HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS		QUICK SEARCH:   [advanced] Author: Keyword(s): Year:  Vol:  Page:

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (19) Citing Articles via Google Scholar Google Scholar Articles by Greenman, J. Articles by El-Maaytah, M. Search for Related Content PubMed PubMed Citation Articles by Greenman, J. Articles by El-Maaytah, M.

Study on the Organoleptic Intensity Scale for Measuring Oral Malodor

¹ Centre for Research in Biomedicine, Faculty of Applied Sciences, University of the West of England, Coldharbour Lane, Bristol, BS16 1QY, UK;
² Maurice and Gabriela Goldschleger School of Dental Medicine, Tel Aviv University, Ramat Aviv, Tel Aviv 69978, Israel;
³ University of Minnesota, 515 Delaware St., SE, Minneapolis, MN 55455;
⁴ University Health Resources Group, 5714, Canterbury Drive, Culver City, CA 90230; and
⁵ School of Dentistry, Jordan University, Amman, Jordan;

^* corresponding author, john.greenman{at}uwe.ac.uk

	ABSTRACT

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
The 0–5 organoleptic scale is used widely in breath research and in trials to measure the efficacy of anti-odor agents. However, the precise relationship between odor scores and gas concentrations of target odorants is unknown. The purpose of this study was to relate mean organoleptic scores from odor judges (n = 7) for pure odorants (n = 8) representative of those found in oral malodor. Judges used a common 0–5 scale to report the odor intensity of sample sets in random order of concentration. Regression analysis of data showed that odor score was proportional to the log concentration of odorant, and comparison of slopes showed H₂S to be the most significant in terms of odor power. Detection thresholds (mol.dm^-3) were: Skatole (7.2 x 10^-13) < methylmercaptan (1.0 x 10^-11) < trimethylamine (1.8 x 10^-11) < isovalerate (1.8 x 10^-11) < butyrate (2.3 x 10^-10) < hydrogen sulphide (6.4 x 10^-10) < putrescine (9.1 x 10^-10) < dimethyl disulphide (5.9 x 10^-8). The study demonstrates the exponential nature of the olfactory response and shows that any single compound’s contribution to malodor depends on odor power and threshold in addition to concentration.

KEY WORDS: oral malodor • organoleptic intensity scale • volatile compounds (VCs) • volatile sulphur componds (VSCs) • odor judges

	INTRODUCTION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Oral malodor (bad breath) affects most people at some time in their lives and in some cases can be both persistent and strong (Scully et al., 1994), causing considerable psychological stress in a minority of subjects (Yaegaki et al., 1996; Rosenberg, 2002). Bad breath is one of the most common complaints patients report to their dental practitioners, and treatment usually consists of recommendations to improve oral hygiene and advice on the use of proprietary OTC treatments such as mouthwashes, toothpastes, lozenges, oral sprays, and films, designed to combat this condition. It is often difficult to assess just how effective many of these treatments are, since few have been subjected to thorough critical examination in well-designed clinical trials.

Measurements of effectiveness rely on well-designed clinical trials where the strength of bad breath is assessed before and at various times following treatment or placebo in volunteer groups large enough to show statistically significant differences following treatment. Both longitudinal and crossover designs have been used. Although instruments including gas chromatography (Tonzetich, 1971; Yaegaki and Sanada, 1992) and sulphide monitors (Rosenberg et al., 1991a,b; Shimura et al., 1996) have been used for measuring bad breath, the majority of breath studies use trained breath judges to measure breath odor subjectively.

Malodor judges recognize two types of dimensions to odors: quality and strength. Methods to estimate the quality of bad breath—so-called "hedonic" methods—score on the basis of how pleasant or unpleasant the odor is (ASTM, 1968). Although the hedonic scale may be useful for measuring the effects of compounds that mask malodor, it gives little information about whether or not treatments interfere with the fundamental malodor processes occurring in the mouth, that is, the biogenic transformation of substrates into volatile compounds (VCs), including volatile sulphur compounds (VSCs). Changes in the rate of production of VC/VSC give rise to different levels in the breath that can be estimated by the organoleptic intensity method, whereby the judges simply assess the intensity (strength) of the target odor.

