JDR JDR Most Cited Articles
HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
QUICK SEARCH:   [advanced]


     

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tanaka, M.
Right arrow Articles by Shizukuishi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tanaka, M.
Right arrow Articles by Shizukuishi, S.
J Dent Res 83(4): 317-321, 2004
© 2004 International and American Associations for Dental Research

RESEARCH REPORT
Clinical

Clinical Assessment of Oral Malodor by the Electronic Nose System

M. Tanaka1,*, H. Anguri1, A. Nonaka1, K. Kataoka1, H. Nagata1, J. Kita2, and S. Shizukuishi1

1 Department of Preventive Dentistry, Graduate School of Dentistry, Osaka University, 1-8, Yamadaoka, Suita, Osaka 565-0871, Japan; and
2 Analytical Instruments Division, Shimadzu Corporation;

* corresponding author, tanakam{at}dent.osaka-u.ac.jp


   ABSTRACT
TOP
 ABSTRACT
INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
A recently developed electronic nose has not yet been clinically applied to evaluations of oral malodor. This investigation sought to determine whether an electronic nose could clinically assess oral malodor. Twenty-nine healthy adults and 49 patients were assessed by results of an actual organoleptic test, a score representing malodor strength with an electronic nose in "top-note" mode (top-note score), and measurements of volatile sulfur compound (VSC) concentrations. The correlation coefficient between top-note and actual organoleptic scores (r = 0.71) was comparable with the log VSC and actual organoleptic scores (r = 0.63). However, the area under the receiver-operating characteristic plots for top-note score was significantly larger than that for log VSC. In logistic regression analyses with top-note score as a dependent variable, probing depth, tongue coating, and plaque control record each had independent associations. Our findings suggest that the top-note score from an electronic nose examination may be useful for the clinical evaluation of oral malodor.

KEY WORDS: oral malodor • clinical assessment • electronic nose • oral health parameters


   INTRODUCTION
TOP
 ABSTRACT
 INTRODUCTION
MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
The two primary methods used for the clinical analysis of oral malodor are organoleptic measurement and instrumental analysis, which involves gas chromatography and a portable sulfide monitor (Rosenberg, 1996). The former is simple, since the examiner directly sniffs expelled mouth air and rates the malodor intensity on a scale from 0 to 5; however, training is required to minimize inter- and intra-examiner variability. In contrast, a gas chromatography unit has highly selective and sensitive flame photometric detection systems specific for volatile sulfur compounds (VSC), and can readily identify hydrogen sulfide (H2S), methylmercaptan (CH3SH), and dimethyl sulfide (CH3SCH3), as well as quantitatively determine VSC levels in breath (Oho et al., 2001; Furne et al., 2002). However, since this instrument is unable to detect other volatile compounds, there remains no ideal objective method by which the degree of oral malodor can be clinically assessed.

In recent years, the advent of chemical sensors and chemical sensor systems (so-called "electronic noses") has made possible fast and simple odor analyses for use in many fields (Mantini et al., 2000). This powerful technology has only recently been introduced in the field of medicine; however, its potential to assist in diagnosis is promising (Thaler et al., 2001). Although several researchers have proposed the application of electronic noses for medical purposes, very few studies have been conducted to investigate their possible use for the clinical assessment of oral malodor. The electronic nose system used in the present study was developed for quantitative measurement of malodor present in food and beverages (Kita et al., 2000), and we applied it clinically to assess oral malodor and to examine the association between oral malodor strength and oral health status.


   MATERIALS & METHODS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
RESULTS
 DISCUSSION
 REFERENCES
 
VSC Preparation as Standard Gases
Using a PD-1B permeation device (Gastec Co., Ltd., Kanagawa, Japan), we prepared H2S, CH3SH, and CH3SCH3 and added them separately to polypropylene bags containing 3000 mL of air (GL Science Inc., Tokyo, Japan). Final H2S, CH3SH, and CH3SCH3 concentrations were adjusted to approximately 0.05, 0.1, 0.5, and 1 ppm, respectively, while an additional mixture of 0.01 ppm of CH3SCH3 was also prepared. These gases were analyzed by means of gas chromatography and the electronic nose in top-note mode in triplicate examinations.

