Oral probiotic Weissella cibaria improves halitosis and reduces periodontal pathogens: a randomized, double-blind, placebo-controlled trial

Abstract

Background

Halitosis is closely linked to periodontal inflammation and affects oral health–related quality of life. This randomized trial investigated the effects of Weissella cibaria CMU (OraCMU) on halitosis in adults with gingivitis and early (Stage I–II) periodontitis.

Methods

Eighty adults were enrolled in a randomized, double-blind, placebo-controlled trial and assigned to OraCMU (n = 40) or placebo (n = 40). Participants consumed two slow-dissolving tablets daily for 8 weeks. The primary endpoint was change in total Volatile sulfur compounds (VSCs) from baseline to week 8. Secondary outcomes included organoleptic test (OLT), bad breath improvement (BBI), gingival index (GI), bleeding on probing (BOP), probing depth, plaque index, and oral health impact profile-14 (OHIP-14). Multiplex quantitative real-time PCR was performed at baseline and week 8 to quantify nine periodontopathogens and to derive the periodontal pathogen index (PPI) and PPI level.

Results

The OraCMU group showed a significantly greater reduction in total VSC (primary endpoint) at week 8 than the placebo group (P = 0.001). Secondary outcomes favored OraCMU for OLT (P = 0.015) and BBI (P = 0.045). Periodontal indices (GI and BOP) improved more in the OraCMU group at week 8 (P = 0.0035 and P = 0.030, respectively). qPCR showed lower levels of Porphyromonas gingivalis, Prevotella intermedia, and Treponema denticola in the OraCMU group at week 8 (P = 0.001, P = 0.046, and P = 0.005), along with lower PPI and PPI level (P = 0.005 and P = 0.002). OHIP-14 improved within the OraCMU group (P < 0.05) but did not differ between groups (P > 0.05).

Conclusions

Eight-weeks of W. cibaria CMU supplementation reduced major halitosis indices and improved periodontal pathogen levels in adults with gingivitis/incipient periodontitis. Longer-term studies, including post-intervention follow-up, are needed to assess the sustainability of the effects.

Trial registration

Clinical Research Information Service (CRIS), KCT0009810, registered 27 September 2024, retrospectively registered.

Introduction

Halitosis is a common oral condition that negatively affects social interactions and overall quality of life [1]. Its primary cause is the production of volatile sulfur compounds (VSC), mainly hydrogen sulfide (H₂S) and methyl mercaptan (CH₃SH), by anaerobic bacteria residing in periodontal pockets and on the dorsum of the tongue [2, 3]. These compounds result from the proteolytic metabolism of sulfur-containing amino acids and are strongly influenced by periodontal inflammation and tongue biofilm accumulation. Accordingly, chronic halitosis is closely associated with gingival inflammation and periodontal disease, reflecting dysbiotic changes in the oral microbiota [4].

Conventional approaches for halitosis management, including mechanical plaque control, tongue scraping, and the use of antiseptic mouth rinses, aim to reduce bacterial load and VSC production. However, these interventions do not consistently address the underlying imbalance of the oral microbial ecosystem [5, 6]. As a result, recurrence of malodor remains common, highlighting the need for strategies that promote sustained microbiological homeostasis.

In this context, oral probiotics have emerged as a promising, non-invasive approach for modulating the oral microbiota and mitigating halitosis [7]. Clinical studies have demonstrated that probiotics can reduce VSC levels and improve malodor-related outcomes, with a favorable safety profile [8]. Nevertheless, the magnitude and durability of these effects have varied across studies, and differences in probiotic strains, dosages, and study designs have limited the generalizability of existing evidence.

Weissella cibaria CMU (Chonnam National University College of Medicine) (OraCMU) is recognized as safe and has been widely used in commercial products. It exerts antibacterial activity against halitosis-associated pathogens, inhibits VSC production, and reduces biofilm formation in vitro [9, 10]. Two placebo-controlled randomized trials reported improvements in halitosis-related outcomes during 4–8 weeks of OraCMU tablet intake [11, 12]; however, effects on objective VSC measurements and organoleptic scores were not fully consistent across studies, and neither trial included post-intervention follow-up. Evidence in adults with concomitant gingivitis/early periodontitis and integrated microbiological profiling also remains limited.

The present study evaluated the effects of 8-week OraCMU supplementation on halitosis-related outcomes and periodontal status in adults with gingivitis/incipient periodontitis and halitosis. By integrating perceptual, biochemical, clinical periodontal indices, and microbiological measures, this randomized, double-blind, placebo-controlled trial aimed to provide more comprehensive evidence regarding the role of W. cibaria CMU as an adjunctive strategy for halitosis management.

Methods

This study was a prospective, randomized, double-blind, placebo-controlled, parallel-group clinical trial. The design, conduct, and reporting adhered to the Consolidated Standards of Reporting Trials (CONSORT) 2010 guidelines for randomized trials, with all methodological elements prepared in accordance with the recommended CONSORT checklist [13]. The study received ethical approval from the Institutional Review Board of Chosun University Dental Hospital (Approval No. CUDHIRB 2401 005) and was registered with the Clinical Research Information Service (CRIS, KCT0009810).

