleptic test (OLT) score and VSC. However, we have used the data from group I and II only for the one week follow‐up period for both the outcome measures.

1c. Mechanical tongue cleaning versus no tongue cleaning: we included two studies (Acar 2019; Wang 2017) in this comparison. Acar 2019 did a two parallel‐arm study. After scaling and polishing, tongue cleaning by using a tongue scraper was compared to no tongue cleaning in 36 patients. The outcome measures were dentist‐reported OLT scores and VSC with a follow‐up of seven days. Wang 2017, a two parallel arm study, compared toothbrushing and mechanical tongue cleaning with toothbrushing and have followed‐up for a period of one, two, four and eight weeks. The outcomes measured were OLT scores and VSC. However, we have used the data for VSC scores after one week follow‐up only (standard deviation (SD) could not be calculated for OLT).

2. Chewing gum/lozenges

We included two studies using chewing gum, one study using candy and one study using lozenges under this category.

2a. 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum and placebo: one study (Tanaka 2010) is included in this comparison which was a three parallel‐arm study comparing high and low concentration eucalyptus chewing gum and a placebo chewing gum, five minutes, five times per day for a period of 12 weeks. The outcome measures used in the study were OLT and VSC scores evaluated by the dentist and were assessed at the end of 4, 8, 12 and 14 weeks. We have used the data for both scores after four weeks follow‐up only and have analysed the outcomes between 0.4% and 0.6% eucalyptus chewing gums and 0.6% eucalyptus chewing gum and placebo groups.

2b. Pycnogenol chewing gum versus placebo chewing gum: one study (Watanabe 2018) compared these two chewing gums, 2.5 mg for 15 minutes, six times daily for a period of four weeks. The outcome measures used were VSC scores of three volatile gases. We have used the data at the end of two weeks.

2c. Abrasive candy; abrasive candy with propolis and abrasive candy with zinc gluconate versus abrasive candy with propolis and zinc:Barak 2012 in their 5‐arms parallel‐group randomised controlled trial (RCT), compared the reduction of halitosis in the subjects using abrasive candy (Breezy candy); abrasive candy with 2% propolis and abrasive candy with 0.5% zinc gluconate versus abrasive candy with 1% propolis and 0.25% zinc. The outcome measured was VSC score using a halimeter. We could not use the results of this study in the meta‐analysis because of the missing SD and P value in the report.

2d. Lactobacillus brevis CD2 lozenges versus placebo lozenges: in a two‐arm parallel‐group RCT conducted by Marchetti 2015, reduction in halitosis was compared between groups consuming Lactobacillus brevis CD2 lozenges and placebo lozenges. The outcome measures were OLT, VSC and breath print scores measured using Rosenberg scale, OralChroma and Bionote. The study did not report any usable data and hence could not be included in the meta‐analysis.

3. Systemic deodorising agent

We have only one comparison of systemic agents under this section.

3a. Champignon extract versus placebo: we included one study (Nishihira 2017) in this comparison. This is a four parallel‐arm study which compared 50 mg/day, 500 mg/day and 1000 mg/day champignon (champignon extract, an extract boiled from the mushroom Agaricus bisporus) with placebo tablets. The follow‐up period was four weeks and the outcome measures were visual analogue scale (VAS) score (0 to 100) which was reported by the study participants and relative of the participant. We have used the data after two weeks follow‐up for 50 mg, 1000 mg, and placebo groups only.

4. Topical agents

We have two comparisons reported by two trials.

4a. Hinokitiol gel versus placebo gel: this comparison was seen in only one study (Iha 2013) which is a two‐arm parallel‐group study comparing hinokitiol gel (hinokitiol C10H12O2 (b‐thujaplicin), a component of the essential oils isolated from Cupressaceae) with placebo gel. The outcome measures were OLT scores, VSC scores for methyl mercaptan and hydrogen sulphide after a follow‐up of 28 days, as reported by the dentist.

4b. Topical G32 versus chlorhexidine gel:Patil 2017 compared topical G32 (ayurvedic preparation consisting of extracts of Mimusops elengi, Acacia catechu, Myrtus caryophyllus, Barleria prionitis) with chlorhexidine digluconate 1% gel in a single‐blind parallel‐designed trial. The study participants crushed 2 to 3 G32 tablets and massaged it on their gums twice a day for five minutes. The outcome measures were VSC and OLT scores reported by the dentist after one week follow‐up. However, we have used the data for VSC only as the OLT score data were not available.

5. Toothpaste

We have seven comparisons reported by seven trials.

5a. Triclosan + polyvinyl methyl ether/maleic acid (PVM/MA) toothpaste versus sodium fluoride toothpaste:Hu 2005 compared the effectiveness of a dentifrice containing 0.3% triclosan, 2% PVM/MA copolymer, 0.243% sodium fluoride (TCF) to a commercially available dentifrice containing 0.243% sodium fluoride (control) for the management of oral malodour in a three‐week, randomised double‐blind, longitudinal clinical trial. The outcome measure was OLT score which was done using a nine‐point hedonic scale (1: most pleasant, 5: neutral, and 9: most unpleasant).

5b. Zinc toothpaste versus placebo toothpaste: two randomised, two‐cell parallel, double‐blind, placebo‐controlled clinical trials were done by Navada 2008. Both the studies compared the efficacy of toothpaste containing 0.2% zinc sulphate to toothpaste without zinc. In the first study, VSC was measured by halimeter and in second, breath freshness was assessed by four odour judges using OLT scores (0: no odour present and 5: extremely foul odour).

5c. Sodium bicarbonate toothpaste versus control toothpaste: a single‐centre, single examiner‐blind, randomised, controlled, two‐treatment, parallel‐group study, with a six‐week intervention period was conducted by Lomax 2017. Toothpaste containing sodium bicarbonate was compared to control toothpaste which did not have sodium bicarbonate. The outcome measure for halitosis was VSC using gas chromatography with flame photometric detection.

5d. Dual zinc + arginine dentifrice versus control toothpaste: in a double‐blind, clinical study done by Hu 2018 a dual zinc plus arginine dentifrice containing zinc oxide and zinc citrate 0.96%, 1.5% arginine and 1450 parts per million (ppm) fluoride (F) as sodium fluoride in a silica base was compared to a regular fluoride dentifrice containing 1450 ppm F as sodium fluoride in a silica base to control halitosis. The outcome measure for halitosis was OLT hedonic scale (1: most pleasant to 9: most unpleasant).

5e. Zinc chloride (ZnCl) + sodium fluoride (NaF) dentifrice versus control dentifrice containing NaF:Payne 2011, in his cross‐over randomised trial, compared the reduction of halitosis in subjects brushing using dentifrice containing ZnCl + NaF with control dentifrice containing NaF. The outcome measure was VSC score using gas chromatography with flame photometric detection. The trial reported the adjusted mean VSC scores in the graph and hence we could not use the data in the meta‐analysis.

5f. Triclosan + PVM/MA copolymer + NaF in a silica base toothpaste versus placebo toothpaste: In a cross‐over trial done by Niles 1999, reduction in halitosis was compared in subjects using 0.3% triclosan + 2.0% PVM/MA copolymer + 0.243% NaF in a silica base toothpaste with a placebo toothpaste. The outcome measure was VSC score using a 565 Tracor gas chromatograph with a flame photometric detector. We could not use the results of this study in the meta‐analysis as there was no correlation coefficient reported and we could not find similar intervention trial to impute the SD of differences.

5g. Crest Complete A dentifrice; Signal Global A dentifrice; Colgate Total A dentifrice and experimental formulation versus negative control: In a five‐arm parallel‐group trial by Nogueira‐Filho 2002, three commercial dentifrices with 0.3% triclosan (Crest Complete A, Signal Global A and Colgate Total A) were compared with similar experimental formulation (0.3% triclosanπ2% PVM/MA 0.75% Zn 4% tetrapotassium pyrophosphate (PPi)) and a negative control dentifrice for reduction of halitosis. The outcome measure was VSC score measured using a halimeter. We could not use the results of this study in the meta‐analysis as there was no correlation coefficient reported and we could not find similar intervention trial to impute the SD of differences.

6. Mouthrinse or mouthwash

Mouthwashes are antiseptic solutions used after brushing. Whereas, a mouthrinse is used before brushing to freshen the breath (Sumanth 2019). However, we are not sure if the study authors have used it synonymously or followed the above described definition. Hence we have used the same terminology as used by the trial authors. We have 17 comparisons reported by 17 trials under this section.

6a. Halita mouthwash versus placebo:Winkel 2003 compared a newly developed mouthrinse (chlorhexidine (0.05%), cetylpyridinium chloride (0.05%) and zinc lactate (0.14%)) to placebo mouthrinse in the treatment of oral halitosis in patients without periodontitis in their dual centre, double‐blind, parallel‐arm, randomised controlled trial. The outcomes were measured after 14 days using VSC (halimeter) and OLT scores (0: no halitosis and 5: offensive halitosis).

6b. Chlorhexidine + zinc acetate mouthwash versus placebo: a randomised, double‐blind, placebo‐controlled, parallel‐group, 6‐month trial, was conducted by Ademovski 2017 comparing chlorhexidine plus zinc acetate mouthwash to placebo mouthwash in patients with halitosis. The outcome measures were OLT score (0: no odour and 5: extremely strong odour), total VSC (halimeter) and hydrogen sulphide (H₂S) and methyl mercaptan (MM) concentration using portable gas chromatograph at the duration of three and six months.

6c. Cetylperidinium chloride mouthwash versus placebo; essential oil mouthwash versus placebo and chlorine dioxide + zinc mouthwash versus placebo:Borden 2002 conducted a randomised, double‐blind, longitudinal clinical trial comparing four different mouthrinses (essential oil, chlorine dioxide + zinc, cetylpyridinium and placebo) for four weeks. The outcome measures for oral halitosis were OLT score (0: no odour and 5: extremely foul odour) and VSC scores (halimeter).

6d. Chlorine dioxide mouthwash versus placebo:Lee 2018 did a cross‐over, double‐blind randomised controlled trial comparing a mouthwash containing 0.1% stabilized chlorine dioxide or a placebo twice daily for a period of eight weeks. The outcome measure was OLT score (0: no odour and 5: extremely strong odour).

6e. Herbal mouthwash versus placebo:Rassameemasmaung 2007 compared the effect of a herbal mouthwash to placebo mouthwash in their double‐blind, randomised, placebo‐controlled trial for two weeks. The outcome measure was VSC score.

6f. Benzethonium chloride mouthwash versus placebo:Iwamura 2016 conducted a randomised, double‐blind pilot study comparing benzethonium chloride mouthwash to placebo mouthwash and no mouthwash. The outcome measures were OLT score (0: absence of odour and 5: extreme malodour) and VSC (OralChroma) for all three components separately.

6g. Green tea mouthwash versus placebo: in a double‐blind, placebo‐controlled trial done by Rassameemasmaung 2012, the effects of a green tea mouthwash were compared to placebo mouthwash for a period of four weeks. The outcome measure was VSC (halimeter).

6h. Lemongrass mouthwash versus placebo:Satthanakul 2014 did a randomised double‐blind clinical study to compare the effects of lemongrass oil mouthwash to placebo mouthwash for eight days. The outcome measure was VSC (halimeter).

6i. Halita mouthrinse versus Perio‐plus mouthrinse:Dadamio 2013 conducted a single‐centre, double‐blind, randomised, parallel‐group clinical trial comparing the efficacy of halita and meridol with and without zinc lactate versus negative and positive control. The outcome measures were OLT score (0 to 5) and VSC determined by a portable gas chromatograph.

6j. Oil water two‐phase mouthwash versus control mouthwash: In a six‐week randomised clinical trial done by Kozlovsky 1996, oil water two‐phase mouthwash containing cetylpyridinium chloride (CPC) was compared to control mouthwash. The outcome measures were OLT score (0: no appreciable odour and 5: extremely foul odour) and VSC (sulphide monitor). However, the report does not give any details of OLT score and hence we have used only VSC score in the meta‐analysis.

6k. Triphala and Ela decoction versus mouthwash:Mamgain 2016 conducted a randomised controlled trial comparing mouthwash containing decoction of Triphala and Ela with chlorhexidine mouthwash for 21 days. The outcome measure was OLT score.

6l. Miswak mouthwash versus chlorhexidine mouthwash: an unpublished clinical trial (NCT02628938) compared miswak (Salvadora persica) mouthwash with chlorhexidine mouthwash twice a day, among female students who had halitosis. The outcome measures were OLT score (0: no appreciable odour and 5: extremely foul odour), VSC (Tanita FitScan HC‐212SF Breath Checker; 0: no odour and 5: intense odour) and patient self‐assessment score (10 cm VAS that is marked as 'no odour' on the 0 cm end, and as 'extremely foul odour' on the 10 cm end) after seven days.

6m. Chlorine dioxide mouthwash versus chlorhexidine mouthwash:Wirthlin 2011, in their double‐blind, randomised, parallel‐group clinical trial, compared tongue scraping + chlorine dioxide mouthwash to tongue scraping + chlorhexidine mouthwash for one week. The outcome measures were VSC (OralChroma) and OLT score (0 to 5). However, we could not include the OLT scores in the analysis as the group‐wise data were not given.