Subjects
Forty-nine patients (mean age, 38.8 yrs) who came to the Oral Malodor Clinic of Osaka University Dental Hospital complaining of oral malodor and with odor-judge ratings of >= 2.0 were selected, after informed consent was obtained. At the initial visit, the patients’ chief complaints, along with brief systemic and dental histories, were obtained, and preliminary oral examinations were conducted. During the second visit, extra-oral, intra-oral, and periodontal examinations were performed. At the third visit, breath malodor was assessed with the use of the electronic nose in top-note mode, VSC measurements were made by gas chromatography, and an organoleptic test was conducted. As control subjects, we recruited 29 healthy adults (mean age, 33.5 yrs), employed by the Osaka University Graduate School of Dentistry and with odor-judge ratings of < 2.0. All subjects were examined for oral health status and breath odor in a manner similar to that used for the patients. Approval for this study was obtained from the Ethical Committee for Clinical Research of Osaka University Graduate School of Dentistry.

Measurement with the Electronic Nose
The FF-1 odor discrimination analyzer (electronic nose) utilized in the present study was composed of a pre-concentrator, an array of 6 metal oxide semiconductor sensors selected for their different sensitivities and selectivities to fragrant substances, and pattern recognition software. The all-note measurement mode is the standard setting used to measure all volatile substances (Kita et al., 2000). However, since our preliminary experiment showed that the correlation coefficient between the top-note measurement and actual organoleptic test results showed the highest value (r = 0.71) of the 3 different measurement modes [all-note (r = 0.42), top-note, and deep-note (r = 0.41)], the top-note measurement mode, which primarily measures volatile substances with a low boiling point, was used in the present study. For that, the gas sample was introduced into the trap tube for 30 sec, during which the sample trapped in the tube was not dried or heated. The trapped odor was then driven to the sensor section with pure nitrogen. The sensor array represented the detected odor as a log [resistance (R)peak/Rbase], and the estimated VSC concentration and the estimated organoleptic score (top-note score) were calculated by the log Rpeak/Rbase according to a multiple linear regression method (Kita et al., 2000).

Statistical Analysis
We analyzed the data utilizing Stat View for Macintosh version 5.0 (SAS Institute Inc., Cary, NC, USA). Relationships among the 3 malodor measurements for each subject were assessed by means of a Spearman signed-rank test. Associations between clinical parameters and the 3 oral malodor measurements were evaluated by bivariate analyses with a Wilcoxon rank-sum test. We used multiple linear regression and logistic regression analyses to determine which variables demonstrated a significant independent effect on oral malodor measurements. Odds ratios and their 95% confidence intervals (CI) were also calculated. Statistical comparisons of top-note score and VSC level (Zweig and Campbell, 1993) were analyzed with the use of receiver-operating characteristic (ROC) plots with Rockit 0.9B Beta Version (1998). Significant differences between each 2 correlation coefficients were also analyzed (Zar, 1996).

The following methods can be found in the Web-only APPENDIX: Oral Gas Sampling, Measurement with Gas Chromatography, Organoleptic Assessment, and Periodontal and Dental Examinations.


   RESULTS
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
DISCUSSION
 REFERENCES
 
The numbers 1 and 2 sensors of the electronic nose yielded an exponential function-like response when exposed to a constant concentration of H2S, CH3SH, and CH3SCH3, whereas sensors 3 and 4 showed a weak response to each VSC (Fig. 1AGo). The concentrations of H2S, CH3SH, and CH3SCH3 estimated by the electronic nose in top-note mode were virtually identical to those found with the gas chromatograph (Fig. 1BGo).