Study design

This 8-week, two-arm clinical trial was conducted at Chosun University Dental Hospital (Gwangju, South Korea) from March 2024 to January 2025. Investigator-assessed halitosis and clinical assessments were performed at baseline, week 4, and week 8. Participants’ bad breath improvement (BBI) scores were assessed at weeks 4 and 8. Periodontal pathogen analysis was performed at baseline and week 8.

Participants

Adults with halitosis and gingivitis/incipient periodontitis were recruited. Inclusion criteria were: age 19–70 years; ≥ 20 natural teeth; an organoleptic test (OLT) score ≥ 2; elevated volatile sulfur compounds (VSCs), defined as H₂S ≥ 1.5 ng/10 mL or CH₃SH ≥ 0.5 ng/10 mL; and a diagnosis of gingivitis or stage I-II periodontitis (BOP in ≥ 10% of sites; at least one tooth with PD > 3 mm and ≤ 5 mm) according to the 2018 AAP/EFP case definition [14]. Exclusion criteria included: periodontal scaling within 1 month; nonsurgical or surgical periodontal treatment within 6 months; ≥ 5 untreated dental caries; Stage III–IV periodontitis; antibiotic use within the previous month; clinically significant cardiovascular, immunological, infectious, or neoplastic disease; medications affecting periodontal status; excessive alcohol intake (> 20 g/day); and smoking. All participants provided written informed consent. Participants were recruited at Chosun University Dental Hospital through onsite postings and screening of volunteers presenting for oral care.

Sample size calculation

The sample size was based on prior study outcomes [15]. A total of 40 participants per group provided 80% power at α = 0.05, assuming a 25% dropout rate. Previous halitosis trials using OraCMU reported VSC reductions of 8–10 ng/10 mL versus placebo [11]. Using these estimates, power exceeded 0.80, supporting the adequacy of the current sample size.

Randomization and blinding

Randomization was performed using a computer-generated block randomization sequence prepared by an independent statistician. Allocation was concealed in sealed opaque envelopes and revealed only after completion of all analyses. Participants, investigators, and outcome assessors remained blinded to group assignment.

Intervention

Tablets (1 g each) were packaged in identical containers, labeled according to randomization codes, and supplied by Oraticx, Inc. (Seoul, South Korea). Participants consumed one tablet (OraCMU 2.0 × 10⁸ CFU or placebo) twice daily for 8 weeks, instructed to dissolve each tablet slowly (approximately 2 min) after toothbrushing and before bedtime to maximize probiotic retention. Tablets in both groups were identical in appearance, weight, and taste. Food, drink, and brushing were restricted for 30 min after tablet intake. Compliance was assessed using returned containers and participant diaries. The probiotic group received tablets containing OraCMU (2.0 × 10⁸ CFU), whereas the placebo group received isomalt-based tablets identical in appearance and taste. Compliance was calculated as intake ≥ 80% confirmed by returned container counts and diaries.

Volatile sulfur compounds (VSCs) measurement

For halitosis assessments, participants were scheduled to attend visits between 9:00 and 10:00 AM and were instructed to refrain from oral hygiene-related behaviors beforehand. VSC concentrations, including H₂S and CH₃SH, major contributors to halitosis (dimethyl sulfide being primarily extraoral), were quantified via Oral Chroma CHM-2 (Fis Inc., Tokyo, Japan) [16]. Participants avoided odor-affecting foods the day prior, attended morning visits fasting, and refrained from brushing and drinking.

Organoleptic test (OLT)

OLT was conducted by a single calibrated and blinded examiner (Lee S.K.), who had completed standardized training based on the International Society for Breath Odor Research guidelines prior to study initiation [17]. Scores ranged from 0 (no odor) to 5 (severe odor).

Bad breath improvement (BBI)

Participants self-evaluated halitosis improvements on a 1-to-5 scale, with lower scores indicating greater improvement [18].

Clinical parameters

Clinical parameters were assessed by a calibrated examiner blinded to group allocation. Clinical parameters were measured at six sites (mesial, central, and distal on the buccal and lingual sides) on six predefined index teeth (#16, 12, 24, 32, 36, and 44). Gingival index (GI) was assessed using a modified scoring system (0–4) on the buccal and lingual sides [19]. Bleeding on probing (BOP) was scored as positive if bleeding was observed within 30 s of probing. Probing depth (PD) was measured as the distance from the gingival margin to the gingival sulcus/pocket floor. Plaque index (PI) was assessed on a scale of 0–5 [20]. Individual mean GI, PD, and PI were calculated by averaging the scores of all examined teeth. BOP% was calculated as the ratio of the number of BOP-positive sites to the total number of examined sites.

Periodontal pathogen index (PPI)

Oral microbial risk associated with periodontitis was assessed using the periodontal pathogen index (PPI) as described by Kim et al. [21]. For sample collection, participants gargled 12 mL of a standardized mouthwash solution (provided by Helixco Inc., Ulsan, South Korea) for more than 30 s. Saliva samples were analyzed via multiplex quantitative real-time polymerase chain reaction (qPCR) targeting nine key periodontopathogens (Aggregatibacter actinomycetemcomitans, Campylobacter rectus, Eikenella corrodens, Fusobacterium nucleatum, Peptostreptococcus anaerobius, Porphyromonas gingivalis, Prevotella intermedia, Tannerella forsythia, and Treponema denticola). Microbiological sampling was performed at baseline and week 8 only (no week 4 sampling was planned or performed). PPI scores were calculated based on the relative abundance of these species and categorized into four risk levels: 1–40 = healthy, 41–60 = moderate, 61–80 = risky, and 81–100 = severe. For categorical analysis, continuous PPI scores were converted into a simplified 4-point PPI level: healthy = 1, caution = 2, risk = 3, and serious = 4.

