Relationship Between Halitosis and Periodontitis: a Pilot Study

Abstract

Objective

Halitosis, or oral malodour, is an unpleasant smell emanating from the oral cavity. It is a common complaint among patients with periodontitis, however, their relationship is not fully elucidated. This study aimed to evaluate the association between halitosis measures, clinical indicators of periodontitis and tongue coating, as well as a novel measure, periodontal inflamed surface area (PISA).

Material and methods

Data of 10 patients with periodontitis and halitosis were included in this study. Halitosis was assessed by the organoleptic method and the portable sulphide monitor, measuring volatiles sulphur compounds. A comprehensive periodontal examination was conducted, and the parameters of probing depth, gingival recession, clinical attachment level, bleeding on probing, plaque and tongue coating were registered. The PISA was calculated using clinical attachment level, gingival recession and bleeding on probing.

Results

A correlation between organoleptic score and tongue coating (r=0.554) and plaque (r=0.614) could be observed. No correlation between measures of halitosis and probing depth or the PISA could be detected. A significant correlation was found between organoleptic scores and volatiles sulphur compounds values (r=0.931).

Conclusion

This pilot study has shown and further reiterated a complex interplay between different factors causative to halitosis in patients affected by periodontitis. The results suggest that tongue coating and oral hygiene may have an important role in halitosis in patients with periodontitis.

Introduction

Periodontitis is a chronic inflammatory disease of the tooth’s supportive tissues. The complex pathogenesis of the disease involves an inflammation triggered by the main causative factor, periodontal pathogenic bacteria (1). A variety of measures have been used in both clinical and research settings for diagnosis, evaluation of the treatment outcomes and long-term monitoring, needed due to the chronic nature of the disease, i.e. periodontal probing depth, gingival recession, clinical attachment level and indices of plaque and bleeding (2). Clinical parameters (indices) of bleeding in particular, have been employed to identify and assess active inflammatory sites (3). In 2008, Nesse et al. introduced a new measure, PISA (Periodontal Inflammed Surface Area), that reflects the surface of the bleeding epithelium of the periodontal pocket (4). The measure was initially developed as a tool that quantifies the inflamed periodontal area and, as such, assesses the inflammatory burden of periodontitis. The authors highlighted the value of PISA in the definition of periodontitis as a risk factor for other systemic diseases. This relatively new, emerging measure was hitherto correlated to the classification of periodontitis (5, 6) and HbA1c levels in healthy and diabetic patients (7, 8) and calculated for chronic kidney disease (9, 10) and rheumatoid arthritis patients (11).

Halitosis is defined as an unpleasant smell emanating from the oral cavity. Quirynen et al. reported on the etiology and characteristics of 2000 patients who visited a specialised, multidisciplinary halitosis clinic in Belgium (12). Almost 80% of halitosis cases were of intra-oral origin. Tongue coating was the most common cause, followed by gingivitis/periodontitis and a combination of the two, accounting, respectively, for 43%, 11% and 18% of the cases.
The olfactory malodourous compounds are primarily the endproducts of microbial degradation of both sulphur-containing (i.e. cysteine, cystine, methionine) and non-sulphur-containing amino acids (i.e. tryptophan, lysine, ornithine). These amino acids are mainly derived from proteins present in bacterial plaque, saliva, blood and tongue coating. Volatile compounds containing sulphur (i.e. volatile sulphur compounds - VSCs), hydrogen sulphide (H₂S), methyl mercaptan (CH₃SH) and dimethyl sulphide [(CH₃)₂S], are the major contributors to halitosis. Non-sulphur-containing volatile compounds, i.e. indole, skatole, amines and ammonia, contribute to halitosis to a lesser extent (13, 14).

