1.

Introduction

Oral malodor is primarily caused by bacteria, such as Gram-negative anaerobic and proteolytic obligate species found in the oral cavity, and their degradation of sulfur-containing amino acids into volatile sulfur compounds (VSCs) results in the observed clinical symptoms.^1–3 It is well-known that the oral cavity provides favorable conditions for the growth of bacteria responsible for oral malodor.^4–6 In particular, the dorsum of the tongue and the interdental areas serve as recipient sites of anaerobic bacteria, which commonly produce VSCs and short-chain organic acids and act as major sites of oral malodor generation.^3,7

The plaque that accumulates in these sites is associated with the oral malodor and the status of plaque influences the severity of the malodor and also acts as an indicator of the oral hygiene status. A strong correlation has been reported between the tongue-coating status and oral malodor,^3,8,9 which highlights the need for a method that can objectively evaluate the tongue-coating status to aid the diagnosis and assessment of oral malodor.¹⁰ Given that the interdental plaque has been shown to consist of oral anaerobic organisms, especially the gingival sulcus and periodontal pockets where plaque is present produce high concentrations of VSCs and are correlated with oral malodor.¹¹ Therefore, bacterial biofilms formed on the tongue and interdental region including the subgingival area influence the production of oral malodor as major etiological factors.^12,13

Various methods for quantifying biofilm level have been proposed for evaluations of oral malodor. These methods have been largely based on visual inspection, bacterial counts, and wet-weight measurements of the biofilm.^14–16 Although visual assessments are predominantly used in clinical applications, they lack reproducibility and objectivity. It has been shown that the opacity of a biofilm can vary with the species constituting the biofilm, and it is difficult to quantify qualitative characteristics related to the pathogenicity of oral malodor.¹⁷

Quantitative light-induced fluorescence (QLF) technology has been used for detecting dental plaque as red fluorescence (RF) that originates from specific bacterial metabolites formed in an oral biofilm, such as endogenous metal-free fluorescent porphyrin,^18,19 when irradiated with 405 nm of visible blue-light. This optical phenomenon of plaque can be explained since porphyrin compounds produced by oral microorganisms show strong fluorescence in the red spectral region when excited with violet-light ranging from 400 to 420 nm.^20,21 The concentration of the porphyrin compounds is reported to be high in Gram-negative oral bacteria, which increases as the dental biofilm becomes more mature.^21–23 Previous studies have found that mature biofilm exhibits RF stronger than initial biofilm, and the RF intensity increases with the biofilm maturation and its cariogenicity.^21,23–25 From recent studies, it has been confirmed that a plaque assessment method based on autofluorescence of oral biofilms can be used to quantify dental plaque and tongue plaque by assessing their RF intensity detected by QLF systems²⁶ and that tongue plaque produces RF that shows a higher-than-moderate correlation with the level of oral malodor.²⁷ If QLF technology can be used to detect bacterial biofilms that mainly comprise anaerobes producing malodorous compounds and to quantify their accumulation status comprehensively, this will make it possible to evaluate oral malodor more precisely than when assessing only a single-bacterial factor.

Therefore, this clinical study examined the fluorescence of tongue and interdental plaque, which are the predominant causes of oral malodor, using QLF technology and assessed the ability of the combined plaque fluorescence (CPF) score to quantify the level of oral malodor.

2.

Materials and Methods

2.5.

Fluorescence Image Acquisition

Plaque fluorescence images were obtained using QLF-D (QLF-D Biluminator™, Inspektor Research Systems, the Netherlands) and proprietary software (C2 version 1.0.0.7, Inspektor Research Systems) for the camera settings. The QLF-D Biluminator™, a device using QLF technology, uses a digital single-lens reflex (SLR) camera with blue and white LED lights with a peak wavelength of $405\pm 20\mathrm{nm}$ (violet) and modified filter set (D007; Inspektor Research Systems, the Netherlands). This device captured the normal white-light images and fluorescence images of the tongue and floss with a “Live-View”-enabled SLR camera (model 550D, Canon, Tokyo, Japan) using the following settings: shutter speed of $1/45\mathrm{s}$, aperture value of 3.2, and ISO 1600 (Fig. 1).

Fig. 1

Example images of tongue plaque (a), (b) and interdental plaque (c), (d) captured by QLF-D using white light (a), (b) and blue light (b), (d).

To ensure that the distance between the tongue and light source remained constant while capturing the whole tongue area, a subsidiary cylindrical ring was equipped with the camera tube, which enables the examiner to easily position a patient’s face and tongue. The subject was instructed to extend his or her tongue out, and the examiner manually extended the tongue as much as possible to both minimize its movements during image acquisition and expose a maximum area of the tongue.³² The examiner fixed the end of the cylindrical ring on the patient’s face and took the tongue images.