Oral health impact profile (OHIP-14)

OHIP-14 was used to assess oral health-related quality of life and consists of 14 items, with two questions for each of seven domains: functional limitation, physical pain, physical disability, psychological discomfort, psychological disability, social disability, and handicap. Each item is scored on a 5-point Likert scale (0 = very often; 1 = quite often; 2 = sometimes; 3 = almost never; 4 = not at all), with higher scores indicating better oral health-related quality of life [22].

Safety assessments

Safety was assessed throughout the study. Adverse events were defined as any untoward medical occurrence after enrollment, regardless of causality. At each visit, participants were asked about symptoms and events since the prior visit, and diaries were reviewed. Events were evaluated by the investigator for severity and relatedness to the intervention.

Statistical analysis

Statistical analysis was performed using SPSS (version 29.0; IBM Corp., Armonk, NY, USA). All tests were two-sided with a significance level of 0.05. Normality was assessed using the Shapiro–Wilk test. For between-group comparisons, a two-sample t-test was used for normally distributed variables; otherwise, a Wilcoxon rank-sum test was applied. Within-group comparisons were evaluated with a paired-samples t-test or Wilcoxon signed-rank test, as appropriate. Categorical variables were analyzed using the chi-square test or Fisher’s exact test. Between-group comparisons of changes in efficacy variables were assessed using analysis of covariance (ANCOVA), adjusting for baseline values. To address multiplicity concerns, the primary endpoint was pre-specified as change in VSC; all other outcomes were treated as secondary/exploratory and interpreted accordingly.

Results

Participant flow and baseline

Of 83 screened participants, 80 were randomized (OraCMU n = 40; placebo n = 40). Three placebo participants withdrew, resulting in 77 completing the study (Fig. 1). No statistically significant differences were observed between groups in baseline characteristics. Compliance was > 91% with no group differences (Table 1). No adverse events occurred.

Fig. 1.

CONSORT flow diagram of participant progression through the clinical trial. A total of 83 participants were assessed for eligibility, and 80 were randomly allocated to oral probiotics (OraCMU; n = 40) or placebo (n = 40). During the study period, three participants in the placebo group withdrew due to antibiotic use. Data from 77 participants were included in the final analysis. OLT, organoleptic test; VSC, volatile sulfur compounds; PPI, periodontal pathogen index; OHIP-14, Oral health impact profile-14; BBI, Bad breath improvement

Table 1.

Baseline demographic and oral health characteristics of the analyzed subjects

Variables	OraCMU n = 40	Placebo n = 37	P-valuea
Gender, n (%)
Male	10 (25.00)	7 (18.90)	0.520 (C)
Female	30 (75.00)	30 (81.10)
Age, years (mean ± SD)	63.63 ± 4.95	64.84 ± 4.69	0.225 (W)
Drinking status†, n (%)
Yes	10 (25.00)	8 (21.60)	0.726 (C)
No	30 (75.00)	29 (78.40)
Frequency of tooth brushing, time/day (mean ± SD)	2.70 ± 0.56	2.84 ± 0.60	0.320 (W)
Use of oral hygiene aids§, n (%)
Yes	37 (92.50)	33 (89.20)	0.705 (F)
No	3 (7.50)	4 (10.80)
Total VSC, ng/10 mL mean ± SD; median (min―max)	16.50 ± 21.23; 9.29 (0.54―84.09)	9.79 ± 12.01; 5.18 (0.50―51.44)	0.113 (W)
OLT, scores mean ± SD; median (min―max)	3.08 ± 0.89; 3.00 (2.00―5.00)	2.89 ± 0.84; 3.00 (2.00―4.00)	0.402 (W)
GI, scores mean ± SD; median (min―max)	0.48 ± 0.31; 0.42 (0.08―1.25)	0.40 ± 0.27; 0.33 (0.08―1.00)	0.282 (W)
BOP, % mean ± SD; median (min―max)	34.93 ± 15.99; 33.33 (8.33―77.78)	32.58 ± 15.86; 30.56 (11.11―83.33)	0.431 (W)
PD, mm mean ± SD; median (min―max)	2.55 ± 0.37; 2.51 (1.94―3.81)	2.56 ± 0.34; 2.53 (1.81―3.42)	0.814 (W)
PI, scores mean ± SD; median (min―max)	0.92 ± 0.55; 0.96 (0.17―2.00)	0.98 ± 0.38; 1.00 (0.08―2.00)	0.440 (W)
Compliance, % (mean ± SD)
Week 4	91.18 ± 9.96	94.33 ± 7.15	0.142 (W)
Week 8	91.13 ± 12.30	92.25 ± 10.40	0.629 (W)
Average	91.16 ± 10.21	93.28 ± 7.77	0.402 (W)

OLT organoleptic test, VSC volatile sulfur compounds, GI gingival index, BOP bleeding on probing, PD probing depth, PI plaque index

^†Drinking status indicates current alcohol consumption (yes/no) based on self-report

^§Oral hygiene aids refer to adjunctive devices such as dental floss, interdental brushes, or mouth rinses in addition to toothbrushing