6n. Triclosan + NaF + ZnCl + alcohol mouthwash; triclosan + NaF + ZnCl mouthwash; zinc lactate + chlorhexidine gluconate + cetylpyridine chloride mouthwash versus placebo mouthwash:López Jornet 2003 conducted a randomised, four‐arm parallel‐group clinical trial comparing triclosan mouthwash with and without alcohol, mouthwash containing zinc lactate 0.14%, chlorhexidine gluconate 0.005% and cetylpyridine chloride 0.05% with placebo mouthwash. The outcome measures were VSC scores obtained from halimeter and OLT score. We could not include the results of this trial in the meta‐analysis because of the missing SD and P value in the results.

6o. Essential oil mouthwash versus placebo mouthwash: in a two‐arm parallel‐group RCT done by Garcia 2014, the group using essential oil mouthwash was compared to the group using placebo mouthwash. The outcome measure was VSC score measured using halimeter. We could not include the results of this trial in the meta‐analysis because of the missing SD and P value in the results.

6p. Cinnamon herbal mouthwash with alcohol; Nanosil mouthwash with hydrogen peroxide; Irsha mouthwash with alcohol versus water (negative control) and zinc solution (positive control):Talebian 2009 did a double‐blind, placebo‐controlled, randomised cross‐over study. The subjects were tested with cinnamon herbal mouthwash with alcohol, Nanosil mouthwash with hydrogen peroxide, Irsha mouthwash with alcohol and compared with a negative control ‐ water and a positive control ‐ zinc solution. The outcome measure was VSC score measured by halimeter. We could not include the results of this trial in the meta‐analysis as the report did not mention any data that could be used.

6q. Sesame oil versus chlorhexidine mouthwash: in a two‐arm, parallel‐group trial done by Asokan 2011, the efficacy of sesame oil was compared with the efficacy of chlorhexidine 0.2% mouthwash in the reduction of halitosis measured with OLT and BANA test. We could not include the results of this trial in the meta‐analysis because they did not report post‐intervention OLT score and P value.

7. Tablets

We have three comparisons using tablets reported by three trials.

7a. Protease cysteine + actinidine tablets versus placebo tablets: in a double‐blind, randomised cross‐over trial done by Nohno 2012, protease cysteine + actinidine tablets were compared to placebo tablets for seven days to reduce the tongue coating and thus the halitosis. The outcome measure was VSC (OralChroma).

7b. Lactobacillus β lactoperoxidase (LPO) tablets versus placebo tablets: in a two‐arm parallel‐group trial conducted by Nakano 2017, halitosis reduction was compared between groups consuming Lactobacillus ß LPO tablets and placebo tablets. The outcome measure was VSC score using OralChroma. The trial did not report any data that could be used in the meta‐analysis and hence could not be included in the analysis.

7c. Lactobacillus salivarius WB21 tablets versus placebo tablets: in a randomised, double‐blind, cross‐over, placebo‐controlled clinical trial with two arms conducted by Suzuki 2014, the reduction in halitosis was compared between subjects consuming Lactobacillus salivarius WB21 tablets versus placebo tablets. The outcome measures were OLT scores and VSC scores (gas chromatography). The trial did not give data that could be used in the meta‐analysis and there was no colour difference in the graph and hence we could not extract the data from the graph.

8. Combination methods

We have seven comparisons reported by seven trials under this section.

8a. Miswak stick versus chlorhexidine mouthwash: an unpublished clinical trial (NCT02628938) compared miswak (Salvadora persica) stick with chlorhexidine mouthwash twice a day, among female students who had halitosis. The outcome measures were OLT score (0: no appreciable odour and 5: extremely foul odour), VSC (Tanita FitScan HC‐212SF Breath Checker; 0: no odour and 5: intense odour) and patient self‐assessment score (10 cm VAS that is marked as 'no odour' on the 0 cm end, and as 'extremely foul odour' on the 10 cm end) after seven days.

8b. Brushing + mouthwash versus brushing + tongue cleaning:Aung 2015 conducted a single‐blind, parallel‐design, randomised controlled trial comparing three oral hygiene regimens for oral malodour reduction. Toothbrushing and mouthwashing with chlorine dioxide mouthwash was compared to toothbrushing and tongue cleaning after four weeks. The outcome measure was VSC using Breathron portable sulphide monitoring device. After four weeks, both the groups used toothbrushing plus mouthwashing with chlorine dioxide plus tongue cleaning and VSC was tested at the end of the fifth week. However, in our review, we have used the data at the end of four weeks only.

8c. Toothbrushing + rinsing with a 0.075% CPC mouthwash versus toothbrushing:Feres 2015 compared the efficacy of toothbrushing with fluoride toothpaste and CPC mouthwash to toothbrushing with fluoride toothpaste in their trial. The outcomes were measured using halimeter and OLT scores (0: no odour present and 5: extremely foul odour) after 21 days.

8d. Brushing + Turkish gall oral rinse versus brushing: a single‐blinded, randomised controlled trial was conducted by An 2011 to compare the effects of toothbrushing and oral rinsing with Turkish gall (traditional Chinese medicine) to toothbrushing alone. The outcome measures used were VSC (halimeter) and OLT scores (0 to 5) reported by the investigator.

8e. Laser with povidone iodine application versus SRP: one study (Kara 2008) was included for this comparison. The study had three arms, SRP (group I), subgingival laser irradiation combined with povidone‐iodine application (group II), and SRP and subgingival laser irradiation (group III). They have followed‐up for one week and four weeks after the intervention. The outcome measures used were OLT score and VSC. However, we have used the data from group I and II only for the one week follow‐up period only for both the outcome measures.

8f. Active rinse and active rinse + tongue scraping versus negative control rinse or negative control rinse + tongue scraping: in a four‐arm, cross‐over trial done by Ademovski 2012, reduction of halitosis was compared in subjects using an active rinse (water, glycerin, sorbitol, alcohol (1.8%), zinc acetate (0.3%), chlorhexidine diacetate (0.025%), sodium fluoride (0.05%), hydrogenated Castro oil, citric acid, acesulphame potassium, menthol and Mentha piperita), active rinse plus tongue scraping with a negative control rinse or negative control rinse plus tongue scraping. The outcome measures were OLT and VSC scores (OralChroma and halimeter). The authors did not report the correlation coefficient and we could not find a similar intervention trial to impute the SD of differences, thus excluding this trial from meta‐analysis.

8g. Toothbrushing with a reference toothpaste, toothbrushing with reference toothpaste + tongue cleaning, and toothbrushing + tongue cleaning with a tooth‐and‐tongue gel:Wilhelm 2012 conducted a single‐centre, examiner‐blind, randomised cross‐over trial in which the participants received each of the three interventions (toothbrushing with a reference toothpaste (1400 ppm F from sodium monofluorophosphate), toothbrushing with reference toothpaste and tongue cleaning, and toothbrushing and tongue cleaning with a tooth‐and‐tongue gel (Meridol halitosis tooth and tongue gel; 1400 ppm F ‐ from amine fluoride/stannous fluoride (ASF), 0.5% zinc lactate, oral malodour counter‐actives (OMCs)). The outcome measures were OLT, VSC (OralChroma CHM‐1, Abilit), and patient satisfaction scores. The authors did not report the correlation coefficient and we could not find a similar intervention trial to impute the SD of differences and thus we could not include this trial results in our meta‐analysis.

Outcome measuring methods

Fifteen trials measured the VSC levels as outcome (Aung 2015; Barak 2012; Caygur 2017; Garcia 2014; Lomax 2017; Nakano 2017; Niles 1999; Nogueira‐Filho 2002; Nohno 2012; Patil 2017; Payne 2011; Rassameemasmaung 2007; Rassameemasmaung 2012; Talebian 2009; Watanabe 2018). Three trials measured OLT scores as the outcome (Hu 2005; Lee 2018; Mamgain 2016). Twenty‐one trials measured both OLT scores as well as VSC levels as outcomes (Acar 2019; Ademovski 2012; Ademovski 2017; An 2011; Borden 2002; Dadamio 2013; Feres 2015; Iha 2013; Iwamura 2016; Kara 2008; Kozlovsky 1996; López Jornet 2003; NCT02628938; Navada 2008; Suzuki 2014; Tanaka 2010; Wang 2017; Wigger‐Alberti 2010; Wilhelm 2012; Winkel 2003; Wirthlin 2011). One study (Hu 2018) used OLT hedonic scores to measure the halitosis and another study (Asokan 2011) measured OLT score as well as self‐assessment of breath by the participants. One study measured VSC level along with self‐assessment as outcomes (Satthanakul 2014). Self‐assessment using VAS was used only in one trial (Nishihira 2017). Three outcome measurements, namely, OLT scores, VSC and breath print analysis were done in one trial (Marchetti 2015). Four studies (Iha 2013; Iwamura 2016; Watanabe 2018; Wirthlin 2011) reported VSC scores of different volatile gases (hydrogen sulphide, methyl mercaptan and methyl sulphide) rather than a compiled VSC score.

Quality of life

None of the included studies reported data on the outcome quality of life.

Adverse events

Seven trials reported adverse events (Borden 2002; Dadamio 2013; Lomax 2017; Patil 2017; Payne 2011; Winkel 2003; Wirthlin 2011). Other studies have either not given the details of adverse events or no adverse events were reported. Adverse events reported by Borden 2002 and Lomax 2017 were not related to the interventions. Dadamio 2013 reported unpleasant feeling and teeth staining in their trial. Patil 2017 trial reported burning mucosa and drying of mouth in few subjects using the control drug (chlorhexidine) and no adverse effects were reported in the intervention group (G32 tablets). Payne 2011 reported 19 non‐oral and 12 oral adverse effects. The oral effects were tingling sensation in lips, dry mouth or sore gums. Winkel 2003 reported discolouration of teeth and Wirthlin 2011 reported altered taste sensation as the adverse effect.

Studies awaiting classification

Eight trials are awaiting classification. Full texts were not available in the British Library for six trials (Cuihua 2009; Dongling 2017; Niles 2003; Rostoka 2012; Shimei 2014; Vazquez 2003), one trial (Gupta 2016) has not mentioned the inclusion criteria for healthy volunteers and we are waiting for translation of one report (Liang 2013).

Among these eight trials, three are in English (Gupta 2016; Niles 2003; Vazquez 2003), one in Russian (Rostoka 2012), and four in Chinese (Cuihua 2009; Dongling 2017; Liang 2013; Shimei 2014). Two are from the USA (Niles 2003; Vazquez 2003), one from India (Gupta 2016), one from Russia (Rostoka 2012), and four from China (Cuihua 2009; Dongling 2017; Liang 2013; Shimei 2014).

Ongoing studies

There are 18 ongoing studies (CTRI/2014/04/004519; CTRI/2018/05/014049; CTRI/2018/06/014686; DRKS00010618; IRCT201105136466N1; IRCT2014121520314N1; IRCT2015030921395N1; IRCT2016012026122N1; ISRCTN67671859; ISRCTN74655176; ISRCTN75902618; NCT02794766; NCT03031756; NCT03053882; NCT03160573; NCT03468595; TCTR20151109001; UMIN000023832).

Excluded studies

We have excluded 99 studies (104 reports): 52 studies were excluded because of short duration of intervention, either single dose or dose for less than a week duration, or outcomes were measured immediately after the intervention (Ademovski 2016; Alqumber 2014; Badanjak 2016; Bordas 2008; Boulware 1984; Carvalho 2004; Chen 2010; DRKS00005334; Farrell 2006; Farrell 2007; Frascella 1998; Frascella 2000; Farrell 2008; Feng 2010; Gerlach 1998; Greenstein 1997; Haas 2007; Leal 2019; Lodhia 2008; Nakano 2016; NCT00250289; NCT03346460; NCT03656419; NCT00655772; NCT00875927; Newby 2008; Pitts 1981; Porciani 2012; Reingewirtz 1999; Roldán 2004; Rolla 2002; Rosenberg 1992; Rosing 2009; Saad 2011; Saad 2016; Schmidt 1978; Seemann 2001; Sharma 1999; Sharma 2007; Shin 2011; Shinada 2008; Sterer 2008; Sterer 2013; Thrane 2010; Tian 2013; Uchida 1973; UMIN000002713; Wild 2001; Wilhelm 2010; Wilhelm 2013; Yaegaki 1992; Yoshimatsu 2007). 16 studies included participants with physiological malodour and thus were excluded (Faveri 2006; Keller 2012; Mendes 2016; NCT00748943; NL3100 (NTR3240); Peruzzo 2007; Peruzzo 2008; Quirynen 2002; Shinada 2010; Soares 2015; Steenberghe 2001; Tolentino 2011; Troccaz 2011; UMIN000002145; Van der Sluijs 2018; Wåler 1997). One study included pregnant women (Sheikh 2016). Six trials checked the outcomes related to the bacterial count (Fine 2005; NCT02194621; Quirynen 2004; Sreenivasan 2003; Sreenivasan 2004; Thaweboon 2011). Nine studies were excluded as the participants had advanced periodontitis (Betsy 2014; Moreno 2005; NCT02789436; Penala 2016; Silveira 2014; Silveria 2017; Soares 2015a; Quirynen 2005; Wang 2015). Six studies induced halitosis by requesting the participants to refrain from brushing (Brunette 1998; Codipilly 2004; Pedrazzi 2004; Seemann 2001a; Tamaki 2007; Yoshimatsu 2006). Five studies included patients with secondary halitosis (post‐surgical) or systemic disease and were excluded (Conceição 2008; EUCTR 2007‐003756‐11; Katsinelos 2007; NCT03591484; Polat 2008). Three studies were not related to halitosis outcome (Hu 2013; Malhotra 2011a; Wessel 2017), and one was an abstract publication (Mousquer 2017).

