Human Breath Odors and Their Use

in Diagnosis
CHRIS L. WHITTLE,a STEVEN FAKHARZADEH,b JASON EADES,a
AND GEORGE PRETIa,b
a MonellChemical Senses Center, 3500 Market Street, Philadelphia,
Pennsylvania 19104, USA
b Department of Dermatology, School of Medicine, University of Pennsylvania,
Philadelphia, Pennsylvania, USA

ABSTRACT: Humans emit a complex array of volatile and nonvolatile

molecules that are influenced by an individual’s genetics, health, diet,
and stress. Olfaction is the most ancient of our distal senses and may be
used to evaluate food and environmental toxins as well as recognize kin
and potential predators. Many body odors evolved to be olfactory mes-
sengers, which convey information between individuals. Consequently,
those practicing the healing arts have used olfaction to aid in their diag-
nosis of disease since the dawn of medical practice. Studies using modern
instrumental analyses have focused upon analysis of breath volatiles for
biomarkers of internal diseases. In these studies, a subject’s oral health
status appears to seldom be considered. However, saliva and properly
collected alveolar air samples must pass over or come in contact with the
posterior dorsal surface of the tongue, a site of bacterial plaque devel-
opment and source of halitosis-related volatiles. Because of our basic re-
search into the nature of human body odors, our lab has received referrals
of people with idiopathic malodor production, from either the oral cavity
or body. We developed a protocol to help differentiate individuals with
chronic halitosis from those with the genetic, odor-producing metabolic
disorder trimethylaminuria (TMAU). In our referred population, TMAU
is the largest cause of undiagnosed body odor. Many TMAU-positive in-
dividuals present with oral symptoms of dysguesia and halitosis as well
as body odor. We present data regarding the presentation of our referred
subjects as well as the analytical results from a small number of these
subjects regarding their oral levels of halitosis-related malodorants and
trimethylamine.

KEYWORDS: human breath; biomarkers; breath condensate; dis-

ease; pathology; trimethylamine; trimethylaminuria; choline; flavin-
containing monooxygenase 3 (FMO3); SPME; GC/MS

Address for correspondence: George Preti, Ph.D., Monell Chemical Senses Center, 3500 Market
Street, Philadelphia, PA 19104. Voice: 215-898-4713; fax: 215-898-2084.
preti@monell.org

Ann. N.Y. Acad. Sci. 1098: 252–266 (2007). 

C 2007 New York Academy of Sciences.

doi: 10.1196/annals.1384.011
252
WHITTLE et al. 253

INTRODUCTION

Humans emit a complex array of nonvolatile and volatile molecules. The

metabolic processes of an individual, and hence the compounds emitted into
the environment, may be influenced by genetics, diet, stress, and the immune
status of the individual. Human olfaction is the most ancient of our distal senses
and does provide information from distant sources in real time. Olfactory in-
formation may be used to detect and evaluate food sources and environmental
toxins as well as to recognize kin and potential predators. In addition, many
body odors evolved to be olfactory cues that convey information between in-
dividuals.1 Large numbers of volatile compounds may be emitted from several
areas of the body that are prone to odor production; these include the scalp,
axillae, feet, groin, and oral cavity.2 Consequently, it is not surprising that
physicians have used their olfactory and gustatory senses to aid in the differ-
ential diagnosis of diseases since the beginning of medical practice.3–5
Hippocrates is reported to have used exhaled breath and smelled the breath
of patients as part of his assessment.6 Breath testing as a scientific tool was
pioneered in the 18th century by Lavoisier and Laplace.7 These pioneers es-
tablished the presence of CO 2 in exhaled breath air. In the second half of the
20th century, the development of more sophisticated analytical techniques,
such as gas chromatography, has allowed the separation and identification of
volatile compounds from complex biological matrices, such as exhaled breath.
In the 1960s, the separating power of gas chromatography was combined with
mass spectrometry to create combined gas chromatography/mass spectrometry
(GC/MS); see Watson for a review.8 This development allowed the separation
of complex mixtures as well as structural identification of separated, volatile
components.
Numerous investigators have captured and concentrated breath and salivary
volatiles. Studies by these researchers have focused upon establishing the nor-
mal breath constituents and searching for biomarkers from the oral cavity to
aid in the diagnosis of illness or severity of disease.9–13
Analysis of exhaled breath provides a unique opportunity to examine the
organic constituents of blood because alveolar breath reflects the concentra-
tion of metabolites that have passively diffused across the pulmonary alveo-
lar membrane. Breath primarily consists of nitrogen, oxygen, carbon dioxide,
inert gases, water vapor, and a trace amount of volatile organic compounds
(VOCs) (e.g., acetone, isoprene, and pentane) that are present in nano- to
picomolar concentrations.14 Because the organic constituents of exhaled breath
are representative of the blood-borne concentrations of metabolites, breath
analysis provides a noninvasive means to examine blood-borne constituents
relative to using blood and/or urine samples. Some of the advantages of us-
ing exhaled breath include the fact that lung air volatiles reflect the arte-
rial concentrations of biological substances; also the VOCs are removed from
complex fluid matrices, such as blood and urine; consequently, the complete
254 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