^aChi-square test (C) or Fisher’s exact test (F) was used for categorical variables. Continuous variables were compared using a two-sample t-test (T) when normally distributed; otherwise, a Wilcoxon rank-sum test (W) was used. The letter in parentheses indicates the test applied

Halitosis-related outcomes

As summarized in Table 2 and Fig. 2, the OraCMU group demonstrated significantly greater improvements in halitosis parameters than the placebo group during the 8-week intervention. Regarding total VSC, the OraCMU group exhibited marked reductions at both week 4 (P = 0.001) and week 8, with the week-8 change (− 9.96 ± 1.60) being significantly greater than that of the placebo group (− 1.96 ± 1.67; P = 0.001). OLT scores showed a significant reduction in the OraCMU group at week 4 (P = 0.000), accompanied by a significant between-group difference (P = 0.003). This improvement persisted through week 8, where a significant difference remained (P = 0.015). For BBI, participant-reported scores showed significant improvement in the OraCMU group at week 8 (P = 0.046), with a corresponding significant between-group difference (P = 0.045).

Table 2.

Changes in halitosis-related variables between the oral probiotic and the placebo groups before and after the intervention

Variables	Time point	OraCMU n = 40	Placebo n = 37	P-valueb	P-valuec
H2S (ng/10 mL)	Baseline	7.96 ± 11.50; 4.23 (0.00―51.41)	4.22 ± 5.79; 2.13 (0.00―21.14)	0.059 (W)
	Week 4 P-valuea	4.08 ± 6.51; 1.97 (0.00―36.47) 0.005 (S)	4.84 ± 6.73; 2.25 (0.00―34.87) 0.315 (S)	0.432 (W)
	Week 8 P-valuea	2.51 ± 5.35; 1.29 (0.00―33.99) 0.000(S)	4.47 ± 5.99; 2.03 (0.00―26.53) 0.987(S)	0.090 (W)	0.003
CH3SH (ng/10 mL)	Baseline	8.54 ± 11.58; 4.73 (0.24―48.99)	5.57 ± 6.95; 3.72 (0.01―36.52)	0.335(W)
	Week 4 P-valuea	3.28 ± 4.29; 1.74 (0.13―18.62) 0.000 (S)	6.10 ± 9.45; 3.23 (0.07―48.67) 0.541 (S)	0.112(W)
	Week 8 P-valuea	1.87 ± 2.84; 1.21 (0.00―16.63) 0.000 (S)	5.71 ± 8.60; 1.96 (0.15―38.65) 0.428 (S)	0.025(W)	0.001
Total VSC (ng/10 mL)	Baseline	16.50 ± 21.23; 9.29 (0.54―84.09)	9.79 ± 12.01; 5.18 (0.50―51.44)	0.113 (W)
	Week 4 P-valuea	7.37 ± 10.01; 4.39 (0.23―55.09) 0.001 (S)	10.94 ± 15.73; 5.73 (0.07―83.54) 0.898 (S)	0.180 (W)
	Week 8 P-valuea	4.37 ± 8.04; 2.71 (0.06―50.62) 0.000 (S)	10.18 ± 14.33; 4.40 (0.22―60.18) 0.624 (S)	0.033 (W)	0.001
OLT (scores)	Baseline	3.08 ± 0.89; 3.00 (2.00―5.00)	2.89 ± 0.84; 3.00 (2.00―4.00)	0.402 (W)
	Week 4 P-valuea	2.23 ± 0.97; 2.00 (1.00―4.00) 0.000 (S)	2.70 ± 0.81; 3.00 (1.00―4.00) 0.167 (S)	0.020 (W)
	Week 8 P-valuea	1.65 ± 0.83; 2.00 (0.00―3.00) 0.000 (S)	2.03 ± 0.99; 2.00 (1.00―4.00) 0.000 (S)	0.131 (W)	0.015
BBI (scores)	Week 4	2.43 ± 0.59; 2.00 (1.00―3.00)	2.27 ± 0.69; 2.00 (1.00―3.00)	0.353 (W)
	Week 8 P-valuea	2.18 ± 0.55; 2.00 (1.00―3.00) 0.046 (S)	2.41 ± 0.64; 2.00 (1.00―4.00) 0.317 (S)	0.096 (W)	0.045

Values are presented as mean ± SD; median (min―max)

H₂S hydrogen sulfide, CH₃SH methyl mercaptan, VSC volatile sulfur compounds (sum of H₂S and CH₃SH), OLT organoleptic test, BBI bad breath improvement

^aCompared within-group; P-value for paired t-test when normally distributed; otherwise, a Wilcoxon signed-rank test (S) was used. The letter in parentheses indicates the test applied

^bCompared between groups; P-value for two-sample t-test when normally distributed; otherwise, a Wilcoxon rank-sum test (W) was used. The letter in parentheses indicates the test applied

^cCompared between groups; P-value for ANCOVA adjusted baseline

Fig. 2.