View larger version (31K):
[in this window]
[in a new window]
  		Figure 1. The mean concentrations of H2S, CH3SH, and CH3SCH3 were measured by gas chromatography and the plots compared with the mean of each finding (log Rpeak/Rbase) from the 6 electronic sensors (• sensor 1, {blacksquare} sensor 2, {blacktriangleup} sensor 3, sensor 4, {square} sensor 5, {circ} sensor 6) (A), as well as mean VSC concentrations estimated with the electronic nose in top-note mode (B).

 
Correlations between oral malodor levels determined from organoleptic and electronic nose top-note scores, and log total VSC by gas chromatography, were examined. The correlation coefficient between top-note scores and actual organoleptic scores was 0.71 (Fig. 2AGo), which was nearly the same as that between log VSC and organoleptic scores (r = 0.63) (Fig. 2BGo). No significant difference was found between these two correlation coefficients (p = 0.366). ROC sensitivity-specificity plots provided an appropriate means for comparison of the top-note scores and log VSC data for classifying subjects with or without a >= 2.0 actual organoleptic score. The area under the ROC plots that represented the top-note scores (0.879) was significantly larger than that for log VSC (0.698) (p = 0.0052).



View larger version (12K):
[in this window]
[in a new window]
  		Figure 2. Correlations between actual organoleptic scores and top-note scores with an electronic nose (A) and log VSC by gas chromatography (B). ROC plots comparing top-note scores and log VSC data used for classifying subjects with or without an actual organoleptic score of >= 2.0 (C).

 
In bivariate analyses using age, gender, smoking, and 3 levels of oral health status as variables (Table 1Go), we found the top-note score to be significantly greater in subjects who showed higher values for age, smoking, probing depth, tongue-coating score, and plaque control record. Further, organoleptic score was significantly associated with age, pocket depth, tongue-coating score, and plaque control record. In contrast, total VSC level had a significant association with only probing depth and tongue-coating score.


View this table:
[in this window]
[in a new window]
  		Table 1. Associations between Clinical Parameters and Three Types of Malodor Measurements
 
The independent effects of those variables associated with top-note score levels were analyzed by multiple linear and logistic regression analyses (Table 2Go). In the linear regression analyses, the percentage of teeth with a 4 mm or greater probing depth, tongue-coating score, and plaque control record had a significantly independent effect on top-note score, with an explanatory power of 41%. Further, in logistic regression analyses of the possibility of subjects being in the upper 25th percentile of top-note score distribution as a dependent variable, gender, percentage of teeth with a 4 mm or greater probing depth, tongue-coating score, and plaque control record were each significantly associated with top-note score, with odds ratios of 22.7, 22.0, 9.8, and 12.5, respectively.


View this table:
[in this window]
[in a new window]
  		Table 2. Multiple Regression Analysis of Top-note Score Measured with Electronic Nose System as a Dependent Variable
 
The independent effects of top-note scores and log VSC, as well as other clinical parameters for actual organoleptic score as dependent variables, were tested by linear regression analysis. As a result, top-note score, plaque control record, log VSC, and age each had a significantly positive independent effect, with an explanatory power of 51%, 7%, 5%, and 3%, respectively (data not shown).


   DISCUSSION
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
REFERENCES
 
We found that at least 2 of the 6 sensors in the electronic nose were sensitive for the detection of VSC, which enabled the device to detect malodorous substances in a manner that was highly correlated with the concentration of each. Thus, with the present device, malodor may be able to be precisely measured as VSC levels in mouth air samples. Further, since the present electronic nose system can also detect such volatile compounds as organic compounds, aromatic compounds, amine-containing compounds, and ammonia derivatives in food and beverages (Kita et al., 2000), those sensors insensitive to VSC may be able to determine other volatile compounds in mouth air samples.