Halitosis is a common complaint in the population of patients affected by periodontitis. Anaerobic gram-negative bacteria of the red complex, Porphyromonas gingivalis, Tanerella forsythia and Treponema denticola, strongly relate to clinical measures of periodontal disease and are also associated with higher production of VSCs (15, 16). However, the literature on the association between the depth of periodontal pockets and VSC production in patients with periodontitis reports conflicting data (17). Furthermore, there is a growing body of evidence that the activity of the disease, i.e. inflammation, expressed through measures of bleeding, is related to oral malodour (18–22). In patients with periodontitis, tongue coating can be another contributing factor to oral malodour since it has been reported that patients with periodontitis produce more tongue coating than healthy individuals (18).

Thus, this pilot study aims to evaluate the association between halitosis measures, clinical indicators of periodontitis and tongue coating. Furthermore, this study explores the correlation between halitosis parameters and a novel measure, periodontal inflamed surface area (PISA).

Material and methods

Study subjects
This pilot study is part of the doctoral dissertation study “Effect of probiotic Lactobacillus reuteri as an adjunct to nonsurgical treatment on halitosis of patients with periodontitis: a randomised placebo-controlled trial”, approved by the Ethics Committee of the School of Dental Medicine of the University of Zagreb (05-PA-15-11/2017). The present study used baseline clinical data obtained from 10 patients who participated in the study. 
Patients that were seeking or were referred for periodontal care to the Department of Periodontology of the School of Dental Medicine of the University of Zagreb were screened for possible inclusion in the study. Furthermore, a simple dichotomous organoleptic assessment (halitosis present / not present) was employed as a preliminary screening method for halitosis. A thorough medical and dental history was then taken, following the recommendations from an international workshop consensus (25) to exclude extraoral halitosis, causative of systemic diseases and/or medications. 
Inclusion criteria were: 1) Systemically healthy patients of both genders; 2) Non-smokers; 3) Presenting at least 20 teeth; 4) Untreated moderate to advanced generalised chronic periodontitis according to the 1999 Classification (23) / Generalised periodontitis stage II-IV according to the 2017 Classification (24); 4) Halitosis of intraoral origin, with a level of total sulphur compounds concentration - tVSC≥160 parts per billion (ppb) measured with a portable sulphide monitor (Halimeter®, Interscan Corporation Chatsworth, CA, USA). Exclusion criteria were 1) Pregnant and lactating females; 2) Antibiotics received for dental or non-dental purposes within 6 months before the start of the study; 3) Presence of systemic diseases or the use of medication that may affect periodontal tissues, the parameter of bleeding that can be the cause of extra-oral malodour; 4) Acute oral or periodontal inflammation or infection (i.e. pericoronitis, necrotising periodontal diseases, etc.) 
A comprehensive periodontal examination was performed by a single, calibrated periodontist (L.M.) The following parameters were measured and registered using a UNC-15 periodontal probe (Hu Friedy, Chicago, IL, USA): periodontal pocket depth (PPD), plaque (calculated as full-mouth plaque score – FMPS), bleeding on probing (calculated as full-mouth bleeding score – FMBS) and gingival recession (REC). A periodontal assessment of the parameters was done at 6 sites per tooth, excluding third molars due to the high prevalence of agenesis, absence due to extraction, malpositioning or infraocclusion, which may present difficulties in measuring periodontal parameters.

Halitosis assessment

Halitosis assessment was performed using an organoleptic method and a portable sulphide monitor, measuring tVSC, i.e. total concentration of the 3 sulphur compounds - hydrogen sulphide, methyl mercaptan and dimethyl sulphide (Halimeter®, Interscan Corporation Chatsworth, CA, USA). Both procedures were performed following recommendations from the literature (26) and the manufacturer’s instructions for the use of device (27). Patients were requested to refrain from eating spicy food, onion and garlic at least 24 hours before the assessment. Furthermore, they were asked not to use scented oral hygiene products on the morning of and have breakfast at least 2 hours before the measurement. All of the measurements were made in the morning, between 08:00 and 11:00 a.m.