It is generally accepted that human beings have a sense of smell that is capable of detecting differences in the strength or concentrations of odor molecules (Engen, 1964). A scale commonly used in breath malodor research is the 0–5 intensity scale described by Allison and Katz (1919) and, more recently, by Rosenberg et al. (1991a,b), where 0 equals a concentration of odorant below threshold, and 5 indicates concentrations that are extremely strong and, it is assumed, close to saturation. The question exists as to the relationship between intensity scores and the real (absolute) concentrations of malodor compounds.

The aim of this study was to compare the mean response from a panel of seven odor judges to a range of concentrations of 8 pure compounds, representative of those important in oral malodor, for better characterization and standardization of the commonly used 0–5 scale as well as for the generation of data on specific odorants for use in modeling of oral malodor processes.

	MATERIALS & METHODS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Chemical Odorants
Butyrate, isovalerate (IVA), putrescine, trimethylamine (TMA), skatole, and dimethyl-disulphide (DMDS) were all obtained from Sigma (Poole, Dorset, UK). Each odorant was weighed and then dissolved in water to give 0.1 M stock solutions. We made amine stocks in 0.01 M NaOH to ensure that they remained in the undissociated (volatile) form during further dilution. Further one-in-ten or one-in-two dilutions were made to give a range of concentrations appropriate for the experiments. Solutions were dispensed as replicate 12-mL vol in Universal bottles (24 mL), with a head space of 12 mL. All weighing and dilutions were carried out in a suitable fume hood, as rapidly as possible (consistent with accuracy), to minimize the loss of odorants as vapor. Hydrogen sulphide (H₂S) (10,000 ppb) and methyl mercaptan (CH₃SH) (1000 ppb) were obtained as calibrated gases diluted in oxygen-free nitrogen (BOC special gases, Reading, UK). Gases were diluted to the concentrations required for the experiments by use of a dynamic flow gas rig (Brooks thermal mass flow controllers, Flotech, Bredbury, UK).

Presentation
For odorants in solution, each bottle was labeled with a code number to allow for subsequent identification of compound and concentration. Each replicate set of odorants (acids, amines, indole, DMDS) consisted of 10 to 12 concentrations ranging from "close to threshold" to "close to saturation". Each set was assigned to a separate rack, and bottle order was randomized.

For VSC, the gas flow rig was adjusted to give a final gas output of 200 mL/min. The gas mixture valves were adjusted to give steady-state VSC concentrations across the smell range. All adjustments were blind to the judges and set according to pre-planned random sequence.

Odor Judges
All experiments were conducted at one study site (UWE Bristol) with seven judges from five different research centers (from the UK, US, Israel, and Jordan). Ethical approval was obtained from the UWE ethical review panel, and all judges gave informed consent to take part. Judges used replicate sets of odorants, and experiments were spaced over a five-day period, with each set being completed within a two-hour session with rest periods between sessions. All judges had some experience with odor judging prior to the study, and three were well-experienced. All judges were tested for full smell acuity by means of a smell identification test (SIT; Sensonics Inc., Haddon Heights, NJ, USA). For all experiments, judges were asked to rate each sample on a 0 to 5 scale and to record each score against the sample code number. The integer descriptives were based on previous work (Rosenberg et al., 1991a, b; Rosenberg and McCulloch, 1992), with slight modification of the highest descriptive. They were: 0, no odor (below the smell threshold); 1, barely noticeable odor; 2, slight odor; 3, moderate odor; 4, strong odor; and 5, extremely strong odor, close to saturation. Judges were allowed to give fractional scores if they thought the strength of the smell was between the integer categories (e.g., 2.5 if between 2 and 3) (Sterer et al., 2002). Score sheets were removed at the end of each experiment. Following sample decoding, the scores for each judge were plotted against the concentration or log concentration of odorant in the liquid or gas phase.

Gas Concentrations
We compared the relative odor of each compound by converting liquid concentrations into corresponding gas concentrations using published Henry’s gas constants (Sander, 1999). The appropriate constants (for 20°C at 1 atmosphere pressure) were converted or expressed as the dimensionless Henry’s Law constant (Kcc). These were taken to be: 3.88 x 10⁴ for butyrate; 2.45 x 10⁴ for IVA; 1.96 x 10² for TMA; and 1.71 x 10¹ for DMDS. Constants for skatole or putrescine could not be found in the literature. Therefore, model constants based on indole (4.08 x 10⁵) (Anon, 2002) and pentylamine (4.10 x 10¹) (Sander, 1999) were used for skatole and putrescine, respectively.