Changes from baseline in halitosis-related parameters after 4 and 8 weeks of supplementation with oral probiotics (OraCMU) or placebo. A hydrogen sulfide (H₂S); B methyl mercaptan (CH₃SH); C total volatile sulfur compounds (VSCs) measured with Oral Chroma; D Investigator-assessed organoleptic test (OLT) scores; and (E) participant-reported bad breath improvement (BBI) scores. Each value is presented as the least squares mean (LSmean) ± standard error (SE). Between-group comparisons of changes from baseline were conducted using analysis of covariance (ANCOVA), with baseline values as covariates. An asterisk (*) denotes a statistically significant within-group difference, and a hash (#) denotes a statistically significant between-group difference (P < 0.05)

Clinical parameters outcomes

Clinical parameters (GI, BOP, PD, and PI) are presented in Table 3. The OraCMU group showed a greater reduction in GI from baseline to week 8 compared with placebo (between-group P = 0.035 for change), and BOP also decreased more in the OraCMU group at week 8 (between-group P = 0.030 for change). There were no significant differences in PD and PI between groups.

Table 3.

Change in clinical parameters by measurement time

Clinical parameters	Time point	OraCMU group n = 40	Placebo group n = 37	P-valueb	P-valuec
GI (scores)	Baseline	0.48 ± 0.31; 0.42 (0.08―1.25)	0.40 ± 0.27; 0.33 (0.08―1.00)	0.282 (W)
	Week 4 P-valuea	0.48 ± 0.54; 0.33 (0.00―2.69) 0.074 (S)	0.39 ± 0.27; 0.33 (0.08―1.50) 0.660 (P)	0.781 (W)
	Week 8 P-valuea	0.27 ± 0.23; 0.17 (0.08―1.00) 0.000 (P)	0.34 ± 0.26; 0.25 (0.00―1.33) 0.524 (S)	0.072 (W)	0.035
BOP (%)	Baseline	34.93 ± 15.99; 33.33 (8.33―77.78)	32.58 ± 15.86; 30.56 (11.11―83.33)	0.431 (W)
	Week 4 P-valuea	29.72 ± 15.41; 27.78 (0.00―75.00) 0.070 (S)	30.93 ± 16.33; 27.78 (0.00―83.33) 0.337 (P)	0.798 (W)
	Week 8 P-valuea	26.73 ± 13.61; 23.61 (2.78―69.44) 0.000 (P)	30.26 ± 13.35; 30.56 (11.11―72.22) 0.235 (P)	0.139 (W)	0.030
PD (mm)	Baseline	2.55 ± 0.37; 2.51 (1.94―3.81)	2.56 ± 0.34; 2.53 (1.81―3.42)	0.814 (W)
	Week 4 P-valuea	2.54 ± 0.29; 2.50 (2.03―3.44) 0.780 (P)	2.62 ± 0.31; 2.53 (2.03―3.53) 0.160 (P)	0.423 (W)
	Week 8 P-valuea	2.52 ± 0.35; 2.47 (1.94―3.61) 0.456 (P)	2.59 ± 0.30; 2.53 (2.19―3.50) 0.606 (P)	0.426 (W)	0.299
PI (scores)	Baseline	0.92 ± 0.55; 0.96 (0.17―2.00)	0.98 ± 0.38; 1.00 (0.08―2.00)	0.440 (W)
	Week 4 P-valuea	0.89 ± 0.55; 0.79 (0.08―2.42) 0.577 (P)	0.91 ± 0.50; 1.00 (0.00―1.92) 0.265 (P)	0.866 (T)
	Week 8 P-valuea	0.76 ± 0.54; 0.67 (0.00―2.08) 0.028 (P)	0.84 ± 0.59; 0.67 (0.00―2.17) 0.043 (S)	0.560 (W)	0.722

Values are presented as mean ± SD; median (min―max)

GI gingival index, BOP bleeding on probing, PD probing depth, PI plaque index

^aCompared within-group; P-value for paired t-test (P) when normally distributed; otherwise, a Wilcoxon signed-rank test (S) was used. The letter in parentheses indicates the test applied

^bCompared between groups; P-value for two-sample t-test (T) when normally distributed; otherwise, a Wilcoxon rank-sum test (W) was used. The letter in parentheses indicates the test applied

^cCompared between groups; P-value for ANCOVA adjusted baseline

Periodontopathogen outcomes

Among the nine target species, F. nucleatum, P. anaerobiusm, P. gingivalis, and P. intermedia showed significant decreases in the OraCMU group at week 8 compared to baseline, while P. gingivalis, P. intermedia, and T. denticola showed significant decreases at week 8 (F. nucleatum, P = 0.034; P. anaerobius, P = 0.020; P. gingivalis, P = 0.023; P. intermedia, P = 0.002). In contrast, E. corrodens, P. gingivalis, and T. denticola showed significant increases at week 8 compared to baseline in the placebo group (E. corrodens, P = 0.032; P. gingivalis, P = 0.043; T. denticola, P = 0.001). P. gingivalis (P = 0.001), P. intermedia (P = 0.046), and T. denticola (P = 0.005) showed significant reductions in the OraCMU group compared to the placebo group at week 8. PPI and PPI level significantly decreased in the OraCMU group but not in the placebo group. Between-group differences were statistically significant at week 8 (PPI, P = 0.003; PPI level, P = 0.002) (Table 4).

Table 4.