The organoleptic assessment and scoring were done by a single clinician (L.M.) The participant was asked to keep the mouth closed for 2 min and breathe through the nose. The examiner positioned approximately 5 – 10 cm from the participant’s mouth, scored the breath with the mouth open while the patient was slowly exhaling the breath. Malodour was given an organoleptic score (OLS) from 0 to 5 as follows: 0: no appreciable odour; 1: barely noticeable odour; 2: slight, but clearly noticeable odour; 3: moderate odour; 4: strong odour; and 5: extremely foul odour, as described by Rosenberg et al. (28).

The measurement of the tVSC (total volatile sulfur compounds) concentration, expressed as parts per billion, was obtained using the portable sulphide monitor. A disposable tube was inserted approximately 3 – 4 cm into the participant’s mouth, slightly opened, without touching any of the oral surfaces. 3 consecutive measurements were taken, and the mean of the measurements was recorded (Figure 1).

Figure 1.

Use of Halimeter®, Interscan Corporation Chatsworth, CA, USA in the clinical setting

Tongue coating evaluation
The tongue coating (TC) assessment was done using the Winkel Tongue Coating Index (WTCI) (29). The dorsum of the tongue was divided into six areas, three in the anterior portion and three in the posterior portion, and each area was then assigned one of the three possible scores. No coating was scored as 0, light coating as 1 and heavy coating as 2, with a total tongue score ranging from 0 to 12.

PISA calculation
PISA was calculated using an available calculator within an Excel spreadsheet, prepared by the authors Nesse et al. (4) and uploaded on the website www.parsprototo.info for research use. The calculation was done in 7 steps, previously reported in detail by the authors. In short: 1) Mean clinical attachment level (CAL) was calculated for each tooth after data input; 2) Calculated mean CAL was translated into attachment loss surface area (ALSA) for each tooth with an appropriate formula; 3) Mean recession was calculated for each tooth after data input; 4) Calculated mean recession was translated into recession surface area (RSA) for each tooth with an appropriate formula; 5) RSA was subtracted from ALSA for each tooth, calculating periodontal epithelial surface area (PESA); 6) PESA of each tooth was multiplied by the proportion of positive BOP sites of that tooth, rendering periodontal inflamed surface area (PISA); 7) Total PISA was calculated through the sum of PISAs of each individual tooth.

Data analysis

The normality of distribution was evaluated using the Shapiro-Wilk test and by inspecting normal Q-Q plots. Due to identified departures from normality, data were presented as medians and interquartile ranges. A non-parametric correlation analysis (Spearman) was used to explore the relationship between variables PISA, FMBS, FMPS, PPD≥4mm, PPD≥6mm, tVSC, OLS, and TC. The statistical analysis was performed using SPSS (version 20, IBM, Armonk, NY, USA). Due to exploratory nature of this pilot study, the level of significance was set at α = 0.1 (30).

Results

General results

A total of 10 participants, 5 male and 5 female, were included in this pilot study. The median age was 38.5 (range 29 – 72). The median, minimum and maximum, and interquartile range of the measured parameters are listed in Table 1. The median PISA value was 1617.8. The halitosis measures median values, OLS and tVSC were 3.0 and 334.5, respectively. The median tongue coating value was 5.0.

Table 1. Median, minimum and maximum, and interquartile range of age, clinical periodontal measures, halitosis measures and tongue coating.

	Median	Minimum	Maximum	Interquartile range
Age (y)	38.5	29.0	72.0	12.0
PISA (mm2)	1617.8	1248.6	2950.7	934.4
FMBS (%)	78.0	61.9	100.0	25.4
FMPS (%)	82.9	75.0	100.0	17.6
PPD≥4mm (N)	57.0	41.0	111.0	53.0
PPD≥6mm (N)	14.5	0.0	63.0	17.0
tVSC (ppb)	334.5	184.0	462.0	176.0
OLS (N)	3.0	2.0	5.0	1.0
TC (N)	5.0	1.0	10.0	4.0
PISA – periodontal inflamed surface area; FMBS – full-mouth bleeding score; FMPS – full-mouth plaque score; PPD≥4mm – pocket probing depth ≥4mm; PPD≥6mm – pocket probing depth ≥6mm; tVSC – total volatile sulphur compounds, measured with Halimeter®; OLS – organoleptic score; TC – tongue coating