Statistics
Group means were plotted against the log concentration of odorant. Linear regression analysis was then used to measure slope, the 95% CI of the slope, and the scatter of the points around the slope (R² values). Graph construction, statistical analyses, and modeling were conducted with the use of GraphPad Prism version 3.02 for Windows (GraphPad Software, San Diego, CA, USA).

	RESULTS

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
The mean organoleptic scores were initially plotted (y-axis) against the liquid or vapor concentrations of each of the different odorants (x-axis). In all cases, a series of rectangular hyperbola curves was produced. All further plots were made according to a log10 transformation of the concentrations. These are shown for butyrate, IVA, and skatole (Fig. 1A), putrescine, TMA, and DMDS (Fig. 1B), and H₂S and CH₃SH (Fig. 1C). Following log transformation, it was clear that points were best fitted by linear regression for all odorants, with the possible exception of TMA, with all regression slopes giving high R² values (Table) and low P-values (P < 0.001) for scatter. All regression slopes were significantly different from zero slopes (P < 0.0001) and significantly different from each other (P < 0.0005). Therefore, with the possible exception of TMA, all odorants gave a simple exponential relationship between odor score and concentration of odorant. TMA (Fig. 1B) gave a more complex relationship, compatible with a steeper gradient at low concentrations (last 5 points toward the threshold) and a lower gradient at higher concentrations (toward saturation). An F-test comparison of one-site vs. two-site binding models showed that, for certain starting assumptions (Kd₁ values for the steeper gradient < 2 x 10^-8 mol.dm^-3; Kd₂ values for the lower gradient > 2 x 10^-3 mol.dm^-3), the two-site model gave the best fit. Computed threshold gas phase concentrations varied from 7.2 x 10^-13 mol/dm³ for skatole to 5.9 x 10^-6 mol/dm³ for DMDS (Table). Computed saturation scores ranged from 4.2 x 10^-8 mol/dm³ for skatole to 1.3 x 10^-2 mol/dm³ for putrescine (Table).

View larger version (28K): [in this window] [in a new window]   Figure 1. Relationship between mean odor score and concentration of VC/VSC in liquid or gas phase. Plots A and B show VCs in liquid phase; C shows VSC in gas phase. Each point shows the mean score of seven odor judges ± SEM error bars.

View larger version (22K): [in this window] [in a new window]   Figure 2. Relationship between odor score and gas phase concentrations. For H2S and CH3SH, the gas concentrations were those set by experiment. For odorants in liquid phase, the equivalent head-space gas phase concentration was calculated according to Henry law constants (see text for details). Regression lines are extrapolated to y = 0 (the odor threshold) and y = 5 (saturation). Points and error bars have been removed for clarity. The slope for TMA (regression line 7) shows alternate projections for one-site binding (unbroken line) or two-site binding (dashed line).

	DISCUSSION

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

On theoretical grounds, it might be expected that the response of the human nose to increasing concentrations of odor molecules would not be simply ‘proportional’ or linear. In the sniffing process, odor molecules are thought to bind with receptors and odor-binding proteins in the olfactory sensory nerves in the nose. Binding is reversible and therefore subject to the laws of mass action, showing saturation kinetics (akin to substrate interaction with enzyme, or agonist drug-receptor in pharmacology). In common with other receptor-ligand interactions, the dose-response curve would be expected to form a rectangular hyperbola (non-log plot) or exponential/sigmoid (log plot). Assuming that the occupancy of binding sites in the olfactory epithelium as a whole is reflected in the rate of firing of neurones, then the magnitude of response (odor scores) vs. log concentration of odorant would also be expected to trace an exponential/sigmoid shape. This was generally seen to be the case, with the possible exception of TMA, where the data could be interpreted as either simple exponential (one-site binding) but with wide scatter, or as multiphasic with two or more binding sites. For a two-site model, the lowest 5 data-points form a much higher slope than the remaining points, showing high-affinity binding when close to threshold and low-affinity binding at higher gas levels. Only the latter model gave threshold concentrations in line with published work (Bedborough and Trott, 1979). TMA may stimulate widely different families of receptors with different affinities. In contrast to the other odorants, TMA at high-test concentrations produced a response the judges described as "acrid". This change of character may indicate a response by trigeminal nerve receptors (Ohloff, 1994), which may down-regulate the responses of the normal olfactory receptors, or else otherwise interfere with the neuronal output at convergence or in parts of the brain, where the signaling process is decoded or interpreted.

