102 JOURNAL OF THE SOCIETY OF COSMETIC CHEMISTS We have previously shown that the instrumental method is capable of assessing the effect of both rinse and dentifrice treatments on mouth odor in humans (4). The purpose of this study was to determine the correlation between the instrumental method and sensory evaluation in a controlled clinical trial. MATERIALS AND METHODS A 220 Tracor gas chromatograph equipped with a flame photometric detector, housing a 394 millimicron filter specific for sulfur, was used to measure the sulfur compounds responsible for mouth odor. A 25-ft teflon (FEP) Supelco column was also used (con- taining 0.05% polyphenyl ether and 0.25% phosphoric acid on a 30/60 mesh- chromosorb T) to specifically separate the primary sulfur components of mouth air. Air was used as a carrier gas to eliminate the air peak and to reduce the analysis time. The isothermal oven temperature was 60øC, with inlet and detector temperatures of 130øC. A standard gas SO2 permeation tube was used to convert the results into nanograms/ milliliter for the total sum of sulfides. The mouth air sample was collected in a 24-ft teflon storage loop (x/s-in o.d. x x/x6-in i.d.) that was wrapped with teflon tape and had a Mininert teflon valve attached to each end (Figure 1). We have shown in previous studies that this loop is capable of storing mouth air without substantial sample loss. Each loop was calibrated to hold 10 cc of mouth air. When collecting the 10-cc sample, a 50-cc polyethylene syringe was attached to one end of the loop while a 3-in sterilized teflon tube was attached to the other end. Figure 1. Storage loop for mouth air samples.

MOUTH ODOR EVALUATION 103 The tube was carefully inserted into the panelist's mouth, both teflon valves were opened, and the mouth air was drawn into the storage loop using the syringe. During the evaluation procedure, panelists were asked to sit for 10 minutes with their mouths closed prior to each evaluation to allow for the buildup of the sulfur volatiles. The stored samples were transported to the laboratory in dry ice and analyzed within five to seven hours following collection. Two samples were obtained from each subject for analysis. The sample storage loop was attached to the gas chromatograph inlet gas sampling valve, where the stored mouth air sample was introduced into the GC. The gas sampling valve apparatus was also calibrated for 10 cc of mouth air. A 50-cc syringe was attached to the outlet portion of the valve on the GC in order to draw the sample into the sampling valve. When the valve was turned, the air carrier gas swept the sample into the column. The instrumental readings were expressed in nanograms per milliliter (ng/ml). The second method for the measurement of mouth odor is a subjective sensory evalu- ation. This evaluation was conducted by two professional judges from the La-Wall Harrison Research Laboratories immediately following the instrumental evaluations. Panelists waited for ten minutes with their mouths closed, similar to the instrumental procedure. A special booth was constructed that allowed the panelists to remain anon- ymous. A sterile thistle tube was inserted into the panelists' mouths while the bell- shaped end was inserted through the screen where the judge evaluated the mouth odor. Panelists were asked not to breath during this evaluation. Each of the two judges evaluated each panelist. The offensiveness of the odor was rated on a scale from 0 to 8 (Table I), where 0 is non-offensive and 8 is considered highly offensive. Two evaluations were conducted on each panelist, one evaluation by each judge. STUDY DESIGN Twenty panelists (ten men and ten women) were qualified to participate in the study by meeting the inclusion/exclusion characteristics of the protocol. The inclusion character- istics consisted of being of adult age, 20 years and above, having good general health, and having a total sulfur volatile baseline greater than 5 ng/ml of mouth air. This level Table I Sensory Evaluation Criteria for Mouth Odor Sensory score Offensiveness of mouth odor 8 7 Strong 6 Definite 5 4 Faint 3 2 Doubtful 1 0 None