The interdental plaque was collected by flossing the area between the first and the second molar in each of the four quadrants. The sampling was omitted if any of the molars were missing from each quadrant. The floss was passed through the contact area and flossed below the gingival margin as deeply as possible, and supra- and subgingival plaque samples were collected simultaneously. The examiner flossed each side of the molar twice in the same way, with a total of four flossing procedures being performed. When removing the floss from the interdental area, the end part of the floss was passed by the contact point to avoid influencing the collected plaque on the middle part of the floss. All sampling procedures were performed by a single-trained examiner (Eun-Song Lee). Disposable floss (WE DEN, Kimpo, Korea) designed for use with a handle and with a constant working length was used. As soon as the collection of plaque samples was completed, normal white-light images and sequential fluorescence images of the interdental plaque were obtained using QLF-D using the same settings used to obtain the tongue images.

3.

Results

About 99 patients included in the present study ranged in age from 19 to 66 years ($36.3\pm 13.1$ years). The organoleptic evaluations revealed that 83.8% of the patients had definite oral malodor ($\mathrm{OLS}=2$ to 5). The OLS score and VSC levels were significantly higher in the obvious-malodor group than in the no-malodor group ($p<0.0001$, Table 1). About $\sim 21\%$ of the subjects had received periodontal treatment within the previous 3 months. About $\sim 39\%$ of the subjects with definite oral malodor had symptoms of gingival bleeding, whereas this was present in $\sim 19\%$ of the subjects without oral malodor (Table 1). The presence of oral malodor was not significantly associated with periodontal treatment, gingival bleeding, or the symptoms of xerostomia ($p>0.05$).

Table 1

Characteristics and clinical parameters of the study population (n=99).

The fluorescence variables and scores of each plaque for the three severity groups classified according to the OLS are listed in Table 2. The CPF score increased with the malodor level and significantly differed among the groups ($p<0.0001$). The fluorescence intensity and integrated score for tongue plaque also differed significantly between the groups ($p<0.0001$). Similarly, there was a significant difference between the strong–severe level group and the other groups for the fluorescence area and the integrated score of the interdental plaque ($p<0.0001$).

Table 2

Median values of fluorescence variables for each type of plaque according to the severity of oral malodor.

$R/G$ values represent the ratios of red to green pixels in RF images of the tongue captured by the QLF-D.

Different letters within the same column indicate significant differences between malodor groups by Bonferroni’s correction for multiple analysis at $\alpha =0.05$.

Also, all plaque fluorescence scores showed significant positive correlations with the OLS and total VSCs level (Table 3, Fig. 2). Among them, the CPF score displayed the strongest correlations with OLS ($r=0.64$, $p<0.01$) and the total VSCs level ($r=0.54$, $p<0.01$). The TPF score was correlated more strongly with the malodor level ($r=0.51$ to 0.55) than with the IPF score ($r=0.38$ to 0.47). Also, the significant correlations were observed between the CPF scores and ${\mathrm{H}}_2\mathrm{S}$ ($r=0.59$, $p<0.01$) and ${\mathrm{CH}}_3\mathrm{SH}$ ($r=0.31$, $p<0.01$) concentrations, while there was no significant correlations between the fluorescence scores and ${({\mathrm{CH}}_3)}_2\mathrm{S}$ concentrations ($r=0.01$, $p=0.42$, Table 3).

Table 3

Correlation coefficients between oral malodor and fluorescence variables of the plaques.

Fig. 2

Scatter plots of the correlations between the OLS and CPF score.

The sensitivity, specificity, and AUC values for the CPF scores regarding the OLS to assess the different severities of oral malodor are presented in Table 4. The diagnostic accuracy of CPF scores for oral malodor was moderate to high ($\mathrm{AUC}=0.77$ to 0.94) and increased with the threshold level. When using a CPF score of 155.7 for detecting subjects with and without definite oral malodor (OLS 0-1/2-5), the sensitivity was 90.7% and the specificity was 62.9%.

Table 4

AUC and optimum sensitivity, specificity, and cutoff values for the CPF score at each diagnostic threshold quantified using the OLS.

4.

Discussion

The main finding of the present study was that oral malodor could be assessed by quantifying fluorescence of oral biofilms, which are considered to be major etiological factors for oral malodor. The two bacterial biofilms targeted in this study comprise VSCs-producing anaerobic bacteria mainly associated with periodontal diseases,^11,12 which means that it is valid and reliable to evaluate the characteristics of these biofilms when assessing malodor.

In terms of optical phenomenon of plaque autofluorescence, RF emission observed on QLF technology is proposed to be the result of excitation of endogenous porphyrins by the violet-blue light at a 405-nm wavelength.^19,33,34 The porphyrin concentration is high in Gram-negative oral bacteria, which indicates the observed RF was produced by mature bacterial biofilms, and its intensity increased with the biofilm maturation and its pathogenicity.^19,23 These previous findings support the results of the present study that the RF of two biofilms was significantly correlated with the severity of oral malodor. Given the associations and the diagnostic accuracy of the CPF score, the combined approach of plaque fluorescence could be used to evaluate the severity of oral malodor with increased accuracy. In addition, the present study confirmed the potential of an assessment method in objectively determining oral hygiene status as well as malodor levels, which is meaningful for both motivating patients to perform plaque control themselves and for clinical applications aimed at reducing oral malodor.