When plotted, the relationships between mean odor scores and log concentrations of odor compounds tend to form a straight line. Valid comparison of different compounds is made possible if all concentrations are expressed in the same gas phase units (mol.dm^-3). The slope is an indication of how the perceived intensity increases with increasing concentration. Extrapolation to intercept estimates the detection thresholds which compare well with published standardized human olfactory thresholds (Devos et al., 1990) giving the same rank order (skatole < CH₃SH < TMA < IVA < butyrate < H₂S < DMDS).

The judges reported that none of the pure odorants alone had the characteristic quality of bad breath. Real breath odor is likely to consist of a wide mix of odor molecules, giving dominant and subdominant odors. Results from this study suggest that an increase in the organoleptic score of one increment would be equivalent to a 4.2- and 7.2-fold increase in gas concentration for H₂S and CH₃SH, respectively, compared with an eight- to 10-fold increase for skatole, DMDS, and butyrate but a much higher 27-, 42-, and 96-fold increase for putrescine, IVA, and TMA, respectively. In the oral cavity, the microbial flora, particularly anaerobes present within the tongue surface biofilm, generate various types of VC (Greenman, 1999), and a one-unit increase on the breath scale relates to an approximately four-fold increase in anaerobes per unit area of tongue surface (Hartley et al., 1999). Assuming that a four-fold increase in cells would increase the production rate of VC four-fold, it suggests that the odorants involved must have high odor power. Furthermore, studies on the dilution of breath samples have shown that an approximately five-fold dilution of breath is required to reduce the odor score (on the 0–5 scale) by one unit (El-Maaytah, 1996), confirming that the main constituents of bad breath have high odor power. Only H₂S has sufficient odor power to account for these findings, thus justifying the role of sulphide monitors in breath odor research.

The contributions of different VCs/VSCs to an overall odor depend on odor threshold, odor power, and gas concentration. For odorants in solution, the head-space gas concentration depends on volatility. Data from this study may be useful for modeling oral malodor processes and show the feasibility of making sets of odor concentrations that equate with the "0 to 5" score increments of the Rosenberg scale. These could be used for the purposes of training, standardization, and calibration of odor judges. Compounds that readily dissolve in water, are relatively stable, can be obtained pure, and show high R² values against odor judges (e.g., butyrate, IVA, putrescine, and skatole) would be the most convenient.

	ACKNOWLEDGMENTS

 
This study was funded by a grant from the Faculty of Applied Sciences, UWE, Bristol.

Received April 2, 2003; Last revision October 22, 2003; Accepted October 24, 2003

	REFERENCES

TOP ABSTRACT INTRODUCTION MATERIALS & METHODS RESULTS DISCUSSION REFERENCES

 
Allison VC, Katz SH (1919). An investigation of stenches and odours for industrial purposes. J Ind Eng Chem 11:336–338.

Anonymous (2002). Clean Air Act Information Network: synthetic organic chemical manufacturing industry report, table 4, p.389. http://envinfo.com/caain/199/socmirnj.pdf

ASTM (1968). Manual on sensory testing methods. STP 434. Philadelphia: American Society for Testing and Materials, pp. 1–77.

Bedborough DR, Trott PE, editors (1979). The sensory measurement of odours by dynamic dilution. Report No. LR 299 (AP). Stevenage: Warren Spring Laboratory, pp. 3–27.

Devos M, Patte F, Rouault J, Laffort P, van Gemert LJ (1990). Standardised human olfactory thresholds. Oxford: IRL Press at Oxford University Press, pp. 35–161.