Correlations between different parameters

In terms of the association between the clinical and halitosis measures, significant correlations were found between the OLS and FMPS (r=0.614, p=0.059), as well as between the OLS and TC (r=0.554, p=0.097) (Table 2). A significant correlation was found between OLS and tVSC values, r=0.931, p<0.001 (Figure 2).
In addition, the PISA values showed a positive correlation with FMBS and FMPS indices, r=0.733, p=0.016 and r=0.839; p=0.002, respectively.

Table 2. Correlation between different parameters.

	PISA	FMBS	FMPS	PPD≥4mm	PPD≥6mm	tVSC	OLS	TC
PISA	-	0.733 (0.016)	0.839 (0.002)	0.572 (0.084)	0.697 (0.025)	-	-	-
FMBS	-	-	0.742 (0.014)	0.652 (0.041)	-	-	-	-
FMPS	-	-	-	0.599 (0.067)	-	-	0.614 (0.059)	-
PPD≥4mm	-	-	-	-	-	-	-	-
PPD≥6mm	-	-	-	-	-	-	-	-
tVSC	-	-	-	-	-	-	0.931 (0.000)	-
OLS	-	-	-	-	-	-	-	0.554 (0.097)
TC	-	-	-	-	-	-	-	-
Values: Spearman’s correlation coefficient - r (p-values) PISA – periodontal inflamed surface area; FMBS – full-mouth bleeding score; FMPS – full-mouth plaque score; PPD≥4mm – pocket probing depth ≥4mm; PPD≥6mm – pocket probing depth ≥6mm; tVSC – total volatile sulphur compounds, measured with Halimeter®; OLS – organoleptic score; TC – tongue coating

Figure 2.

Correlation between the organoleptic score and total volatile sulphur compounds

Discussion

The present pilot study explored the correlation between the parameters of halitosis and clinical indicators of periodontitis, tongue coating and, for the first time, the PISA.

Rizzo (31) and Tonzetich (32) were among the first to report a correlation between the production of sulfur compounds and periodontal disease, particularly the presence of periodontal pockets. While other research groups published similar findings, the correlation between the halitosis parameters and periodontal pocket depth was not always exclusive. A study by Liu et al., on a large population of 2000 participants, reported a significant correlation between the OLS and tVSC levels and pocket depth, however, tongue coating was highlighted as the most significant variable related to the oral malodour measures (19). Yaegaki & Sanada (18) showed the association between a higher concentration of methyl mercaptan in patients with pocket depths above 4 mm and bleeding on probing. The severity of periodontitis correlated with halitosis as reported by Söder et al. (33), as well as measures of oral hygiene, expressed through plaque and calculus indices. In our study, however, the number of pockets above 4 mm and 6mm, respectively, showed no significant correlation with halitosis measures.

It is important to emphasize the fact that not all evidence supports the existence of correlation between the number of periodontal pockets and depth with halitosis measures, thus questioning the causative relationship between periodontitis and halitosis. Bosy et al. (34) found a correlation between tongue coating and floss odour after insertion in the interdental area with tVSC and OLS values. However, no significant difference in halitosis measures could be observed among patients with periodontitis and healthy groups of participants. In the present study, an association between OLS and FMPS was also observed. While the indices measuring plaque, i.e. oral hygiene, are not the same as ours and the abovementioned study, it has been suggested in the literature that poor oral hygiene could be a source of halitosis of patients with periodontitis irrespective of the presence of periodontal pockets (35).