Risk of bias in included studies

We assessed three studies as at low risk of bias overall (Feres 2015; Lee 2018; Marchetti 2015). 16 studies had high risk of bias (An 2011; Asokan 2011; Aung 2015; Caygur 2017; Dadamio 2013; Iha 2013; Iwamura 2016; Mamgain 2016; NCT02628938; Nishihira 2017; Nohno 2012; Patil 2017; Payne 2011; Satthanakul 2014; Wang 2017; Wilhelm 2012) and the remaining 25 studies had unclear risk of bias (Figure 2).

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Allocation

32 studies had unclear risk of bias either in random sequence generation or allocation concealment processes (Acar 2019; Ademovski 2012; Ademovski 2017; Asokan 2011; Borden 2002; Caygur 2017; Garcia 2014; Hu 2005; Hu 2018; Iwamura 2016; Kara 2008; Kozlovsky 1996; Lomax 2017; López Jornet 2003; Mamgain 2016; Nakano 2017; Navada 2008; NCT02628938; Niles 1999; Nogueira‐Filho 2002; Nohno 2012; Rassameemasmaung 2007; Rassameemasmaung 2012; Satthanakul 2014; Suzuki 2014; Talebian 2009; Tanaka 2010; Wang 2017; Watanabe 2018; Wigger‐Alberti 2010; Wilhelm 2012; Winkel 2003). One study (Payne 2011) had high risk of selection bias and 11 studies had low risk of selection bias (Figure 3).

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Blinding

12 studies had either unclear risk of performance bias or unclear risk of detection bias (Acar 2019; Ademovski 2017; Garcia 2014; Hu 2005; Kara 2008; Kozlovsky 1996; López Jornet 2003; Nishihira 2017; Nogueira‐Filho 2002; Nohno 2012; Payne 2011; Talebian 2009). 11 studies had high risk of performance or detection bias (An 2011; Asokan 2011; Aung 2015; Caygur 2017; Iha 2013; Mamgain 2016; NCT02628938; Patil 2017; Satthanakul 2014; Wang 2017; Wilhelm 2012), and the remaining 21 studies had low risk of bias in blinding (Figure 3)

Incomplete outcome data

12 studies had unclear risk of attrition bias (An 2011; Borden 2002; Garcia 2014; Kozlovsky 1996; López Jornet 2003; Mamgain 2016; NCT02628938; Payne 2011; Satthanakul 2014; Wigger‐Alberti 2010; Wilhelm 2012; Winkel 2003). None of the studies had high risk of attrition bias and the remaining 32 studies had low risk of attrition bias (Figure 3).

Selective reporting

26 studies had unclear risk of reporting bias (Acar 2019; Ademovski 2012; Ademovski 2017; An 2011; Asokan 2011; Aung 2015; Barak 2012; Caygur 2017; Garcia 2014; Hu 2005; Hu 2018; Iha 2013; Kara 2008; Kozlovsky 1996; López Jornet 2003; Mamgain 2016; Navada 2008; Niles 1999; Nogueira‐Filho 2002; Nohno 2012; Patil 2017; Satthanakul 2014; Suzuki 2014; Wang 2017; Watanabe 2018; Wilhelm 2012), and one study had high risk of reporting bias (Nishihira 2017). The remaining 17 studies had low risk of reporting bias (Figure 3).

Other potential sources of bias

Six studies had unclear risk of other biases (Ademovski 2017; Aung 2015; Nishihira 2017; Patil 2017; Wigger‐Alberti 2010; Wirthlin 2011). Three studies had high risk of other biases (Dadamio 2013; Iwamura 2016; Nohno 2012), and the remaining 35 studies had low risk of other biases (Figure 3).

Effects of interventions

See: Summary of findings for the main comparison Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis; Summary of findings 2 0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis; Summary of findings 3 1000 mg champignon compared to placebo for managing halitosis; Summary of findings 4 Hinokitiol gel compared to placebo gel for managing halitosis; Summary of findings 5 0.3% triclosan toothpaste compared to control toothpaste for managing halitosis; Summary of findings 6 Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis; Summary of findings 7 Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis

Out of 44 included studies, we could analyse results from only 30 studies (38 reports).
We could not use the data from the other 14 studies (17 reports): we could not analyse data from Barak 2012; Garcia 2014 and López Jornet 2003 because of the missing SD and P values. Three studies (Marchetti 2015; Nakano 2017; Talebian 2009) did not give any data that could be used in the meta‐analysis. Asokan 2011 did not give post‐intervention OLT score and P value and hence could not be included in the meta‐analysis. Suzuki 2014 did not give data that could be used in the meta‐analysis and there was no colour difference in the graph and hence we could not extract the data from the graph. Payne 2011 gave the adjusted mean VSC scores in a graph and hence we could not use the data in the meta‐analysis. We could not calculate mean difference and impute correlation coefficient for Ademovski 2012; Niles 1999; Nogueira‐Filho 2002 and Wilhelm 2012 as there was no correlation coefficient reported. We could not find a similar intervention trial to impute the SD of differences. We calculated the data from the graph (Additional Table 1), however, we did not include the data from Wigger‐Alberti 2010 as the report did not give details of sample size per group.

Table 1. Wigger‐Alberti 2010 data

Mouthwash used	Median	Q1	Q2	Mean (calculated)
ASF	2.419	0.835	3.568	3.69
CHX + CPC + Zn	2.046	0.714	3.994	4.43
CHX	2.143	0.281	4.275	5.39
Tap water	2.695	1.147	4.719	4.39

Median, Q1, Q3 for 7 days follow‐up calculated from the graph using PlotDigitizer software.
ASF = amine fluoride/stannous fluoride; CHX = chlorhexidine; CPC = cetylpyridinium chloride; Zn = zinc.

We categorised the interventions found in the included studies under eight broad types and have explained the results based on the type of intervention as follows.

Mechanical debridement.
Chewing gum.
Systemic deodorising agent.
Topical agents.
Toothpaste.
Mouthrinse/mouthwash.
Tablets.
Combination methods.

We could not combine the interventions because the majority of the included trials had heterogenous interventions or control. As most of the trials reported data at multiple time points, the clinically relevant follow‐up time was considered in the meta‐analysis as described by the Cochrane Handbook for Systematic Reviews of Interventions Section 9.3.4 (Higgins 2011). We analysed data from cross‐over trials and parallel‐arm trials separately (Section 16.4.7, Higgins 2011). We did 'Summary of findings' tables for the most commonly used interventions with clinical outcomes. We discussed all the outcomes separately in the individual comparisons for ease of understanding. None of the included studies reported data on the outcome quality of life. Seven studies reported adverse events, other studies have either not given the details of adverse events or no adverse events were reported.

1. Mechanical debridement

1a. SRP + air polishing versus SRP: under this comparison, we had one trial (Caygur 2017) in which VSC was the outcome assessed. In this trial, SRP along with glycine powder air polishing was compared with SRP alone to see the effect on the VSC in halitosis patients. Using glycine powder air polishing adjunctively with SRP had no beneficial effects on halitosis when compared to SRP alone (mean difference (MD) ‐3.87; 95% confidence interval (CI) ‐17.93 to 10.19; 1 trial; 60 participants; 30 days follow‐up; Analysis 1.1).

1b. SRP + laser versus SRP: one trial (Kara 2008) assessed VSC under this comparison as outcome. The effect estimate for this outcome showed improvement in SRP + laser group compared to SRP group. However, the confidence interval crossed the line of no effect (MD ‐3.30; 95% CI ‐9.38 to 2.78; 1 trial; 40 participants; 4 weeks follow‐up; Analysis 2.1).

1c. Mechanical tongue cleaning versus no tongue cleaning: we have three trials under this comparison with two outcomes, VSC and OLT score. Acar 2019 and Wang 2017 were parallel‐arm trials and were analysed together.

VSC scores showed improvement in the intervention group with wider confidence intervals crossing the line of no effect (MD ‐7.69; 95% CI ‐47.08 to 31.69; 2 trials; 46 participants; 1 week follow‐up; Analysis 3.1).

OLT scores in parallel‐arm and cross‐over trials showed improvement in the intervention group (MD ‐0.20; 95% CI ‐0.34 to ‐0.07; 2 trials; 46 participants; 1 week follow‐up; Analysis 3.2).

2. Chewing gum

2a. 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum: one trial (Tanaka 2010) under this comparison reported two outcomes, VSC and OLT score. The effect estimate for OLT scores showed marginal improvement with wide confidence intervals crossing the line of no effect, in the 0.6% chewing gum group compared to 0.4% chewing gum group (MD ‐0.10; 95% CI ‐0.37 to 0.17; 1 trial; 64 participants; 4 weeks follow‐up; Analysis 4.1). However, the effect estimate for VSC did not show any improvement in the 0.6% chewing gum group (MD 0.00; 95% CI ‐0.21 to 0.21; 1 trial; 64 participants; 4 weeks follow‐up; Analysis 4.2).

2b. 0.6% eucalyptus chewing gum versus placebo chewing gum: one trial (Tanaka 2010) under this comparison reported two outcomes, VSC and OLT score. The effect estimate for OLT scores showed marginal improvement with wide confidence intervals crossing the line of no effect, in the 0.6% chewing gum group compared to placebo chewing gum group (MD ‐0.10; 95% CI ‐0.31 to 0.11; 1 trial; 65 participants; 4 weeks follow‐up; Analysis 5.1). However, the effect estimate for VSC did not show any improvement in the 0.6% chewing gum group (MD 0.00; 95% CI ‐0.21 to 0.21; 1 trial; 65 participants; 4 weeks follow‐up; Analysis 5.2).

2c. Pycnogenol chewing gum versus placebo chewing gum: one trial (Watanabe 2018) reported each component of VSC as the outcome under this comparison. The effect estimates are given for hydrogen sulphide, methyl mercaptan and methyl sulphide separately. All three components of VSC decreased in the intervention group compared to the placebo group. In the hydrogen sulphide outcome, the confidence intervals were wide and did not cross the line of no effect (MD ‐114.90; 95% CI ‐206.59 to ‐23.21; 1 trial; 21 participants; 4 weeks follow‐up; Analysis 6.1). However, in methyl mercaptan and methyl sulphide outcomes, the effect estimates showed wider confidence intervals crossing the line of no effect (MD ‐8.40; 95% CI ‐24.95 to 8.15; 1 trial; 21 participants; 4 weeks follow‐up; Analysis 6.2 and MD ‐4.70; 95% CI ‐27.01 to 17.61; 1 trial; 21 participants; 4 weeks follow‐up; Analysis 6.3 respectively).

3. Systemic deodorising agent

3a. 1000 mg champignon extract versus placebo: one trial (Nishihira 2017) reported patient score and patient's family member score in VAS. The effect estimates for both the outcomes showed marginal decrease of halitosis in the intervention group with wide confidence intervals crossing the line of no effect (MD ‐1.07; 95% CI ‐14.51 to 12.37; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 7.1 and MD ‐1.74; 95% CI ‐15.52 to 12.04; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 7.2 respectively).

3b. 1000 mg champignon versus 50 mg champignon extract: one trial (Nishihira 2017) reported patient score and patient's family member score in VAS. The effect estimates for both the outcomes showed marginal decrease of halitosis in the intervention group with wide confidence intervals crossing the line of no effect (MD ‐5.32; 95% CI ‐18.14 to 7.50; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 8.1 and MD ‐0.61; 95% CI ‐15.58 to 14.36; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 8.2 respectively).

4. Topical agents

4a. Hinokitiol gel versus placebo gel: one trial (Iha 2013) reported OLT scores and scores of two components of VSC (hydrogen sulphide and methyl mercaptan) for this comparison. The effect estimates for all three outcomes showed marginal decrease of halitosis in the hinokitiol gel group compared to placebo group with wide confidence intervals crossing the line of no effect (MD ‐0.27; 95% CI ‐1.26 to 0.72; 1 trial; 18 participants; 28 days follow‐up; Analysis 9.1, MD ‐2.13; 95% CI ‐5.33 to 1.08; 1 trial; 18 participants; 28 days follow‐up; Analysis 9.2 and MD ‐1.64; 95% CI ‐5.77 to 2.49; 1 trial; 18 participants; 28 days follow‐up; Analysis 9.3 respectively).

4b. G32 versus chlorhexidine gel: one trial (Patil 2017) reported VSC score as the outcome under this comparison. The effect estimate for this outcome showed marginal improvement in control (chlorhexidine gel) group compared to the intervention (G32 tablets) group (MD 0.05; 95%CI ‐0.28 to 0.38; 1 trial; 40 participants; 1 week follow‐up; Analysis 10.1). Regarding adverse events, Patil 2017 reported burning mucosa and drying of mouth in a few subjects using the control drug (chlorhexidine) and no adverse effects were reported in the intervention group (G32 tablets).

5. Toothpaste

5a. Triclosan + PVM/MA toothpaste versus control toothpaste: one trial (Hu 2005) reported OLT scores as reported by odour judges. The effect estimate for this outcome showed improvement in the intervention group compared to control group (MD ‐3.48; 95% CI ‐3.77 to ‐3.19; 1 trial; 81 participants; 1 week follow‐up; Analysis 11.1).