The results obtained in the present study demonstrate that the newly developed CPF score was more strongly correlated with oral malodor than each individual plaque score (Table 3). This is consistent with previous findings suggesting that the sites that most commonly produce oral malodor—the tongue dorsum, interdental area, and gingival sulcus—are also the main anatomical sources of VSCs.⁵ These areas may constitute specific environments that are relatively free from saliva flushing and have low oxygen levels, thereby favoring more anaerobic species and their bacterial biofilms containing high proportions of anaerobic Gram-negative bacteria.^35,36 Moreover, an association of plaque fluorescence with oral malodor is also supported by previous findings that the anaerobic species within the biofilms can actively produce malodorous VSCs and short-chain organic acids via metabolic interactions.¹³ For these reasons, the assessment of bacterial samples obtained from the tongue and interdental sites has been used to diagnose oral malodor, such as using the benzoyl-DL-arginine-naphthylamide (BANA) test quantifying bacterial activity and the polymerase chain reaction. In addition, it is well known that oral malodor is induced by interactions between multiple species of bacteria, which support the measurement of fluorescence resulting from the metabolic activity of all bacteria for assessing oral malodor.

This is the first study to have visualized and quantified the amount of interdental plaque using an objective detection method. To assess the pathogenicity of interdental plaque, which is difficult to detect, it is imperative to collect plaque samples while preserving their mechanical and biochemical properties. Therefore, the interdental plaque in individual patients was sampled by flossing using the same method with standardized floss, and the accumulation status of the interdental plaque on the floss was quantified based on the fluorescence intensity and area. As the tendency for RF varied between the participants, it showed variations in the fluorescence intensity and distribution patterns (Fig. 1), we have proposed the use of an individual fluorescence score obtained by multiplying two fluorescence properties. In terms of the interdental plaque being a predictor for gingival inflammation and the periodontal status in an individual, the findings of the present study suggest that the method using QLF technology would be useful for early detection of plaque-related disease.

In the present study, the IPF score was significantly correlated with the OLS and the VSC levels, especially of ${\mathrm{CH}}_3\mathrm{SH}$ and ${\mathrm{H}}_2\mathrm{S}$ (Table 3), which indicates that evaluating the amount of interdental plaque could be useful for diagnosing oral malodor. Emphasizing the role of interdental plaque as a status indicator for oral hygiene and gingival health, the present findings are consistent with previous reports of oral malodor being primarily associated with the tongue coating and gingival inflammation.³⁷ Previous researchers found that the interdental plaque and subgingival plaque of patients with oral malodor contain significantly higher proportions of anaerobic bacteria that produce VSCs, and that patients with periodontal pocket and inflamed gingiva showed significantly higher concentrations of VSCs.^13,38,39 Moreover, there was a strong relationship between plaque-induced periodontal disease and the severity of oral malodor, since periodontal pathogens such as Porphyromonas gingivalis, Prevotella intermedia, Tannerella denticola, and Veillonella alcalescens in the subgingival area can produce malodorous compounds. This further suggests that assessment of interdental plaque can be used to evaluate oral malodor as a useful predictor of the periodontal status of an individual patient.

It is presumed that the two biofilms have similar bacterial compositions and, in particular, share the presence of red-fluorescencing bacteria, confirming the significant correlation between the two plaque fluorescence scores in this study. Based on previous studies, the microbes identified in the tongue plaque are almost the same as those found in the subgingival plaque,^{11,12,40–42} and the tongue-coating volume is significantly correlated with the percentage area of bleeding-on-probing sites and the severity of periodontal disease.^37,41 We speculate that there is a relationship between the microflora of the tongue plaque and of the interdental plaque, where the accumulation of anaerobic bacteria in the biofilms can affect the production of oral malodor.

The present study has revealed the potential of QLF technology for detecting bacterial biofilms in diagnosing oral malodor by objectively assessing the comprehensive properties of pathogenic biofilms that are closely involved in the generation of oral malodor. The clinical use of QLF technology may make it possible for patients to check their own oral hygiene status and provide a quantitative numerical value that accurately reflects the severity of oral malodor. However, some improvements are necessary to the convenience and accuracy when applying this method in clinical applications. For example, analysis software based on algorithms for calculating the combined fluorescence score found in this study needs to be developed, and additional equipment is needed to adequately fix the tongue position when capturing the QLF images.

5.

Conclusions

It can be concluded that the red autofluorescence from tongue and interdental plaque detected using QLF technology was significantly correlated with the level of oral malodor. Therefore, the plaque fluorescence score which represents comprehensive fluorescence properties of oral biofilms could be used to quantify the level of oral malodor and to aid its diagnosis.