Quantitative qPCR outputs for the nine target periodontopathogens and periodontal pathogen index

Clinical parameters	Time point	OraCMU group n = 40	Placebo group n = 37	P-valueb	P-valuec
Aggregatibacter actinomycetemcomitans	Baseline	363,734.13 ±  1,734,506.01;1.00 (1.00―10,568,175)	31,128.24 ±  188,293.29;1.00 (1.00―1,145,513)	0.879 (W)
	Week 8 P-valuea	142,632.6  ± 856,174.36;1.00 (1.00―5,420,009) 0.877 (S)	83,329.78  ± 505,676.12;1.00 (1.00―3,076,097) 0.272 (S)	0.081 (W)	0.693
Campylobacter rectus	Baseline	122,205.48  ± 139,431.57; 69,903.5 (2,588―583,445)	104,249.16  ± 115,600.86; 74,989 (2,218―532,108)	0.632 (W)
	Week 8 P-valuea	140,719.53  ± 198,380.5; 59,597.5 (1,169―868,960) 0.819 (S)	185,894.92  ± 229,998.16; 91,201 (1,538―1,137,627) 0.158 (S)	0.814 (W)	0.242
Eikenella corrodens	Baseline	92,826.58  ± 132,997.42; 53,053 (2,339―781,628)	63,646.46  ± 85,767.24; 28,973 (2,291―370,681)	0.119 (W)
	Week 8 P-valuea	105,110.30  ± 109,630.61; 88,819.5 (1,384―535,797) 0.197 (S)	155,567.84  ± 196,994.87; 88,716 (1,531―1,039,920) 0.032 (S)	0.466 (W)	0.121
Fusobacterium nucleatum	Baseline	1,980,326.65  ± 1,751,684.65; 1,361,737 (121,339―6,934,258)	1,587,691.35  ± 1,521,835.39; 1,174,898 (1.00―5,662,393)	0.223 (W)
	Week 8 P-valuea	1,404,001.18  ± 1,926,081.64; 826,671 (35,727―10,209,395) 0.034 (S)	2,121,474.54  ± 2,576,554.7; 1,044,720 (40,179―13,551,894) 0.338 (S)	0.093 (W)	0.059
Peptostreptococcus anaerobius	Baseline	690,831.03  ± 1,151,725.09; 156,210.5 (1,026―4,666,594)	649,240.49  ± 1,097,665.59; 224,388 (1.00―4,943,107)	0.767 (W)
	Week 8 P-valuea	390,808.70  ± 1,105,077.19; 68,280.5 (1.00―6,854,882) 0.020 (S)	1,102,210.50  ± 2,273,060.51; 475,335 (3,715―13,061,709) 0.219 (S)	0.009 (W)	0.052
Porphyromonas gingivalis	Baseline	343,126.1  ± 459,407.54; 198,419 (49.00―1,972,423)	297,908.32  ± 363,946.74; 197,242 (1.00―1,402,814)	0.764 (W)
	Week 8 P-valuea	201,965.25  ± 35,4993.58; 90,704.5 (37.00―2,018,945) 0.023 (S)	630,728.59  ± 913,629.79; 295,121 (1.00―3,899,420) 0.043 (S)	0.016 (W)	0.001
Prevotella intermedia	Baseline	295,254.98  ± 1,165,020.8; 1,239 (1.00―7,227,698)	229,342.62  ± 601,629.7; 24,717 (1.00―3,365,116)	0.440 (W)
	Week 8 P-valuea	134,964.98  ± 711,779.18;1.00 (1.00―4,487,454) 0.002 (S)	248,388.81  ± 589,301.56; 31,623 (1.00―2,779,713) 0.610 (S)	0.560 (W)	0.046
Tannerella forsythia	Baseline	11,570.50  ± 23,055.39; 2,311.5 (1.00―113,501)	10,456.76  ± 15,974.24; 2,460 (1.00―57,148)	0.680 (W)
	Week 8 P-valuea	11,326.15  ± 24,785.65; 2,229.5 (1.00―144,877) 0.607 (S)	25,517.00  ± 49,840.03; 2,667 (1.00―212,814) 0.354 (S)	0.537 (W)	0.069
Treponema denticola	Baseline	14,588.58  ± 45,926.07;1,921 (1.00―280,543)	18,248.27  ± 42,739.3; 2,404 (1.00―247,742)	0.713 (W)
	Week 8 P-valuea	6,280.58  ± 13,701.54; 216.5 (1.00―54,954) 0.185 (S)	27,240.03  ± 44,135.38; 8,318 (1.00―229,615) 0.001 (S)	0.003 (W)	0.005
PPI	Baseline	75.59 ± 19.90; 74.85 (20.80―100.00)	74.74 ± 24.54; 79.30 (21.50―100.00)	0.411(W)
	Week 8 P-valuea	62.65 ± 24.20; 61.70 (15.30―100.00) 0.033 (S)	76.52 ± 24.52; 80.00 (4.60―100.00) 0.758 (S)	0.007 (W)	0.005
PPI level	Baseline	3.15 ± 0.89; 3.00 (1.00―4.00)	3.11 ± 1.05; 3.00 (1.00―4.00)	0.930 (W)
	Week 8 P-valuea	2.63 ± 1.13; 3.00 (1.00―4.00) 0.002 (S)	3.19 ± 0.94; 3.00 (1.00―4.00) 0.499 (S)	0.025 (W)	0.002

Values are presented as mean ± SD; median (min―max)

PPI periodontal pathogen index, PPI level periodontal pathogen index level

^aCompared within-group; P-value for paired t-test when normally distributed; otherwise, a Wilcoxon signed-rank test (S) was used. The letter in parentheses indicates the test applied

^bCompared between groups; P-value for two-sample t-test when normally distributed; otherwise, a Wilcoxon rank-sum test (W) was used. The letter in parentheses indicates the test applied

^cCompared between groups; P-value for ANCOVA adjusted baseline

Oral health-related quality of life

Pre- and post-intervention differences between the OraCMU and placebo groups on the OHIP-14 are shown in Fig. 3. After the 8-week intervention, the total score difference for all OHIP-14 items increased statistically significantly in the OraCMU group to 3.43 ± 1.00 (P = 0.002), but between-group differences were not significant (P = 0.733). Five items, including functional limitations, physical pain, physical disability, psychological discomfort, and psychological disability, also showed statistically significant increases in the oral probiotic group after the 8-week intervention compared to the pre-intervention level (P < 0.05). However, no statistically significant differences were observed between the two groups (P > 0.05).