El-Maaytah MA (1996). A clinical and microbial study of oral malodour (PhD thesis). Bristol, UK: University of the West of England, pp. 77–79.

Engen T (1964). Psychophysical scaling of odor intensity and quality. Ann NY Acad Sci 116:504–516.

Greenman J (1999). The microbial aetiology of halitosis. In: Dental plaque revisited: oral biofilms in health and disease. Newman HN, Wilson M, editors. Cardiff: BioLine Publications, pp. 419–442.

Hartley MG, McKenzie C, Greenman J, El-Maaytah MA, Scully C, Porter S (1999). Tongue microbiota and malodour: effects of metronidazole mouthrinse on tongue microbiota and breath odour. Micro Ecol Health Dis 11:226–233.

Ohloff G (1994). Scents and fragrances. Heidelberg: Springer-Verlag, p. 6.

Rosenberg M (2002). The science of bad breath. Sci Am 286:72–79.

Rosenberg M, McCulloch CA (1992). Measurements of oral malodor: current methods and future prospects. J Periodontol 63:776–782.[ISI][Medline]

Rosenberg M, Septon I, Eli I, Bar-Ness R, Gelernter I, Brenner S, et al. (1991a). Halitosis measurement by an industrial sulphide monitor. J Periodontol 62:487–489.[ISI][Medline]

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

Sander R (1999). Compilation of Henry’s Law constants for inorganic and organic species of potential importance in environmental chemistry (version 3). http://www.mpch-mainz.mpg.de/~sander/res/henry.html

Scully C, Porter S, Greenman J (1994). What to do about halitosis (editorial). BMJ 308:217–218.[Free Full Text]

Shimura M, Yasuno Y, Iwakura M, Shimada Y, Sakai S, Suzuki K, et al. (1996). A new monitor with a zinc-oxide thin film semiconductor sensor for the measurement of volatile sulphur compounds in mouth air. J Periodontol 67:396–402.[ISI][Medline]

Sterer N, Greenstein RB, Rosenberg M (2002). Beta-galactosidase activity in saliva is associated with oral malodor. J Dent Res 81:182–185.[Abstract/Free Full Text]

Stevens SS (1957). On the psychophysical law. Psychol Rev 64:153–181.[ISI][Medline]

Tonzetich J (1971). Direct gas chromatographic analysis of sulphur compounds in mouth air in man. Arch Oral Biol 16:587–597.[ISI][Medline]

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

Yaegaki K, Matudaira H, Sano S, Kitamura T (1996). Attitudes towards one’s and others’ oral malodour. In: Bad breath: a multidisciplinary approach. van Steenberghe D, Rosenberg M, editors. Belgium: Leuven University Press, pp. 217–230.

This article has been cited by other articles:

				N. Sterer and M. Rosenberg Streptococcus salivarius Promotes Mucin Putrefaction and Malodor Production by Porphyromonas gingivalis. J. Dent. Res., October 1, 2006; 85(10): 910 - 914. [Abstract] [Full Text] [PDF]

				S R Porter and C Scully Oral malodour (halitosis) BMJ, September 23, 2006; 333(7569): 632 - 635. [Full Text] [PDF]

				D. L. Eubanks Canine Oral Malodor J. Am. Anim. Hosp. Assoc., January 1, 2006; 42(1): 77 - 79. [Abstract] [Full Text] [PDF]

				J. GREENMAN, M. EL-MAAYTAH, J. DUFFIELD, P. SPENCER, M. ROSENBERG, D. CORRY, S. SAAD, P. LENTON, G. MAJERUS, and S. NACHNANI Assessing the relationship between concentrations of malodor compounds and odor scores from judges J Am Dent Assoc, June 1, 2005; 136(6): 749 - 757. [Abstract] [Full Text] [PDF]

This Article Abstract Figures Only Full Text (PDF) Services Similar articles in this journal Similar articles in ISI Web of Science Similar articles in PubMed Alert me to new issues of the journal Download to citation manager Citing Articles Citing Articles via HighWire Citing Articles via ISI Web of Science (19) Citing Articles via Google Scholar Google Scholar Articles by Greenman, J. Articles by El-Maaytah, M. Search for Related Content PubMed PubMed Citation Articles by Greenman, J. Articles by El-Maaytah, M.