5b. Zinc toothpaste versus placebo toothpaste: one trial (Navada 2008) reported OLT score and VSC score for this comparison. The effect estimates for OLT and VSC outcomes showed improvement in the intervention group compared to the control group (MD ‐1.31; 95% CI ‐1.39 to ‐1.23; 1 trial; 187 participants; 4 weeks follow‐up; Analysis 12.1 and MD ‐11.30; 95% CI ‐20.45 to ‐2.15; 1 trial; 188 participants; 4 weeks follow‐up; Analysis 12.2 respectively).

5c. Sodium bicarbonate toothpaste versus control toothpaste: one trial (Lomax 2017) reported VSC score for this comparison. The effect estimate showed improvement in the placebo group compared to the intervention group with wide confidence interval crossing the line of no effect (MD 105.80; 95% CI ‐16.20 to 227.80; 1 trial; 148 participants; 6 weeks follow‐up; Analysis 13.1). Adverse events reported by Lomax 2017 were not related to the interventions.

5d. Dual zinc + arginine dentifrice versus control dentifrice: one trial (Hu 2018) reported OLT hedonic ratings. The effect estimate showed improvement in the intervention group compared to the control group (MD ‐2.00; 95% CI ‐2.19 to ‐1.81; 1 trial; 80 participants; 3 weeks follow‐up; Analysis 14.1).

6. Mouthrinse/mouthwash

Intervention mouthwash versus placebo mouthwash

6a. Halita versus placebo mouthwash: one trial (Winkel 2003) reported OLT score and VSC score as outcomes for this comparison. Both the effect estimates showed improvement in the intervention group compared to placebo group (MD ‐1.00; 95% CI ‐1.65 to ‐0.35; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 15.1 and MD ‐188.00; 95% CI ‐308.29 to ‐67.71; 1 trial; 40 participants; 2 weeks follow‐up; Analysis 15.2 respectively). Regarding adverse events, Winkel 2003 reported tongue staining was seen in patients who gargled, rather than rinsed in the halita mouthwash group.

6b. Chlorhexidine + zinc acetate mouthwash versus placebo mouthwash: one trial (Ademovski 2017) reported OLT score as outcome for this comparison. The effect estimate showed improvement in the intervention group compared to placebo group, however the confidence intervals crossed the line of no effect (MD ‐0.20; 95% CI ‐0.58 to 0.18; 1 trial; 44 participants; 3 months follow‐up; Analysis 16.1).

6c. Cetylperidinium chloride mouthwash versus placebo mouthwash: one trial (Borden 2002) reported OLT and VSC scores as outcomes for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group (MD ‐0.50; 95% CI ‐0.83 to ‐0.17; 1 trial; 47 participants; 2 weeks follow‐up; Analysis 17.1 and MD ‐20.04; 95% CI ‐37.71 to ‐2.37; 1 trial; 47 participants; 2 weeks follow‐up; Analysis 17.2 respectively). Adverse events reported by Borden 2002 were not related to the interventions.

6d. Essential oil mouthwash versus placebo mouthwash: one trial (Borden 2002) reported OLT and VSC scores as outcomes for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group, however the confidence intervals crossed the line of no effect (MD ‐0.09; 95% CI ‐0.47 to 0.29; 1 trial; 45 participants; 2 weeks follow‐up; Analysis 18.1 and MD ‐5.13; 95% CI ‐32.94 to 22.68; 1 trial; 45 participants; 2 weeks follow‐up; Analysis 18.2 respectively). Adverse events reported by Borden 2002 were not related to the interventions.

6e. Chlorine dioxide + zinc mouthwash versus placebo mouthwash: one trial (Borden 2002) reported both OLT and VSC scores as outcomes for this comparison. In this trial, both outcomes (OLT and VSC) showed improvement in the intervention group, however the confidence interval was crossing the line of no control in OLT outcome (MD ‐0.17; 95% CI ‐0.59 to 0.25; 1 trial; 41 participants; 2 weeks follow‐up; Analysis 19.1 and MD ‐20.53; 95% CI ‐38.52 to ‐2.54; 1 trial; 41 participants; 2 weeks follow‐up; Analysis 19.2 respectively). Adverse events reported by Borden 2002 were not related to the interventions.

6f. Chlorine dioxide mouthwash versus placebo mouthwash: one trial (Lee 2018) reported OLT score as outcome for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group (MD ‐0.61; 95% CI ‐0.73 to ‐0.49; 1 trial; 47 participants; 3 weeks follow‐up; Analysis 20.1). However, when week 6 data were used, the effect estimate favoured the intervention group with 95% CI crossing the line of no effect.

6g. Herbal mouthwash versus placebo mouthwash: one trial (Rassameemasmaung 2007) reported VSC score as outcome for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group (MD ‐70.29; 95% CI ‐121.01 to ‐19.57; 1 trial; 60 participants; 15 days follow‐up; Analysis 21.1).

6h. Benzethonium chloride mouthwash versus placebo mouthwash: one trial (Iwamura 2016) reported VSC scores of individual gases as outcomes for this comparison. The effect estimate showed no improvement in the intervention group compared to placebo group for the VSC score of methyl mercaptan (MD 7.20; 95% CI ‐24.92 to 39.32; 1 trial; 20 participants; 9 days follow‐up; Analysis 22.1). The effect estimates showed improvement in the intervention group compared to placebo group for the VSC scores of hydrogen sulphide and dimethyl sulphide and the confidence intervals crossed the line of no effect for both the outcomes (MD ‐125.10; 95% CI ‐286.32 to 36.12; 1 trial; 20 participants; 9 days follow‐up; Analysis 22.2 and MD ‐0.03; 95% CI ‐0.63 to 0.57; 1 trial; 20 participants; 9 days follow‐up; Analysis 22.3 respectively).

6i. Green tea mouthwash versus placebo mouthwash: one trial (Rassameemasmaung 2012) reported VSC score as the outcome for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group, however the confidence intervals crossed the line of no effect (MD ‐57.39; 95% CI ‐184.63 to 69.85; 1 trial; 60 participants; 28 days follow‐up; Analysis 23.1).

6j. Lemongrass mouthwash versus placebo mouthwash: one trial (Satthanakul 2014) reported VSC score as the outcome for this comparison. The effect estimates showed improvement in the intervention group compared to placebo group (MD ‐26.66; 95% CI ‐43.39 to ‐9.93; 1 trial; 20 participants; 8 days follow‐up; Analysis 24.1).

Intervention mouthwash versus control mouthwash

6k. Cetylpyridinium chloride mouthwash versus chlorhexidine + zinc mouthwash: one trial (Borden 2002) reported OLT and VSC scores as the outcomes for this comparison. The OLT effect estimate showed improvement in the cetylpyridinium chloride group compared to chlorhexidine + zinc group and the VSC score showed no improvement in the intervention group compared to control group. However, the confidence intervals in both the outcomes crossed the line of no effect (MD ‐0.33; 95% CI ‐0.72 to 0.06; 1 trial; 44 participants; 2 weeks follow‐up; Analysis 25.1 and MD 0.49; 95% CI ‐8.68 to 9.66; 1 trial; 44 participants; 2 weeks follow‐up; Analysis 25.2 respectively). Adverse events reported by Borden 2002 were not related to the interventions.

6l. Halita mouthrinse versus Perio‐plus mouthrinse: one trial (Dadamio 2013) reported OLT and VSC scores as the outcomes for this comparison. The effect estimates for both the outcomes showed improvement in the halita group compared to Perio‐plus mouthrinse, however the confidence intervals crossed the line of no effect (MD ‐0.20; 95% CI ‐0.86 to 0.46; 1 trial; 36 participants; 8 days follow‐up; Analysis 26.1 and MD ‐25.00; 95% CI ‐64.21 to 14.21; 1 trial; 36 participants; 8 days follow‐up; Analysis 26.2 respectively). Regarding adverse events, one patient from each group reported unpleasant feeling after the use of the product and one patient from the halita mouthrinse group reported tooth staining. There were no severe adverse events reported.

6m. Oil water two‐phase mouthwash versus control mouthwash: one trial (Kozlovsky 1996) reported VSC score as the outcome for this comparison. The effect estimates showed improvement in the oil water two‐phase mouthwash group compared to control group, however the confidence intervals crossed the line of no effect (MD ‐11.00; 95% CI ‐25.26 to 3.26; 1 trial; 50 participants; 1 week follow‐up; Analysis 27.1).

6n. Triphala and Ela decoction versus chlorhexidine mouthwash: one trial (Mamgain 2016) reported OLT score as the outcome for this comparison. The effect estimates showed no improvement in the Triphala and Ela decoction group compared to chlorhexidine group (MD 0.20; 95% CI 0.09 to 0.31; 1 trial; 60 participants; 14 days follow‐up; Analysis 28.1).

6o. Miswak mouthwash versus chlorhexidine mouthwash: one unpublished trial (NCT02628938) reported OLT and VSC scores and patient self‐assessment scores as the outcomes for this comparison. The effect estimate of the OLT outcome showed marginal improvement and effect estimates of the other two outcomes showed no improvement for the miswak mouthwash compared to chlorhexidine mouthwash (MD 0.01; 95% CI ‐0.95 to 0.97; 1 trial; 21 participants; 1 week follow‐up; Analysis 29.1, MD ‐0.20; 95% CI ‐1.03 to 0.63; 1 trial; 21 participants; 1 week follow‐up; Analysis 29.2 and MD ‐0.18; 95% CI ‐1.59 to 1.23; 1 trial; 21 participants; 1 week follow‐up; Analysis 29.3 respectively).

6p. Chlorine dioxide mouthrinse versus chlorhexidine mouthwash: one trial (Wirthlin 2011) reported VSC scores of individual gases as outcomes for this comparison. The effect estimate showed improvement in the chlorine dioxide group compared to chlorhexidine group for the VSC score of hydrogen sulphide and no improvement for the other two components of VSC (MD ‐11.00; 95% CI ‐31.61 to 9.61; 1 trial; 22 participants; 1 week follow‐up; Analysis 30.1, MD 7.63; 95% CI ‐1.70 to 16.96; 1 trial; 22 participants; 1 week follow‐up; Analysis 30.2 and MD 22.80; 95% CI ‐33.18 to 78.78; 1 trial; 22 participants; 1 week follow‐up; Analysis 30.3 respectively).Wirthlin 2011 reported altered taste sensation as the adverse event.

7. Tablets

7a. Protease cysteine + actinidine versus placebo tablets: one trial (Nohno 2012) reported VSC score as the outcome for this comparison. The effect estimates showed improvement in the intervention tablets group compared to placebo group however, the confidence intervals crossed the line of no effect (MD ‐45.80; 95% CI ‐258.38 to 166.78; 1 trial; 14 participants; 1 week follow‐up; Analysis 31.1).

8. Combination methods

8a. Miswak stick versus chlorhexidine mouthwash: one unpublished trial (NCT02628938) reported OLT and VSC scores and patient self‐assessment scores as the outcomes for this comparison. The effect estimate of all three outcomes showed improvement in the miswak mouthwash compared to chlorhexidine mouthwash. However, the confidence intervals crossed the line of no effect in OLT and patient self‐assessment scores (MD ‐0.55; 95% CI ‐1.33 to 0.23; 1 trial; 24 participants; 1 week follow‐up; Analysis 32.1, MD ‐0.77; 95% CI ‐1.19 to ‐0.35; 1 trial; 24 participants; 1 week follow‐up; Analysis 32.2 and MD ‐0.26; 95% CI ‐1.16 to 0.64; 1 trial; 24 participants; 1 week follow‐up; Analysis 32.3 respectively).

8b. Brushing + mouthwash versus brushing + tongue cleaning: one trial (Aung 2015) reported VSC score as the outcome for this comparison. The effect estimates showed improvement in the brushing plus mouthwash group compared to brushing plus tongue cleaning group (MD ‐81.87; 95% CI ‐140.12 to ‐23.62; 1 trial; 30 participants; 4 weeks follow‐up; Analysis 33.1).

8c. Brushing + cetylpyridium mouthwash versus brushing: one trial (Feres 2015) reported OLT and VSC score as the outcomes for this comparison. The effect estimates showed improvement in the brushing plus cetylpyridium mouthwash group compared to brushing group (MD ‐0.48; 95% CI ‐0.72 to ‐0.24; 1 trial; 70 participants; 3 weeks follow‐up; Analysis 34.1 and MD ‐8.04; 95% CI ‐15.87 to ‐0.21; 1 trial; 70 participants; 3 weeks follow‐up; Analysis 34.2 respectively).

8d. Turkish gall oral rinse versus brushing: one trial (An 2011) reported OLT and VSC score as the outcomes for this comparison. The effect estimates showed improvement in the Turkish gall oral rinse group compared to brushing group. However, the confidence intervals crossed the line of no effect (MD ‐0.10; 95% CI ‐0.50 to 0.30; 1 trial; 66 participants; 2 weeks follow‐up; Analysis 35.1 and MD ‐211.47; 95% CI ‐503.58 to 80.64; 1 trial; 66 participants; 2 weeks follow‐up; Analysis 35.2 respectively).

8e. Laser + povidone iodine versus SRP: one trial (Kara 2008) compared subgingival Nd:YAG laser irradiation combined with povidone‐iodine application with SRP. Both the effect estimates for OLT scores and VSC showed lesser improvement in laser group compared to SRP group (MD 0.49; 95% CI 0.30 to 0.68; 1 trial; 40 participants; 4 weeks follow‐up; Analysis 36.1 and MD 70.00; 95% CI 63.88 to 76.12; 1 trial; 40 participants; 4 weeks follow‐up; Analysis 36.2 respectively).

