website: 86th General Session & Exhibition of the IADR

ABSTRACT: 3359  

Prediction of halitosis-treatment outcome using malodor intensities by electronic nose

M. TANAKA1, M. NODA1, M. TOE1, J. KITA2, and S. SHIZUKUISHI3, 1Osaka University, Suita, Japan, 2Shimadzu Corporation, Kyoto, Japan, 3Osaka University Graduate School of Dentistry, Japan

Objectives: In the previous study, we classified halitosis patients by cluster analysis using malodor-intensity values of seven category-gases in breath air as measured with electronic nose (EN). The purpose of this investigation was to determine whether this classification is reliable in the development of an accurate prognosis for oral malodor treatment.

Methods: In sixty-three patients who complained of oral malodor, forty-seven patients with ³ 0.25ppm of total volatile sulfur compound (VSC) level using gas chromatograph and ³ 2 of organoleptic test (OLT) score were treated in a similar manner. Improved treatment outcome was defined as <0.25ppm of total VSC level and <2 of OLT score 8 months after baseline. At both the initial and final visit, oral malodor was assessed by the malodor-intensity values of seven gases using EN, OLT and VSC level using gas chromatograph. Oral health status was evaluated as probing pocket depth, tongue coating score and plaque index. Additionally, proportions of six anaerobes in saliva were assessed by real-time PCR method. The halitosis patients were classified into four clusters using malodor-intensity values.

Results: All subjects in cluster 1 and 2 and the subjects except one in cluster 3 showed ³ 0.25ppm VSC level with gas chromatograph and ³ 2 of OLT score, but no subject in cluster 4 revealed oral malodor. Twenty-six patients demonstrated improvement due to the treatment. In bivariate analysis, cluster 1, probing pocket depth, tongue coating score and proportion of total six anaerobes at baseline each displayed significant positive association with “not improved outcome” due to treatment. Furthermore, in multiple logistic regression analysis of “not improved outcome” as a dependent variable, only cluster 1 showed independent association and its odds ratio was 35.1.

Conclusion: The classification of halitosis using malodor intensities measured with EN may be effective for prediction of outcome of malodor treatment.

Back to Top