In the present study, we measured oral malodor strength using an electronic nose system in top-note measurement mode. In a comparison with all-note and deep-note measurement modes, the top-note score showed the highest correlation with organoleptic score, which suggested that the main compounds related to oral malodor are volatile substances with a low boiling point. There is the possibility that volatile components of mouthrinses or toothpastes could influence top-note vales, but this was not observed with the toothpaste in our preliminary experiment. A traditional problem with the use of electronic noses is the influence of water vapor. In this study, a water vapor trap was used that eliminated this problem. The current instrumentation has been used on a nearly daily basis for 4 yrs, and no evidence of any effects of water vapor on the sensors has been observed. Further, the correlation coefficient between the top-note and organoleptic scores (r = 0.71) was nearly the same as that between log VSC by gas chromatography and organoleptic score (r = 0.63), suggesting that the electronic nose system may be useful for the measurement of non-sulfur gaseous compounds as well as VSC. The area under the ROC curve representing top-note score was significantly larger than that for log VSC. This result suggests that the top-note score determined with the electronic nose had a higher level of accuracy than log VSC for classifying subjects with and without an actual organoleptic score of >= 2.0.

In the present subjects, bivariate analyses showed that age, smoking, probing depth, tongue-coating score, and plaque control record each had a significant association with top-note score. However, only probing depth and tongue-coating score were identified as having significant relationships to VSC level measured with gas chromatography, while age, probing depth, tongue coating, and plaque control record were each associated with actual organoleptic score. In linear regression analyses, the explanatory power for top-note score was 41%, while that for total VSC level was 15%, and for organoleptic score, 45% (data not shown). In addition, in a linear regression analysis with actual organoleptic score as a dependent variable, the explanatory power of top-note score was much higher (51%) than that of log VSC (7%), suggesting that top-note score may be more strongly associated with oral malodor than VSC by gas chromatography.

Associations between oral health parameters and malodor level have also been shown in previous studies, though with conflicting results. Using both sulfide measurements and organoleptic tests, Bosy et al.(1994) reported that oral malodor was not associated with periodontal probing depth, gingival index, or plaque index, whereas a significant correlation was observed between periodontal condition and oral malodor with both methods in other studies (Rosenberg et al., 1991; Kozlovsky et al., 1994; Morita and Wang, 2001). Further, tongue coating was found to have an association with oral malodor in organoleptic tests and portable sulfide monitor results (Miyazaki et al., 1995; Morita and Wang, 2001; Oho et al., 2001). Yaegaki and Sanada (1992) found a greater association of tongue coating with VSC formation in periodontitis subjects than in healthy individuals. Plaque accumulation has also been weakly associated with sulfide measurement and organoleptic test results in one report (Rosenberg et al., 1991), but with only organoleptic test results in another (Kozlovsky et al., 1994). Our present findings suggest that the effectiveness of the electronic nose can be best attributed to the cumulative effect of several factors, including the amount of supragingival plaque, the depth of periodontal pockets, and tongue coating.

We concluded that top-note score as determined with the use of an electronic nose can provide objective halitosis-related measurements, though those obtained by the multiple linear regression method showed only relative and not absolute values. Further investigations regarding the absolute expression of malodor intensity with an electronic nose will be conducted.


   ACKNOWLEDGMENTS
 
This work was supported by a Grant-in-Aid for Scientific Research from the Ministry of Education, Science, Sports, and Culture of Japan (B-15390650).


   FOOTNOTES
 
A supplemental appendix to this article is published electronically only at http://www.dentalresearch.org.

Received April 28, 2003; Last revision January 23, 2004; Accepted January 26, 2004


   REFERENCES
TOP
 ABSTRACT
 INTRODUCTION
 MATERIALS & METHODS
 RESULTS
 DISCUSSION
 REFERENCES
 
Bosy A, Kulkarni GV, Rosenberg M, McCulloch CA (1994). Relationship of oral malodor to periodontitis: evidence of independence in discrete subpopulations. J Periodontol 65:37–46.[ISI][Medline]

Furne J, Majerus G, Lenton P, Springfield J, Levitt DG, Levitt MD (2002). Comparison of volatile sulfur compound concentrations measured with a sulfide detector vs. gas chromatography. J Dent Res 81:140–143.[Abstract/Free Full Text]

Kita J, Aoyama Y, Kinoshita M, Nakano H, Akamaru H (2000). Quantification of the MOS sensor based electronic nose utilizing trap tube. In: Technical digest of the 17th Sensor Symposium, May 30, 2000. Tokyo: The Institute of Electrical Engineers of Japan, pp. 301–305.