Discussion

Summary of main results

We found 36 comparisons in this Cochrane Review which were grouped for ease of understanding as comparisons related to eight broad interventions: mechanical debridement, chewing gum, systemic deodorising agent, topical agents, toothpaste, mouthrinse/mouthwash, tablets, and combination of methods.

The majority of the included trials presented the results for reduction in halitosis as self‐perceived or dentist‐perceived outcome or both in terms of organoleptic test (OLT) or visual analogue scale (VAS). Some studies used other methods like halimeter and breath print analysis. Adverse events were reported only in seven trials. Other included trials either not mentioned the adverse events or there were no adverse events. None of the trials reported quality of life as an outcome. These outcomes were reported for a minimum follow‐up period of seven days to a maximum follow‐up period of three months, with one to two weeks as the most commonly reported duration.

We produced 'Summary of findings' tables for the most commonly used interventions with findings also summarized for the rest of important comparisons and included under Additional tables. We considered only the clinical outcomes of commonly used interventions for the summary of main results. We could not find any COMET 2019 recommendations for the most important outcome measures. Hence, in this review, we considered OLT score reported by odour judges as the gold standard outcome measure (Quirynen 2018) followed by patient self‐assessment scores as the most important clinical outcomes. Certainty of the evidence was assessed for only these outcome measures and adverse events when reported. We found low‐ to very low‐certainty evidence for all the interventions included in this review and therefore we cannot draw any conclusions regarding the superiority of any of the interventions.

See summary of findings Table for the main comparison; summary of findings Table 2; summary of findings Table 3; summary of findings Table 4; summary of findings Table 5; summary of findings Table 6; summary of findings Table 7 and Additional Table 2; Table 3; Table 4; Table 5; Table 6; Table 7; Table 8; Table 9; Table 10; Table 11; Table 12; Table 13; Table 14; Table 15; Table 16.

Table 2. 0.6% eucalyptus chewing gum compared to 0.4% eucalyptus chewing gum for managing halitosis

0.6% eucalyptus chewing gum compared to 0.4% eucalyptus chewing gum for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 0.6% eucalyptus chewing gum Comparison: 0.4% eucalyptus chewing gum
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with 0.4% eucalyptus chewing gum	Risk with 0.6% eucalyptus chewing gum
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 1.60 units	MD 0.10 units lower (0.37 lower to 0.17 higher)	‐	64 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aTanaka 2010.
^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment.
^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

Table 3. 1000 mg champignon compared to 50 mg champignon for managing halitosis

1000 mg champignon compared to 50 mg champignon for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 1000 mg champignon Comparison: 50 mg champignon
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with 50 mg champignon	Risk with 1000 mg champignon
Dentist‐reported OLT score assessed with dentist's perception	‐	‐	‐	‐	‐	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 100 Follow‐up: mean 2 weeks	The mean patient‐reported VAS was 67.72 units	MD 5.32 units lower (18.14 lower to 7.50 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aNishihira 2017.
^bDowngraded for risk of bias ‐ unclear risk of performance and detection bias and high risk of reporting bias.
^cDowngraded for imprecision ‐ wide confidence interval crossing the line of no effect, low sample size and event rate.

Table 4. Toothpaste with 0.2% zinc sulphate compared to placebo toothpaste for managing halitosis

Toothpaste with 0.2% zinc sulphate compared to placebo toothpaste for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: toothpaste with 0.2% zinc sulphate Comparison: placebo toothpaste
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo toothpaste	Risk with toothpaste with 0.2% zinc sulphate
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT scores was 2.85 units	MD 1.31 units lower (1.39 lower to 1.23 lower)	‐	187 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aNavada 2008.
^bDowngraded for risk of bias ‐ unclear risk of selection bias in random sequence generation and allocation concealment.
^cDowngraded for imprecision ‐ low sample size and event rate.

Table 5. Dual zinc + arginine dentifrice compared to control dentifrice for managing halitosis

Dual zinc + arginine dentifrice compared to control dentifrice for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: dual zinc+ arginine dentifrice Comparison: control dentifrice
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with control dentifrice	Risk with dual zinc + arginine dentifrice
Dentist‐reported OLT hedonic ratings assessed with dentist's perception Scale from: 1 to 9 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT hedonic rating was 6.49 units	MD 2.00 units lower (2.19 lower to 1.81 lower)	‐	80 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aHu 2018.
^bDowngraded for risk of bias ‐ unclear risk of bias in random sequence generation and lack of allocation concealment details.
^cDowngraded for imprecision ‐ low sample size and event rate.

Table 6. Halita mouthwash compared to placebo mouthwash for managing halitosis

Halita mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: halita mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with halita mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 2.50 units	MD 1.00 units lower (1.65 lower to 0.35 lower)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	None reported	Tongue staining was seen in patients who gargled, rather than rinsed	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aWinkel 2003.
^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment and attrition bias.
^cDowngraded for imprecision ‐ low sample size and event rate.

Table 7. Cetylperidium chloride mouthwash compared to placebo mouthwash for managing halitosis

Cetylperidium chloride mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: cetylperidium chloride mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with cetylperidium chloride mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.50 units lower (0.83 lower to 0.17 lower)	‐	47 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aBorden 2002.
^bDowngraded for risk of bias ‐ unclear risk of bias due to random sequence generation, lack of allocation concealment, and attrition bias.
^cDowngraded for imprecision ‐ low sample size and event rate.

Table 8. Mouthwash containing essential oil compared to placebo mouthwash for managing halitosis

Mouthwash containing essential oil compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing essential oil Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing essential oil
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.09 units lower (0.47 lower to 0.29 higher)	‐	45 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aBorden 2002.
^bDowngraded for risk of bias ‐ unclear risk of bias due to random sequence generation, lack of allocation concealment and attrition bias.
^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate

Table 9. Mouthwash containing chlorine dioxide and zinc compared to placebo mouthwash for managing halitosis

Mouthwash containing chlorine dioxide and zinc compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing chlorine dioxide and zinc Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing chlorine dioxide and zinc
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.17 units lower (0.59 lower to 0.25 higher)	‐	41 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aBorden 2002.
^bDowngraded for risk of bias ‐ unclear risk of bias in random sequence generation, lack of allocation concealment and attrition bias.
^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Table 10. Chlorine dioxide mouthwash compared to placebo mouthwash for managing halitosis

Chlorine dioxide mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: chlorine dioxide mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with chlorine dioxide mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT score was 3.19 units	MD 0.61 units lower (0.73 lower to 0.49 lower)	‐	47 (1 RCT)^a	⊕⊕⊝⊝ LOW^b	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aLee 2018.
^bDowngraded for imprecision ‐ low sample size and event rate. However, when week 6 data were used, the effect estimate favoured placebo group with 95% CI crossing the line of no effect.

Table 11. Cetylpyridinium mouthwash compared to mouthwash containing chlorhexidine and zinc for managing halitosis

Cetylpyridinium mouthwash compared to mouthwash containing chlorhexidine and zinc for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: cetylpyridinium mouthwash Comparison: mouthwash containing chlorhexidine and zinc
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with mouthwash containing chlorhexidine and zinc	Risk with cetylpyridinium mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.03 units	MD 0.33 units lower (0.72 lower to 0.06 higher)	‐	44 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

Table 12. Halita mouthrinse compared to Perio‐plus mouthrinse for managing halitosis

Halita mouthrinse compared to Perio‐plus mouthrinse for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: halita mouthrinse Comparison: Perio‐plus mouthrinse
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with Perio‐plus mouthrinse	Risk with halita mouthrinse
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 8 days	The mean dentist‐reported OLT score was 1.40 units	MD 0.20 units lower (0.86 lower to 0.46 higher)	‐	36 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	1 patient reported unpleasant feeling after the use of the product. There were no severe adverse events reported	1 patient reported unpleasant feeling after the use of the product and 1 involving tooth staining. There were no severe adverse events reported	‐	36 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aDadamio 2013.
^bDowngraded for risk of bias ‐ high risk of other bias.
^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Table 13. Mouthwash containing Triphala and Ela decoction compared to chlorhexidine mouthwash for managing halitosis

Mouthwash containing Triphala and Ela decoction compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing Triphala and Ela decoction Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with mouthwash containing Triphala and Ela decoction
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 3.40 units	MD 0.20 units higher (0.09 higher to 0.31 higher)	‐	60 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aMamgain 2016.
^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, detection and attrition bias and high risk of performance bias.
^cDowngraded for imprecision ‐ low sample size and event rate.

Table 14. Miswak mouthwash compared to chlorhexidine mouthwash for managing halitosis

Miswak mouthwash compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: miswak mouthwash Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with miswak mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 1 week	The mean dentist‐reported OLT score was 1.09 units	MD 0.01 units higher (0.95 lower to 0.97 higher)	‐	21 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 10 Follow‐up: mean 1 week	The mean patient‐reported VAS was 2.18 units	MD 0.18 units lower (1.59 lower to 1.23 higher)	‐	21 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aNCT02628938.
^bDowngraded for risk of bias ‐ unclear risk of selection and attrition bias and high risk of performance bias.
^cDowngraded for imprecision ‐ low sample size and event rate; wide confidence intervals crossing the line of no effect.

Table 15. Miswak stick compared to chlorhexidine mouthwash for managing halitosis

Miswak stick compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: miswak stick Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with miswak stick
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 1 week	The mean dentist‐reported OLT score was 1.09 units	MD 0.55 units lower (1.33 lower to 0.23 higher)	‐	24 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 10 Follow‐up: mean 1 week	The mean patient‐reported VAS was 2.18 units	MD 0.26 units lower (1.16 lower to 0.64 higher)	‐	24 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

Table 16. Laser + povidone iodine compared to SRP alone for managing halitosis

Laser + povidone iodine compared to SRP alone for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: laser + povidone iodine Comparison: SRP alone
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with SRP alone	Risk with laser + povidone iodine
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 0.07 units	MD 0.49 units higher (0.30 higher to 0.68 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; SRP: scaling and root planing
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect

^aKara 2008.
^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, performance and detection bias.
^cDowngraded for imprecision ‐ low sample size and event rate.

Overall completeness and applicability of evidence

We systematically searched for trials according to the methodology written in the protocol. We checked all cross references of included articles and other systematic reviews on the management of halitosis to be sure that we did not miss any article. Two pairs of review authors did data extraction in duplicate. Trials which were not included in the meta‐analysis were explained qualitatively. We selected trials with adult participants treated for halitosis and included all types of interventions and concentrations. We included comparisons with placebo and control. All clinically relevant outcomes of interest were analysed. We also included trials in which herbal and alternative medicines were tested.

We did not exclude any trial due to missing data. For trials reporting data in graphs, we derived the data using PlotDigitizer software. When mean and standard error (SE) were given, we calculated the standard deviation (SD) as given in the Cochrane Handbook for Systematic Reviews of Interventions Section 7.7.3.3 (Higgins 2011). In cross‐over trials, mean difference (MD) and SE were calculated using the MD, imputing correlation coefficient (Corr) method as described in the Cochrane Handbook for Systematic Reviews of Interventions Section 16.4.6.3 (Higgins 2011). When mean and P value were given, SD was calculated according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions Section 7.7.3.3 (Higgins 2011). When median and interquartile range were given, we used the data to calculate mean and SD according to the methods described in the Cochrane Handbook for Systematic Reviews of Interventions Section 7.7.3.5 (Higgins 2011).

We have used data from the clinically relevant time points as the therapeutic effect could be at different durations. For scaling and root planing (SRP), we considered 4 weeks as time point as this is the time taken for epithelial and initial connective tissue healing. For antiseptic mouthwashes, we considered the therapeutic effect to happen at a minimum of seven days and this time point was considered in this review. We considered the reported minimal time point for other interventions (champignon, G32, sesame oil, protease cysteine + actinidine tablets, miswak, Turkish gall) for which we were not sure of the therapeutic effect time period.

In cross‐over trials, we used the data before and after cross‐over to see the consistency in the results. If the results were not consistent, we mentioned the same in the footnotes of the meta‐analysis.

In case of completed and unpublished trials, we cross‐checked the trial registry for any updated results and used it in the meta‐analysis.

We could not use the data from 14 trials (17 reports) due to improper reporting and missing details. 12 trials out of these 14 were published after the publication of the CONSORT Statement (Moher 2001). It is surprising to note that the reporting of trials is still an issue and not standardised.

Although we had 44 trials (55 reports) included in this review, most of the comparisons were single trials and could not be combined in meta‐analyses due to varying methods of intervention and concentrations. The evidence generated was also of very low to low quality for most of the comparisons testing an intervention versus placebo or control, and hence the results cannot be considered with certainty.

Most of the trials reported on short‐term improvement of halitosis (ranging from one to four weeks). Long‐term follow‐up (three months) was reported in one trial only (Ademovski 2017). The results cannot reflect the retention period for the improvement in halitosis as the oral hygiene maintenance issues would determine the long‐term success. However, the review encourages further high‐quality randomised controlled trials (RCTs) to be conducted by standardising methods of interventions, concentrations, and dosage.

Quality of the evidence

The certainty of the evidence for all comparisons was low to very low for the considered outcomes. We downgraded the trials mainly for two reasons: risk of bias and imprecision. Most of the trials were downgraded by one level for unclear risk of bias, by two levels for high risk of bias, and downgraded by two levels for imprecision as most were single trials with limited number of participants and low event rates.