Calil et al. and De Boever & Loesche (36, 37) reported an association between halitosis and tongue coating measures, yet not with periodontal disease measures. 
As seen in the general population-based cross-sectional studies (12, 19, 20), tongue coating is suggested to be the primary cause of halitosis among periodontally healthy subjects. This is primarily attributed to the large surface of the tongue’s dorsum and its morphology (38), thus contributing to the accumulation of dead microbial and host cells, the source of halitosis. Furthermore, taking into account the evidence highlighted hitherto, it has been suggested that tongue coating is also an important contributing factor to halitosis in patients with periodontitis, since they produce more tongue coating than healthy individuals (39). They also harbour greater counts of the VSC-producing periodontal pathogen, Porphyromonas gingivalis, on the dorsum of the tongue (40). Our study showed a correlation between tongue coating and OLS values, but not tVSC. While sulphide monitors such as Halimeter® and OralChroma™ are an objective measuring of tVSCs and previous studies confirmed their clinical value, they fail to capture the contributions of non-sulfur compounds to the overall halitosis (14, 41, 42). An organoleptic assessment is considered the gold standard for halitosis detection because the human nose can evaluate a greater variety of malodourous compounds/odourants (41). According to the manufacturer of the Halimeter, patients are diagnosed with halitosis when the level of tVSCs exceeds 160 ppb, but as suggested by Vandekerckhove et al. (41) this cut-off should be lowered in order to improve the sensitivity of the instrument with regard to organoleptic testing. The range of OLS and tVSC values in our study ranged from 2 - 5 and 184.0 - 462.0 ppb, respectively, and a strong correlation was found between the two halitosis variables (r = 0.931). This is considered an important finding, considering that this pilot study was performed as a preliminary investigation for the upcoming comprehensive study “Effect of probiotic Lactobacillus reuteri as an adjunct to nonsurgical treatment on halitosis of patients with periodontitis: a randomised placebo-controlled trial”, in which halitosis measurements will be done by OLS assessment, and Halimeter® as an adjunct.

As indicated before, bleeding expressed through various bleeding and inflammation indices was found to be correlated to halitosis in several published studies (18, 19, 21, 43, 44). Bleeding is a sign of inflammation of the connective tissue of the gingiva. Thus, a variety of bleeding indices are employed to assess the presence of inflammation. In particular, when bleeding is associated with pocket probing, it indicates a site with active periodontal disease (3, 45). A possible explanation for the association of bleeding pockets and halitosis is the presence of periodontal pathogens, VSC-producing anaerobic bacteria at such sites (46). Indeed, as highlighted in a study by Torresyap et al., intra-pocket sulfide levels, measured with a specific Diamond Probe/Perio 2000 system, are higher in deeper pockets with higher counts of orange- and red-complex bacteria (47). Since PISA is a parameter that quantifies the surface of the bleeding epithelium of the pocket, one of the hypotheses of this study was that a correlation between this novel measure and halitosis might be observed. However, the present study did not confirm this hypothesis since no significant observations with OLS or tVSC values were found. Nonetheless, the PISA values were positively correlated to values of FMPS and FMBS, as previously reported by Park et al. (5).
Leira et al. and Nesse et al. (4, 6) correlated the PISA values and periodontal status, highlighting the increase in PISA values with disease severity, the former also suggesting a cut-off PISA value of ≥130.33mm² as a predictor for the presence of periodontitis. In our study, the values of PISA ranged from 1248.6 to 2950.7 mm². These values correspond to the values calculated for severe periodontitis (6).

The intrinsic limitations of this pilot study are the small sample size and the absence of a control group. Due to the sample of n=10 representing a specific population of patients, i.e. only highly inflamed patients with a progressed periodontitis forms and identifiable intraoral halitosis, the analysis might not have captured the correlation between all of the parameters. A sequel study with a larger sample size will be performed, including a control group of patients with periodontitis and no halitosis.

Conclusion

This pilot study shows and further reiterates a complex interplay between different factors causative to halitosis in patients affected by periodontitis. OLS values correlated to measures of oral hygiene and tongue coating, however, no correlation to probing depth could be observed.