Fig. 3.

Changes from baseline in oral health-related quality of life after 4 and 8 weeks of oral probiotics (OraCMU) or placebo supplementation. A Oral health impact profile (OHIP)-14 total scores; B functional limitation; C physical pain; D physical disability; E psychological discomfort; F psychological disability; G social disability; H handicap. Each value represents the least squares mean (LSmean) ± standard error (SE). Between-group comparisons of changes from baseline were performed using analysis of covariance (ANCOVA), with baseline values as the covariate. No significant differences were found between groups (p > 0.05). (*) indicates a statistically significant difference within a group (P < 0.05)

Discussion

Halitosis is predominantly of intraoral origin and is closely linked to periodontal inflammation and anaerobic biofilms [2, 23]. VSCs, particularly H₂S and CH₃SH, are generated by proteolytic metabolism of sulfur-containing substrates by Gram-negative anaerobes, and CH₃SH is often elevated in periodontitis [2, 23]. Accordingly, strategies that reduce pathogenic biofilms and VSC production remain central to halitosis management.

The World Health Organization defines probiotics as “live microorganisms which, when administered in adequate amounts, confer a health benefit on the host [24].” To date, most clinical studies on the efficacy of oral probiotics for halitosis management have demonstrated only modest effects, and the underlying mechanisms remain incompletely elucidated [8].

OraCMU has the ability to reduce pathogenic biofilms [9] and has been suggested as a potential benefit in the management of halitosis in two previous randomized, double-blind, placebo-controlled clinical trials (RCTs), but the available evidence was limited and heterogeneous. In both previous studies, a daily intake of 1 × 10⁸ CFU for 8 weeks showed a significant decrease in BBI compared to the placebo group, but no significant decrease in OLT and VSC changes [11, 12].

While universally accepted minimum clinically important differences for VSC concentrations are not established, changes on the 0–5 organoleptic scale are generally considered perceptible to patients and clinicians. Furthermore, BBI is a subjective, but individual-perceived, measure of halitosis improvement, making it an important variable for assessing halitosis improvement.

This RCT is the first to comprehensively evaluate the efficacy of OraCMU in reducing halitosis as well as clinical indices in adults with early-stage periodontal inflammation, corresponding to stage I-II periodontitis. In this study, the daily intake of OraCMU was evaluated to be 4 × 10⁸ CFU. After 8 weeks, the OraCMU group showed additional improvements in several halitosis-related indices compared to the placebo group, including investigator-assessed halitosis score (OLT), VSC levels measured by Oral Chroma, and participant-reported halitosis index (BBI). In addition, significant improvements were observed in GI and BOP, which are represented gingival health evaluation indices. This suggests that the observed effects are likely to be clinically noticeable, although confirmation in longer-term studies is needed.

Given the well-established association between periodontitis and halitosis, various methods for the quantitative analysis and risk assessment of periodontal pathogens have been developed. One such method is the PPI, which is derived from multiplex qPCR targeting nine high- and intermediate-risk periodontal species, including A. actinomycetemcomitans, C. rectus, E. corrodens, F. nucleatum, P. anaerobius, P. gingivalis, P. intermedia, T. forsythia, and T. denticola. This grading system has been validated using mouthwash samples from Korean adults [21].

In the present study, the PPI and PPI level both significantly decreased in the OraCMU group after 8 weeks, with statistically significant differences compared to the placebo group. Notably, PPI level scores improved from the “risk” to the “caution” category, reflecting a clinically favorable shift in microbial composition toward reduced pathogenic burden.

Several oral probiotics, including commercially available strains such as Streptococcus salivarius, Limosilactobacillus reuteri, Lactobacillus brevis, and Ligilactobacillus salivarius, have been reported to reduce VSC levels, periodontal pathogens, and clinical parameters [25–29]. However, few have demonstrated both halitosis reduction and direct pathogen decreases within the same trial. In this study, halitosis-related variables consistently improved following OraCMU intervention. Importantly, the significant reduction in periodontal pathogen levels, as measured by the PPI, reinforces the hypothesis that OraCMU supplementation not only alleviates halitosis but also promotes a healthier oral microbial environment. Although improvements in OHIP-14 were observed, no significant between-group difference was found; thus, these should be interpreted as exploratory.