Potential biases in the review process

We have taken steps to minimise bias in every step of the review. We searched all the above mentioned databases, conference proceedings, and trial registries to include all relevant reports. We included reports not in the English language in our review. We contacted trial authors for missing data through emails, peer‐contacts,Google search and university/hospital websites where they were previously affiliated. Nevertheless, there could be unpublished data which we could not trace with the above methods. We checked all cross‐references in the included articles and other systematic reviews conducted on interventions for halitosis and found articles which were missed in the search. Two review authors independently reviewed data extraction forms obtained from translators and cross‐checked doubtful areas using Google translator.

Agreements and disagreements with other studies or reviews

We found six systematic reviews published in the last two years. All these six reviews were limited to any one particular type of intervention. We could not find a review which covered all types of interventions.

Kellesarian 2017 conducted a systematic review to assess the efficacy of laser therapy and antimicrobial photodynamic therapy as an adjunct to mechanical debridement to manage halitosis. The review included six RCTs and concluded that the efficacy of laser therapy and antimicrobial photodynamic therapy to manage halitosis is unclear due to moderate to high risk of bias in the included trials. This review included advanced periodontitis cases and adolescents unlike the present Cochrane Review in which these two were excluded.

Muniz 2017 did a systematic review to analyse the impact of chewing gum on halitosis parameters. They concluded that chewing gums containing probiotics Lactobacillus, zinc acetate and magnolia bark extract, eucalyptus extract and allylisothiocyanate (AITC) with zinc lactate may be suitable for halitosis management. However, there is lack of evidence in applying these findings clinically due to high heterogeneity and low number of included studies. Unlike this Cochrane Review, Muniz 2017 included single‐dose interventions and follow‐ups of few hours.

Deutscher 2018 published a systematic review to give the best available evidence on the impact of professional tooth cleaning and SRP on oral halitosis in patients with periodontal diseases. They concluded that the professional tooth cleaning and SRP in combination with oral hygiene instructions reduced volatile sulphur compound (VSC) values in patients with oral halitosis or periodontal diseases or both, independent of tongue cleaning and the use of mouthrinses. Only controlled clinical trials (CCTs) were included in this review, unlike the present Cochrane Review, which included RCTs only. The difference in conclusions could be due to the patient‐related outcomes evaluated in the present Cochrane Review.

Wu 2018 reported a systematic review and meta‐analysis which evaluated the effective rate of Chinese medicine and combined Chinese and Western medicine on halitosis. It included 17 RCTs (10 intraoral halitosis and 7 extraoral halitosis) and concluded that both Chinese medicine and the combined Chinese and Western medicines have significantly better effect on halitosis than Western medicine alone. The authors included intraoral and extraoral halitosis, searched the Chinese biomedical databases CNKI, Wanfang and CBM, and had different exclusion criteria. The differences in the conclusion between Wu 2018 and our review could be because of these reasons.

Tahani 2018 did a systematic review to evaluate the clinical effect of green tea on halitosis. The search was limited to English language publications and included RCTs and quasi‐RCTs. The review included two RCTs out of which, one evaluated the short‐term effects and the other the long‐term effects of green tea. Due to the small number of included studies, this review was inconclusive.

Yoo 2019 published a systematic review and meta‐analysis of RCTs to summarize the evidence on the effect of probiotics on halitosis. The review concluded that the Lactobacillus strain given for an average of two weeks has a moderate effect on the OLT outcome and did not confirm the effect on the reduction of VSC. However, the authors included trials testing morning breath and patients with advanced periodontitis, unlike this Cochrane Review.

Figure 1

Study flow diagram.

Figure 2

Risk of bias graph: review authors' judgements about each risk of bias item presented as percentages across all included studies.

Figure 3

Risk of bias summary: review authors' judgements about each risk of bias item for each included study.

Analysis 1.1

Comparison 1 SRP + air polishing versus SRP, Outcome 1 VSC.

Analysis 2.1

Comparison 2 SRP + laser versus SRP, Outcome 1 VSC.

Analysis 3.1

Comparison 3 Mechanical tongue cleaning versus no tongue cleaning, Outcome 1 VSC.

Analysis 3.2

Comparison 3 Mechanical tongue cleaning versus no tongue cleaning, Outcome 2 Dentist‐reported OLT score.

Analysis 4.1

Comparison 4 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum, Outcome 1 Dentist‐reported OLT score.

Analysis 4.2

Comparison 4 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum, Outcome 2 VSC.

Analysis 5.1

Comparison 5 0.6% eucalyptus chewing gum versus placebo chewing gum, Outcome 1 Dentist‐reported OLT score.

Analysis 5.2

Comparison 5 0.6% eucalyptus chewing gum versus placebo chewing gum, Outcome 2 VSC.

Analysis 6.1

Comparison 6 Pycnogenol chewing gum versus placebo chewing gum, Outcome 1 VSC (hydrogen sulphide).

Analysis 6.2

Comparison 6 Pycnogenol chewing gum versus placebo chewing gum, Outcome 2 VSC (methyl mercaptan).

Analysis 6.3

Comparison 6 Pycnogenol chewing gum versus placebo chewing gum, Outcome 3 VSC (methyl sulphide).

Analysis 7.1

Comparison 7 1000 mg champignon versus placebo, Outcome 1 Patient‐reported VAS.

Analysis 7.2

Comparison 7 1000 mg champignon versus placebo, Outcome 2 Patient's relative‐reported VAS.

Analysis 8.1

Comparison 8 1000 mg champignon extract versus 50 mg champignon extract, Outcome 1 Patient‐reported VAS.

Analysis 8.2

Comparison 8 1000 mg champignon extract versus 50 mg champignon extract, Outcome 2 Patient's relative‐reported VAS.

Analysis 9.1

Comparison 9 Hinokitiol gel versus placebo gel, Outcome 1 Dentist‐reported OLT score.

Analysis 9.2

Comparison 9 Hinokitiol gel versus placebo gel, Outcome 2 VSC (methyl mercaptan).

Analysis 9.3

Comparison 9 Hinokitiol gel versus placebo gel, Outcome 3 VSC (hydrogen sulphide).

Analysis 10.1

Comparison 10 G32 versus chlorhexidine gel, Outcome 1 VSC.

Analysis 11.1

Comparison 11 Triclosan + PVM/MA toothpaste versus control toothpaste, Outcome 1 Dentist‐reported breath odour score.

Analysis 12.1

Comparison 12 Zinc toothpaste versus placebo toothpaste, Outcome 1 Dentist‐reported OLT score.

Analysis 12.2

Comparison 12 Zinc toothpaste versus placebo toothpaste, Outcome 2 VSC.

Analysis 13.1

Comparison 13 Sodium bicarbonate toothpaste versus control toothpaste, Outcome 1 VSC.

Analysis 14.1

Comparison 14 Dual zinc + arginine dentifrice versus control dentifrice, Outcome 1 OLT hedonic ratings.

Analysis 15.1

Comparison 15 Halita versus placebo mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 15.2

Comparison 15 Halita versus placebo mouthwash, Outcome 2 VSC.

Analysis 16.1

Comparison 16 Chlorhexidine + zinc acetate mouthwash versus placebo mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 17.1

Comparison 17 Cetylperidinium chloride mouthwash versus placebo, Outcome 1 Dentist‐reported OLT score.

Analysis 17.2

Comparison 17 Cetylperidinium chloride mouthwash versus placebo, Outcome 2 VSC.

Analysis 18.1

Comparison 18 Essential oil mouthwash versus placebo mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 18.2

Comparison 18 Essential oil mouthwash versus placebo mouthwash, Outcome 2 VSC.

Analysis 19.1

Comparison 19 Chlorine dioxide + zinc mouthwash versus placebo mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 19.2

Comparison 19 Chlorine dioxide + zinc mouthwash versus placebo mouthwash, Outcome 2 VSC.

Analysis 20.1

Comparison 20 Chlorine dioxide mouthwash versus placebo, Outcome 1 Dentist‐reported OLT score.

Analysis 21.1

Comparison 21 Herbal mouthwash versus placebo mouthwash, Outcome 1 VSC.

Analysis 22.1

Comparison 22 Benzethonium chloride mouthwash versus placebo mouthwash, Outcome 1 VSC (methyl mercaptan).

Analysis 22.2

Comparison 22 Benzethonium chloride mouthwash versus placebo mouthwash, Outcome 2 VSC (hydrogen sulphide).

Analysis 22.3

Comparison 22 Benzethonium chloride mouthwash versus placebo mouthwash, Outcome 3 VSC (dimethyl sulphide).

Analysis 23.1

Comparison 23 Green tea mouthwash versus placebo mouthwash, Outcome 1 VSC.

Analysis 24.1

Comparison 24 Lemongrass mouthwash versus placebo mouthwash, Outcome 1 VSC.

Analysis 25.1

Comparison 25 Cetylpyridinium chloride mouthwash versus chlorine dioxide + zinc mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 25.2

Comparison 25 Cetylpyridinium chloride mouthwash versus chlorine dioxide + zinc mouthwash, Outcome 2 VSC.

Analysis 26.1

Comparison 26 Halita mouthrinse versus Perio‐plus, Outcome 1 Dentist‐reported OLT score.

Analysis 26.2

Comparison 26 Halita mouthrinse versus Perio‐plus, Outcome 2 VSC.

Analysis 27.1

Comparison 27 Oil water 2‐phase mouthwash versus control mouthwash, Outcome 1 VSC.

Analysis 28.1

Comparison 28 Triphala and Ela decoction versus chlorhexidine mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 29.1

Comparison 29 Miswak mouthwash versus chlorhexidine mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 29.2

Comparison 29 Miswak mouthwash versus chlorhexidine mouthwash, Outcome 2 VSC.

Analysis 29.3

Comparison 29 Miswak mouthwash versus chlorhexidine mouthwash, Outcome 3 Patient self‐assessment score.

Analysis 30.1

Comparison 30 Chlorine dioxide mouthrinse versus chlorhexidine mouthwash, Outcome 1 VSC (hydrogen sulphide).

Analysis 30.2

Comparison 30 Chlorine dioxide mouthrinse versus chlorhexidine mouthwash, Outcome 2 VSC (methyl mercaptan).

Analysis 30.3

Comparison 30 Chlorine dioxide mouthrinse versus chlorhexidine mouthwash, Outcome 3 VSC (methyl sulphide).

Analysis 31.1

Comparison 31 Protease cysteine + actinidine versus placebo tablets, Outcome 1 VSC.

Analysis 32.1

Comparison 32 Miswak stick versus chlorhexidine mouthwash, Outcome 1 Dentist‐reported OLT score.

Analysis 32.2

Comparison 32 Miswak stick versus chlorhexidine mouthwash, Outcome 2 VSC.

Analysis 32.3

Comparison 32 Miswak stick versus chlorhexidine mouthwash, Outcome 3 Patient self‐assessment score.

Analysis 33.1

Comparison 33 Brushing + mouthwash versus brushing + tongue cleaning, Outcome 1 VSC.

Analysis 34.1

Comparison 34 Brushing + cetylpyridium mouthwash versus brushing, Outcome 1 Dentist‐reported OLT score.

Analysis 34.2

Comparison 34 Brushing + cetylpyridium mouthwash versus brushing, Outcome 2 VSC.

Analysis 35.1

Comparison 35 Turkish gall oral rinse versus brushing alone, Outcome 1 Dentist‐reported OLT score.

Analysis 35.2

Comparison 35 Turkish gall oral rinse versus brushing alone, Outcome 2 VSC.

Analysis 36.1

Comparison 36 Laser + povidone iodine versus SRP, Outcome 1 Dentist‐reported OLT score.

Analysis 36.2

Comparison 36 Laser + povidone iodine versus SRP, Outcome 2 VSC.

Summary of findings for the main comparison. Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis

Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mechanical tongue cleaning Comparison: no tongue cleaning
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with no tongue cleaning	Risk with mechanical tongue cleaning
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 1.804 units	MD 0.20 units lower (0.34 lower to 0.07 lower)	‐	46 (2 RCTs)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aAcar 2019; Wang 2017. ^bDowngraded for imprecision ‐ low sample size and event rate. ^cDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, selection bias, detection bias, and reporting bias. High risk of performance bias.

Summary of findings for the main comparison. Mechanical tongue cleaning compared to no tongue cleaning for managing halitosis

Summary of findings 2. 0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis

0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis
Patient or population: patients reporting halitosis Setting: University hospital Intervention: 0.6% eucalyptus chewing gum Comparison: placebo chewing gum
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo chewing gum	Risk with 0.6% eucalyptus chewing gum
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported organoleptic score was 1.60 units	MD 0.10 units lower (0.31 lower to 0.11 higher)	‐	65 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aTanaka 2010. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment. ^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

Summary of findings 2. 0.6% eucalyptus chewing gum compared to placebo chewing gum for managing halitosis

Summary of findings 3. 1000 mg champignon compared to placebo for managing halitosis

1000 mg champignon compared to placebo for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 1000 mg champignon Comparison: placebo
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo	Risk with 1000 mg champignon
Dentist‐reported OLT score assessed with dentist's perception	‐	‐	‐	‐	‐	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 100 Follow‐up: mean 2 weeks	The mean patient‐reported VAS was 63.47 units	MD 1.07 units lower (14.51 lower to 12.37 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNishihira 2017. ^bDowngraded for risk of bias ‐ unclear risk of performance and detection bias and high risk of bias in reporting bias. ^cDowngraded for imprecision ‐ wide confidence interval crossing the line of no effect, low sample size and event rate.