Kozlovsky A, Gordon D, Gelernter I, Loesche WJ, Rosenberg M (1994). Correlation between the BANA test and oral malodor parameters. J Dent Res 73:1036–1042.[Abstract/Free Full Text]

Mantini A, Di Natale C, Macagnano A, Paolesse R, Finazzi-Agro A, D’Amoco A (2000). Biomedical application of an electronic nose. Crit Rev Biomed Eng 28:481–485.[ISI][Medline]

Miyazaki H, Sakao S, Katoh Y, Takehara T (1995). Correlation between volatile sulphur compounds and certain oral health measurements in the general population. J Periodontol 66:679–684.[ISI][Medline]

Morita M, Wang HL (2001). Relationship between sulcular sulfide level and oral malodor in subjects with periodontal disease. J Periodontol 72:79–84.[ISI][Medline]

Oho T, Yoshida Y, Shimazaki Y, Yamashita Y, Koga T (2001). Characteristics of patients complaining of halitosis and the usefulness of gas chromatography for diagnosing halitosis. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 91:531–534.[ISI][Medline]

Rockit 0.9B Beta Version (computer program) (1998). IBM-compatible version. Metz, EC, Department of Radiology, The University of Chicago.

Rosenberg M (1996). Clinical assessment of bad breath: current concepts. J Am Dent Assoc 127:475–482.[Abstract/Free Full Text]

Rosenberg M, Kulkarni GV, Bosy A, McCulloch CA (1991). Reproducibility and sensitivity of oral malodor measurements with a portable sulphide monitor. J Dent Res 70:1436–1440.[Abstract/Free Full Text]

Tanaka M, Anguri H, Nishida N, Ojima M, Nagata H, Shizukuishi S (2003). Reliability of clinical parameters for predicting the outcome of oral malodor treatment. J Dent Res 82:518–522.[Abstract/Free Full Text]

Thaler ER, Kennedy DW, Hanson CW (2001). Medical applications of electronic nose technology: review of current status. Am J Rhinol 15:291–295.[ISI][Medline]

Yaegaki K, Sanada K (1992). Volatile sulfur compounds in mouth air from clinically healthy subjects and patients with periodontal disease. J Periodontol Res 27:233–238.[ISI][Medline]

Zar JH (1996). Biostatistical analysis. 3rd ed. Upper Saddle River, NJ: Prentice-Hall, Inc., pp. 380–382.

Zweig MH, Campbell G (1993). Receiver-operating characteristic (ROC) plots: a fundamental evaluation tool in clinical medicine. Clin Chem 39:561–577.[Abstract/Free Full Text]




This article has been cited by other articles:


Home page
 Journal of the American Animal Hospital AssociationHome page
D. L. Eubanks
Canine Oral Malodor
J. Am. Anim. Hosp. Assoc., January 1, 2006; 42(1): 77 - 79.
[Abstract] [Full Text] [PDF]

This Article
Right arrow Abstract Freely available
Right arrow Figures Only
Right arrow Full Text (PDF)
Right arrow Appendix
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in ISI Web of Science
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Citing Articles
Right arrow Citing Articles via HighWire
Right arrow Citing Articles via ISI Web of Science (6)
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Tanaka, M.
Right arrow Articles by Shizukuishi, S.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Tanaka, M.
Right arrow Articles by Shizukuishi, S.

HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
IADR Journals Advances in Dental Research ®
Journal of Dental Research ® Critical Reviews (1990-2004)