The observed efficacy of OraCMU in this study is supported by several previously proposed mechanisms, as illustrated in Fig. 4. First, OraCMU produces antimicrobial substances such as hydrogen peroxide, organic acids (e.g., lactic, acetic, and citric acids), and N-acetylmuramidase, which exhibit broad-spectrum activity against Gram-negative anaerobes [30]. Second, it may inhibit biofilm formation by coaggregating with harmful bacteria [31, 32]. Third, it has been reported to inhibit bacterial adhesion [33] and persist in the oral cavity in prior studies [11, 12]; however, strain retention and colonization were not directly assessed in the current trial. Finally, OraCMU suppresses mgl gene expression and methionine γ-lyase (METase) activity, thereby reducing VSC production from sulfur-containing amino acids such as methionine and cysteine [34].

Fig. 4.

Proposed mechanisms of halitosis inhibition by Weissella cibaria CMU (OraCMU). OraCMU exerts multiple actions to inhibit oral malodor by targeting Gram-negative anaerobic bacteria responsible for producing volatile sulfur compounds (VSCs) such as hydrogen sulfide (H₂S) and methyl mercaptan (CH₃SH). (1) It shows antibacterial activity against major oral pathogens including Porphyromonas gingivalis, Treponema denticola, Prevotella intermedia, Fusobacterium nucleatum, and Peptostreptococcus anaerobius via the production of organic acids (lactic, acetic, and citric), hydrogen peroxide (H₂O₂), and antibacterial enzymes (e.g., N-acetylmuramidase) [30]; (2) it modulates biofilm formation through coaggregation with oral bacteria [31, 32]; (3) it competitively inhibits pathogen attachment to oral epithelial surfaces [33]; (4) it colonizes the oral cavity, contributing to ecological balance [11, 12]; (5) furthermore, it suppresses mgl gene expression and methionine γ-lyase (METase) activity, reducing VSC production from sulfur-containing amino acids such as methionine and cysteine [34]. Only clinical outcomes and qPCR-based pathogen measures were assessed in this trial; the other mechanisms are proposed based on prior studies

Although probiotics may not act as rapidly as antibiotics or chemical agents, they are linked to fewer side effects and a lower risk of resistance. However, their efficacy depends on factors such as strain stability, dosage, and delivery method, and long-term continuous use is required [35, 36]. This study differs from prior trials by including gingivitis participants and integrating microbiological analyses. The findings align with previous research [25, 26, 28] while expanding evidence in older adults.

Despite these promising results, this study has some limitations. First, although statistically powered for the primary endpoint, the trial was relatively small and conducted at a single center, which may limit generalizability. Second, microbiological profiling was limited to targeted qPCR and did not include broader microbiome-wide analyses (e.g., 16S rRNA sequencing). Third, no post-intervention washout/follow-up period was included; therefore, durability of the clinical and microbiological effects and persistence/retention of the probiotic strain could not be evaluated. Finally, self-reported assessments for subjective halitosis and oral health-related quality of life introduce the potential for reporting bias. Future studies with larger and more diverse populations, longer intervention durations, and post-intervention follow-up are warranted to confirm and expand upon these findings.

Conclusions

Supplementation with W. cibaria CMU (OraCMU) for 8 weeks reduced gingival and halitosis indices and improved periodontal pathogen counts in adults with stage I–II periodontitis. These results suggest that OraCMU may be a useful adjuvant probiotic for halitosis management by modulating the oral microbiota. Larger, longer-term follow-up studies are needed to determine the sustainability of the effects and their clinical applicability.

Abbreviations

BBI

Bad breath improvement

CFU

Colony-forming units

H₂S

Hydrogen sulfide

CH₃SH

Methyl mercaptan

OHIP-14

Oral health impact profile-14

OLT

Organoleptic test

OraCMU

Weissella cibaria CMU

PPI

Periodontal pathogen index

VSC

Volatile sulfur compounds

Authors’ contributions

Seung-Kyu Lee: Investigation; Formal analysis; Data curation; Writing – original draft; Visualization. Mi-Sun Kang: Resources; Funding acquisition. Kyeong-Ok Lim: Investigation; Data curation; Writing – review and editing. Keon-Il Yang: Writing – review and editing. Sang-Joun You: Writing – review and editing. Won-Pyo Lee: Conceptualization; Methodology; Validation; Supervision; Project administration; Writing – review and editing.

Funding

This study was funded by the Health Functional Food Development Support Program (No. RS-2022–00167206), supported by the Ministry of Small and Medium Enterprises (SMEs) and Startups (MSS, Republic of Korea).

Data availability

The datasets generated and analyzed during the current study are not publicly available due to ethical restrictions and the need to protect participant privacy. However, the data underlying the findings of this study are available from the corresponding author upon reasonable request, subject to approval by the institutional review board.

Declarations

Ethics approval and consent to participate

This study was conducted at Chosun University Dental Hospital and approved by the Institutional Review Board of Chosun University Dental Hospital (Approval No. CUDHIRB 2401 005). This clinical trial was registered at the Clinical Research Information Service (CRIS), Republic of Korea (KCT0009810). All procedures involving human participants were performed in accordance with the ethical standards of the institutional and national research committee, the Declaration of Helsinki, and local statutory requirements. Written informed consent was obtained from all participants prior to enrollment, and all participants provided consent for the publication of anonymized data.

Consent for publication

Not applicable.

Competing interests

Mi-Sun Kang is an employee of Oraticx Inc. and served as the principal investigator for the government-funded project that supported this study. Her involvement in this research was limited to funding acquisition and the provision of study resources, and she had no role in participant recruitment, randomization, data collection, data analysis, data interpretation, or manuscript preparation. All other authors declare no competing interests.