Summary of findings 3. 1000 mg champignon compared to placebo for managing halitosis

Summary of findings 4. Hinokitiol gel compared to placebo gel for managing halitosis

Hinokitiol gel compared to placebo gel for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: hinokitiol gel Comparison: placebo gel
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo gel	Risk with hinokitiol gel
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 2.10 units	MD 0.27 units lower (1.26 lower to 0.72 higher)	‐	18 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aIha 2013. ^bDowngraded for risk of bias ‐ high risk of performance and detection bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Summary of findings 4. Hinokitiol gel compared to placebo gel for managing halitosis

Summary of findings 5. 0.3% triclosan toothpaste compared to control toothpaste for managing halitosis

0.3% triclosan toothpaste compared to control toothpaste for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 0.3% triclosan toothpaste Comparison: control toothpaste
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with control toothpaste	Risk with 0.3% triclosan toothpaste
Dentist‐reported breath odour score assessed with dentist's perception Scale from: 1 to 9 Follow‐up: mean 1 week	The mean dentist‐reported breath odour score was 7.14 units	MD 3.48 units lower (3.77 lower to 3.19 lower)	‐	81 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aHu 2005. ^bDowngraded for risk of bias ‐ unclear risk of bias due to improper selection, lack of allocation concealment, performance, detection and reporting. ^cDowngraded for imprecision ‐ low sample size and event rate.

Summary of findings 5. 0.3% triclosan toothpaste compared to control toothpaste for managing halitosis

Summary of findings 6. Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis

Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing chlorhexidine and zinc acetate Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing chlorhexidine and zinc acetate
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 months	The mean dentist‐reported OLT score was 2.30 units	MD 0.20 units lower (0.58 lower to 0.18 higher)	‐	44 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aAdemovski 2017. ^bDowngraded for risk of bias ‐ unclear risk of selection bias, detection bias and other bias. ^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

Summary of findings 6. Mouthwash containing chlorhexidine and zinc acetate compared to placebo mouthwash for managing halitosis

Summary of findings 7. Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis

Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: brushing + cetylpyridium mouthwash Comparison: brushing
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with brushing	Risk with brushing + cetylpyridium mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT score was 1.37 units	MD 0.48 units lower (0.72 lower to 0.24 lower)	‐	70 (1 RCT)^a	⊕⊕⊝⊝ LOW^b	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aFeres 2015. ^bDowngraded for imprecision ‐ low sample size and event rate.

Summary of findings 7. Brushing + cetylpyridium mouthwash compared to brushing for managing halitosis

Table 1. Wigger‐Alberti 2010 data

Mouthwash used	Median	Q1	Q2	Mean (calculated)
ASF	2.419	0.835	3.568	3.69
CHX + CPC + Zn	2.046	0.714	3.994	4.43
CHX	2.143	0.281	4.275	5.39
Tap water	2.695	1.147	4.719	4.39
Median, Q1, Q3 for 7 days follow‐up calculated from the graph using PlotDigitizer software. ASF = amine fluoride/stannous fluoride; CHX = chlorhexidine; CPC = cetylpyridinium chloride; Zn = zinc.

Table 1. Wigger‐Alberti 2010 data

Table 2. 0.6% eucalyptus chewing gum compared to 0.4% eucalyptus chewing gum for managing halitosis

0.6% eucalyptus chewing gum compared to 0.4% eucalyptus chewing gum for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 0.6% eucalyptus chewing gum Comparison: 0.4% eucalyptus chewing gum
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with 0.4% eucalyptus chewing gum	Risk with 0.6% eucalyptus chewing gum
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 1.60 units	MD 0.10 units lower (0.37 lower to 0.17 higher)	‐	64 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aTanaka 2010. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment. ^cDowngraded for imprecision ‐ wide confidence intervals, low sample size and event rate.

Table 2. 0.6% eucalyptus chewing gum compared to 0.4% eucalyptus chewing gum for managing halitosis

Table 3. 1000 mg champignon compared to 50 mg champignon for managing halitosis

1000 mg champignon compared to 50 mg champignon for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: 1000 mg champignon Comparison: 50 mg champignon
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with 50 mg champignon	Risk with 1000 mg champignon
Dentist‐reported OLT score assessed with dentist's perception	‐	‐	‐	‐	‐	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 100 Follow‐up: mean 2 weeks	The mean patient‐reported VAS was 67.72 units	MD 5.32 units lower (18.14 lower to 7.50 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNishihira 2017. ^bDowngraded for risk of bias ‐ unclear risk of performance and detection bias and high risk of reporting bias. ^cDowngraded for imprecision ‐ wide confidence interval crossing the line of no effect, low sample size and event rate.

Table 3. 1000 mg champignon compared to 50 mg champignon for managing halitosis

Table 4. Toothpaste with 0.2% zinc sulphate compared to placebo toothpaste for managing halitosis

Toothpaste with 0.2% zinc sulphate compared to placebo toothpaste for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: toothpaste with 0.2% zinc sulphate Comparison: placebo toothpaste
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo toothpaste	Risk with toothpaste with 0.2% zinc sulphate
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT scores was 2.85 units	MD 1.31 units lower (1.39 lower to 1.23 lower)	‐	187 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNavada 2008. ^bDowngraded for risk of bias ‐ unclear risk of selection bias in random sequence generation and allocation concealment. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 4. Toothpaste with 0.2% zinc sulphate compared to placebo toothpaste for managing halitosis

Table 5. Dual zinc + arginine dentifrice compared to control dentifrice for managing halitosis

Dual zinc + arginine dentifrice compared to control dentifrice for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: dual zinc+ arginine dentifrice Comparison: control dentifrice
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with control dentifrice	Risk with dual zinc + arginine dentifrice
Dentist‐reported OLT hedonic ratings assessed with dentist's perception Scale from: 1 to 9 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT hedonic rating was 6.49 units	MD 2.00 units lower (2.19 lower to 1.81 lower)	‐	80 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aHu 2018. ^bDowngraded for risk of bias ‐ unclear risk of bias in random sequence generation and lack of allocation concealment details. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 5. Dual zinc + arginine dentifrice compared to control dentifrice for managing halitosis

Table 6. Halita mouthwash compared to placebo mouthwash for managing halitosis

Halita mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: halita mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with halita mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 2.50 units	MD 1.00 units lower (1.65 lower to 0.35 lower)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	None reported	Tongue staining was seen in patients who gargled, rather than rinsed	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aWinkel 2003. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment and attrition bias. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 6. Halita mouthwash compared to placebo mouthwash for managing halitosis

Table 7. Cetylperidium chloride mouthwash compared to placebo mouthwash for managing halitosis

Cetylperidium chloride mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: cetylperidium chloride mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with cetylperidium chloride mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.50 units lower (0.83 lower to 0.17 lower)	‐	47 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aBorden 2002. ^bDowngraded for risk of bias ‐ unclear risk of bias due to random sequence generation, lack of allocation concealment, and attrition bias. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 7. Cetylperidium chloride mouthwash compared to placebo mouthwash for managing halitosis

Table 8. Mouthwash containing essential oil compared to placebo mouthwash for managing halitosis

Mouthwash containing essential oil compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing essential oil Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing essential oil
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.09 units lower (0.47 lower to 0.29 higher)	‐	45 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aBorden 2002. ^bDowngraded for risk of bias ‐ unclear risk of bias due to random sequence generation, lack of allocation concealment and attrition bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate

Table 8. Mouthwash containing essential oil compared to placebo mouthwash for managing halitosis

Table 9. Mouthwash containing chlorine dioxide and zinc compared to placebo mouthwash for managing halitosis

Mouthwash containing chlorine dioxide and zinc compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing chlorine dioxide and zinc Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with mouthwash containing chlorine dioxide and zinc
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.20 units	MD 0.17 units lower (0.59 lower to 0.25 higher)	‐	41 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aBorden 2002. ^bDowngraded for risk of bias ‐ unclear risk of bias in random sequence generation, lack of allocation concealment and attrition bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Table 9. Mouthwash containing chlorine dioxide and zinc compared to placebo mouthwash for managing halitosis

Table 10. Chlorine dioxide mouthwash compared to placebo mouthwash for managing halitosis

Chlorine dioxide mouthwash compared to placebo mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: chlorine dioxide mouthwash Comparison: placebo mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with placebo mouthwash	Risk with chlorine dioxide mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 3 weeks	The mean dentist‐reported OLT score was 3.19 units	MD 0.61 units lower (0.73 lower to 0.49 lower)	‐	47 (1 RCT)^a	⊕⊕⊝⊝ LOW^b	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aLee 2018. ^bDowngraded for imprecision ‐ low sample size and event rate. However, when week 6 data were used, the effect estimate favoured placebo group with 95% CI crossing the line of no effect.

Table 10. Chlorine dioxide mouthwash compared to placebo mouthwash for managing halitosis

Table 11. Cetylpyridinium mouthwash compared to mouthwash containing chlorhexidine and zinc for managing halitosis

Cetylpyridinium mouthwash compared to mouthwash containing chlorhexidine and zinc for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: cetylpyridinium mouthwash Comparison: mouthwash containing chlorhexidine and zinc
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with mouthwash containing chlorhexidine and zinc	Risk with cetylpyridinium mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 4.03 units	MD 0.33 units lower (0.72 lower to 0.06 higher)	‐	44 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aBorden 2002. ^bDowngraded for risk of bias ‐ unclear risk of bias in random sequence generation, lack of allocation concealment and attrition bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Table 11. Cetylpyridinium mouthwash compared to mouthwash containing chlorhexidine and zinc for managing halitosis

Table 12. Halita mouthrinse compared to Perio‐plus mouthrinse for managing halitosis

Halita mouthrinse compared to Perio‐plus mouthrinse for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: halita mouthrinse Comparison: Perio‐plus mouthrinse
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with Perio‐plus mouthrinse	Risk with halita mouthrinse
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 8 days	The mean dentist‐reported OLT score was 1.40 units	MD 0.20 units lower (0.86 lower to 0.46 higher)	‐	36 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	1 patient reported unpleasant feeling after the use of the product. There were no severe adverse events reported	1 patient reported unpleasant feeling after the use of the product and 1 involving tooth staining. There were no severe adverse events reported	‐	36 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aDadamio 2013. ^bDowngraded for risk of bias ‐ high risk of other bias. ^cDowngraded for imprecision ‐ wide confidence interval, low sample size and event rate.

Table 12. Halita mouthrinse compared to Perio‐plus mouthrinse for managing halitosis

Table 13. Mouthwash containing Triphala and Ela decoction compared to chlorhexidine mouthwash for managing halitosis

Mouthwash containing Triphala and Ela decoction compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: mouthwash containing Triphala and Ela decoction Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with mouthwash containing Triphala and Ela decoction
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 2 weeks	The mean dentist‐reported OLT score was 3.40 units	MD 0.20 units higher (0.09 higher to 0.31 higher)	‐	60 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aMamgain 2016. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, detection and attrition bias and high risk of performance bias. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 13. Mouthwash containing Triphala and Ela decoction compared to chlorhexidine mouthwash for managing halitosis

Table 14. Miswak mouthwash compared to chlorhexidine mouthwash for managing halitosis

Miswak mouthwash compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: miswak mouthwash Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with miswak mouthwash
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 1 week	The mean dentist‐reported OLT score was 1.09 units	MD 0.01 units higher (0.95 lower to 0.97 higher)	‐	21 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 10 Follow‐up: mean 1 week	The mean patient‐reported VAS was 2.18 units	MD 0.18 units lower (1.59 lower to 1.23 higher)	‐	21 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNCT02628938. ^bDowngraded for risk of bias ‐ unclear risk of selection and attrition bias and high risk of performance bias. ^cDowngraded for imprecision ‐ low sample size and event rate; wide confidence intervals crossing the line of no effect.

Table 14. Miswak mouthwash compared to chlorhexidine mouthwash for managing halitosis

Table 15. Miswak stick compared to chlorhexidine mouthwash for managing halitosis

Miswak stick compared to chlorhexidine mouthwash for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: miswak stick Comparison: chlorhexidine mouthwash
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with chlorhexidine mouthwash	Risk with miswak stick
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 1 week	The mean dentist‐reported OLT score was 1.09 units	MD 0.55 units lower (1.33 lower to 0.23 higher)	‐	24 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported VAS assessed with patient's perception Scale from: 0 to 10 Follow‐up: mean 1 week	The mean patient‐reported VAS was 2.18 units	MD 0.26 units lower (1.16 lower to 0.64 higher)	‐	24 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; VAS: visual analogue scale
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aNCT02628938. ^bDowngraded for risk of bias ‐ unclear risk of selection and attrition bias and high risk of performance bias. ^cDowngraded for imprecision ‐ low sample size and event rate; wide confidence intervals crossing the line of no effect.

Table 15. Miswak stick compared to chlorhexidine mouthwash for managing halitosis

Table 16. Laser + povidone iodine compared to SRP alone for managing halitosis

Laser + povidone iodine compared to SRP alone for managing halitosis
Patient or population: patients reporting halitosis Setting: university hospital Intervention: laser + povidone iodine Comparison: SRP alone
Outcomes	*Anticipated absolute effects^ (95% CI)**		Relative effect (95% CI)	Number of participants (studies)	Certainty of the evidence (GRADE)	Comments
	Risk with SRP alone	Risk with laser + povidone iodine
Dentist‐reported OLT score assessed with dentist's perception Scale from: 0 to 5 Follow‐up: mean 4 weeks	The mean dentist‐reported OLT score was 0.07 units	MD 0.49 units higher (0.30 higher to 0.68 higher)	‐	40 (1 RCT)^a	⊕⊝⊝⊝ VERY LOW^b,c	‐
Patient‐reported OLT score assessed with patient's perception	‐	‐	‐	‐	‐	‐
Adverse events	‐	‐	‐	‐	‐	‐
^*The risk in the intervention group (and its 95% CI) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: confidence interval; MD: mean difference; OLT: organoleptic test; RCT: randomised controlled trial; SRP: scaling and root planing
GRADE Working Group grades of evidence High certainty: we are very confident that the true effect lies close to that of the estimate of the effect Moderate certainty: we are moderately confident in the effect estimate: the true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different Low certainty: our confidence in the effect estimate is limited: the true effect may be substantially different from the estimate of the effect Very low certainty: we have very little confidence in the effect estimate: the true effect is likely to be substantially different from the estimate of effect
^aKara 2008. ^bDowngraded for risk of bias ‐ unclear risk of bias due to lack of allocation concealment, performance and detection bias. ^cDowngraded for imprecision ‐ low sample size and event rate.

Table 16. Laser + povidone iodine compared to SRP alone for managing halitosis

Comparison 1. SRP + air polishing versus SRP

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	60	Mean Difference (IV, Random, 95% CI)	‐3.87 [‐17.93, 10.19]

Comparison 1. SRP + air polishing versus SRP

Comparison 2. SRP + laser versus SRP

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐3.30 [‐9.38, 2.78]

Comparison 2. SRP + laser versus SRP

Comparison 3. Mechanical tongue cleaning versus no tongue cleaning

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	2	46	Mean Difference (IV, Random, 95% CI)	‐7.69 [‐47.08, 31.69]

2 Dentist‐reported OLT score Show forest plot	2	46	Mean Difference (IV, Random, 95% CI)	‐0.20 [‐0.34, ‐0.07]

Comparison 3. Mechanical tongue cleaning versus no tongue cleaning

Comparison 4. 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	64	Mean Difference (IV, Random, 95% CI)	‐0.10 [‐0.37, 0.17]

2 VSC Show forest plot	1	64	Mean Difference (IV, Random, 95% CI)	0.0 [‐0.21, 0.21]

Comparison 4. 0.6% eucalyptus chewing gum versus 0.4% eucalyptus chewing gum

Comparison 5. 0.6% eucalyptus chewing gum versus placebo chewing gum

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	65	Mean Difference (IV, Random, 95% CI)	‐0.10 [‐0.31, 0.11]

2 VSC Show forest plot	1	65	Mean Difference (IV, Random, 95% CI)	0.0 [‐0.21, 0.21]

Comparison 5. 0.6% eucalyptus chewing gum versus placebo chewing gum

Comparison 6. Pycnogenol chewing gum versus placebo chewing gum

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC (hydrogen sulphide) Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	‐114.90 [‐206.59, ‐23.21]

2 VSC (methyl mercaptan) Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	‐8.4 [‐24.95, 8.15]

3 VSC (methyl sulphide) Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	‐4.70 [‐27.01, 17.61]

Comparison 6. Pycnogenol chewing gum versus placebo chewing gum

Comparison 7. 1000 mg champignon versus placebo

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Patient‐reported VAS Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐1.07 [‐14.51, 12.37]

2 Patient's relative‐reported VAS Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐1.74 [‐15.52, 12.04]

Comparison 7. 1000 mg champignon versus placebo

Comparison 8. 1000 mg champignon extract versus 50 mg champignon extract

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Patient‐reported VAS Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐5.32 [‐18.14, 7.50]

2 Patient's relative‐reported VAS Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐0.61 [‐15.58, 14.36]

Comparison 8. 1000 mg champignon extract versus 50 mg champignon extract

Comparison 9. Hinokitiol gel versus placebo gel

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	18	Mean Difference (IV, Random, 95% CI)	‐0.27 [‐1.26, 0.72]

2 VSC (methyl mercaptan) Show forest plot	1	18	Mean Difference (IV, Random, 95% CI)	‐2.13 [‐5.33, 1.08]

3 VSC (hydrogen sulphide) Show forest plot	1	18	Mean Difference (IV, Random, 95% CI)	‐1.64 [‐5.77, 2.49]

Comparison 9. Hinokitiol gel versus placebo gel

Comparison 10. G32 versus chlorhexidine gel

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	0.05 [‐0.28, 0.38]

Comparison 10. G32 versus chlorhexidine gel

Comparison 11. Triclosan + PVM/MA toothpaste versus control toothpaste

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported breath odour score Show forest plot	1	81	Mean Difference (IV, Random, 95% CI)	‐3.48 [‐3.77, ‐3.19]

Comparison 11. Triclosan + PVM/MA toothpaste versus control toothpaste

Comparison 12. Zinc toothpaste versus placebo toothpaste

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	187	Mean Difference (IV, Random, 95% CI)	‐1.31 [‐1.39, ‐1.23]

2 VSC Show forest plot	1	188	Mean Difference (IV, Random, 95% CI)	‐11.30 [‐20.45, ‐2.15]

Comparison 12. Zinc toothpaste versus placebo toothpaste

Comparison 13. Sodium bicarbonate toothpaste versus control toothpaste

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	148	Mean Difference (IV, Random, 95% CI)	105.80 [‐16.20, 227.80]

Comparison 13. Sodium bicarbonate toothpaste versus control toothpaste

Comparison 14. Dual zinc + arginine dentifrice versus control dentifrice

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 OLT hedonic ratings Show forest plot	1	80	Mean Difference (IV, Random, 95% CI)	‐2.0 [‐2.19, ‐1.81]

Comparison 14. Dual zinc + arginine dentifrice versus control dentifrice

Comparison 15. Halita versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐1.0 [‐1.65, ‐0.35]

2 VSC Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	‐188.0 [‐308.29, ‐67.71]

Comparison 15. Halita versus placebo mouthwash

Comparison 16. Chlorhexidine + zinc acetate mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	44	Mean Difference (IV, Random, 95% CI)	‐0.20 [‐0.58, 0.18]

Comparison 16. Chlorhexidine + zinc acetate mouthwash versus placebo mouthwash

Comparison 17. Cetylperidinium chloride mouthwash versus placebo

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	47	Mean Difference (IV, Random, 95% CI)	‐0.5 [‐0.83, ‐0.17]

2 VSC Show forest plot	1	47	Mean Difference (IV, Random, 95% CI)	‐20.04 [‐37.71, ‐2.37]

Comparison 17. Cetylperidinium chloride mouthwash versus placebo

Comparison 18. Essential oil mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	45	Mean Difference (IV, Random, 95% CI)	‐0.09 [‐0.47, 0.29]

2 VSC Show forest plot	1	45	Mean Difference (IV, Random, 95% CI)	‐5.13 [‐32.94, 22.68]

Comparison 18. Essential oil mouthwash versus placebo mouthwash

Comparison 19. Chlorine dioxide + zinc mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	41	Mean Difference (IV, Random, 95% CI)	‐0.17 [‐0.59, 0.25]

2 VSC Show forest plot	1	41	Mean Difference (IV, Random, 95% CI)	‐20.53 [‐38.52, ‐2.54]

Comparison 19. Chlorine dioxide + zinc mouthwash versus placebo mouthwash

Comparison 20. Chlorine dioxide mouthwash versus placebo

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	47	Mean Difference (Random, 95% CI)	‐0.61 [‐0.73, ‐0.49]

Comparison 20. Chlorine dioxide mouthwash versus placebo

Comparison 21. Herbal mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	60	Mean Difference (IV, Random, 95% CI)	‐70.29 [‐121.01, ‐19.57]

Comparison 21. Herbal mouthwash versus placebo mouthwash

Comparison 22. Benzethonium chloride mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC (methyl mercaptan) Show forest plot	1	20	Mean Difference (IV, Random, 95% CI)	7.20 [‐24.92, 39.32]

2 VSC (hydrogen sulphide) Show forest plot	1	20	Mean Difference (IV, Random, 95% CI)	‐125.10 [‐286.32, 36.12]

3 VSC (dimethyl sulphide) Show forest plot	1	20	Mean Difference (IV, Random, 95% CI)	‐0.03 [‐0.63, 0.57]

Comparison 22. Benzethonium chloride mouthwash versus placebo mouthwash

Comparison 23. Green tea mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	60	Mean Difference (IV, Random, 95% CI)	‐57.39 [‐184.63, 69.85]

Comparison 23. Green tea mouthwash versus placebo mouthwash

Comparison 24. Lemongrass mouthwash versus placebo mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	20	Mean Difference (IV, Random, 95% CI)	‐26.66 [‐43.39, ‐9.93]

Comparison 24. Lemongrass mouthwash versus placebo mouthwash

Comparison 25. Cetylpyridinium chloride mouthwash versus chlorine dioxide + zinc mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	44	Mean Difference (IV, Random, 95% CI)	‐0.33 [‐0.72, 0.06]

2 VSC Show forest plot	1	44	Mean Difference (IV, Random, 95% CI)	0.49 [‐8.68, 9.66]

Comparison 25. Cetylpyridinium chloride mouthwash versus chlorine dioxide + zinc mouthwash

Comparison 26. Halita mouthrinse versus Perio‐plus

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	36	Mean Difference (IV, Random, 95% CI)	‐0.20 [‐0.86, 0.46]

2 VSC Show forest plot	1	36	Mean Difference (IV, Random, 95% CI)	‐25.0 [‐64.21, 14.21]

Comparison 26. Halita mouthrinse versus Perio‐plus

Comparison 27. Oil water 2‐phase mouthwash versus control mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	50	Mean Difference (IV, Random, 95% CI)	‐11.0 [‐25.26, 3.26]

Comparison 27. Oil water 2‐phase mouthwash versus control mouthwash

Comparison 28. Triphala and Ela decoction versus chlorhexidine mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	60	Mean Difference (IV, Random, 95% CI)	0.20 [0.09, 0.31]

Comparison 28. Triphala and Ela decoction versus chlorhexidine mouthwash

Comparison 29. Miswak mouthwash versus chlorhexidine mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	0.01 [‐0.95, 0.97]

2 VSC Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	‐0.20 [‐1.03, 0.63]

3 Patient self‐assessment score Show forest plot	1	21	Mean Difference (IV, Random, 95% CI)	‐0.18 [‐1.59, 1.23]

Comparison 29. Miswak mouthwash versus chlorhexidine mouthwash

Comparison 30. Chlorine dioxide mouthrinse versus chlorhexidine mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC (hydrogen sulphide) Show forest plot	1	22	Mean Difference (IV, Random, 95% CI)	‐11.0 [‐31.61, 9.61]

2 VSC (methyl mercaptan) Show forest plot	1	22	Mean Difference (IV, Random, 95% CI)	7.63 [‐1.70, 16.96]

3 VSC (methyl sulphide) Show forest plot	1	22	Mean Difference (IV, Random, 95% CI)	22.80 [‐33.18, 78.78]

Comparison 30. Chlorine dioxide mouthrinse versus chlorhexidine mouthwash

Comparison 31. Protease cysteine + actinidine versus placebo tablets

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	14	Mean Difference (Random, 95% CI)	‐45.8 [‐258.38, 166.78]

Comparison 31. Protease cysteine + actinidine versus placebo tablets

Comparison 32. Miswak stick versus chlorhexidine mouthwash

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	24	Mean Difference (IV, Random, 95% CI)	‐0.55 [‐1.33, 0.23]

2 VSC Show forest plot	1	24	Mean Difference (IV, Random, 95% CI)	‐0.77 [‐1.19, ‐0.35]

3 Patient self‐assessment score Show forest plot	1	24	Mean Difference (IV, Random, 95% CI)	‐0.26 [‐1.16, 0.64]

Comparison 32. Miswak stick versus chlorhexidine mouthwash

Comparison 33. Brushing + mouthwash versus brushing + tongue cleaning

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 VSC Show forest plot	1	30	Mean Difference (IV, Random, 95% CI)	‐81.87 [‐140.12, ‐23.62]

Comparison 33. Brushing + mouthwash versus brushing + tongue cleaning

Comparison 34. Brushing + cetylpyridium mouthwash versus brushing

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	70	Mean Difference (IV, Random, 95% CI)	‐0.48 [‐0.72, ‐0.24]

2 VSC Show forest plot	1	70	Mean Difference (IV, Random, 95% CI)	‐8.04 [‐15.87, ‐0.21]

Comparison 34. Brushing + cetylpyridium mouthwash versus brushing

Comparison 35. Turkish gall oral rinse versus brushing alone

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	66	Mean Difference (IV, Random, 95% CI)	‐0.10 [‐0.50, 0.30]

2 VSC Show forest plot	1	66	Mean Difference (IV, Random, 95% CI)	‐211.47 [‐503.58, 80.64]

Comparison 35. Turkish gall oral rinse versus brushing alone

Comparison 36. Laser + povidone iodine versus SRP

Outcome or subgroup title	No. of studies	No. of participants	Statistical method	Effect size
1 Dentist‐reported OLT score Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	0.49 [0.30, 0.68]

2 VSC Show forest plot	1	40	Mean Difference (IV, Random, 95% CI)	70.0 [63.88, 76.12]

Comparison 36. Laser + povidone iodine versus SRP