sample of all compounds are present in the collected sample with no

work-up required prior to analysis of the sample. However, breath analy-
sis also has limitations. One issue that hinders routine breath tests in clin-
ical practice is a lack of standardized collection and analysis methods and
the need for complex and unfamiliar (for clinicians) instrumentation. In
addition, simple and automated chemical tests that are routinely used for
blood and urine analyses are relatively inexpensive when compared to the
most commonly used instrumentation for breath analyses, GC/MS. The lat-
ter requires an expensive initial investment (> $70,000) and maintenance
(>$10,000/year) as well as skilled operators who must be able to interpret
data from large complex data sets.9,15
A large number of studies using modern instrumental analyses have fo-
cused upon analysis of breath volatiles in search of biomarkers for disease
states. The vast majority of these studies have examined respiratory system
diseases such as asthma. Nitric oxide is the most extensively exhaled marker
investigated, and has been linked to a host of respiratory ailments including
chronic obstructive pulmonary disease,16,17 rhinitis,18,19 rhinorrhea,20 chronic
cough (primary),21,22 asthma,23,24 cystic fibrosis,25 bronchiectasis,26,27 and
lung cancer.28,29 Numerous exhaled markers of disease have been studied,
such as several leukotrienes and nitrogenous compounds.30–33
Many recent studies focused upon branched or methylated hydrocarbons or
VOCs with broad diagnostic potential. These have been postulated to increase
in breath effluvia as a function of oxidative stress or damage (see Miekisch et al.
for a review).11 Although hydrocarbons are commonly found in the air of urban
areas, where many of the above-cited studies have taken place, investigators
appear certain that these products are products of human biological activity
and not exogenous, environmental contaminants. These include the straight-
chained hydrocarbons, ethane and pentane, which are markers of oxidative
damage and have also been linked to diseases, such as asthma,34,35 lung and
breast cancer,33,36 interstitial lung disease,37 chronic obstructive pulmonary
disease,35 and heart rejection.38 Numerous other volatile compounds have been
found in exhaled breath and have been linked to different diseases. Although not
exhaustive, TABLE 1 illustrates the range of VOCs that may serve as biomarkers
for disease states.
In these studies, however, a subject’s oral health status appears to seldom
be considered. Halitosis is one of the most frequent complaints expressed
by dental patients. Approximately 90% of oral malodor is thought to orig-
inate from the oral cavity, with the remaining 10% originating from distal
points in the digestive and respiratory systems.39–41 Persistent halitosis may
be indicative of underlying medical conditions, such as diabetes, leukemia,
gastrointestinal ulcers, lung cancer, trimethylaminuria (TMAU), and several
other idiopathic conditions.39,42 Even correctly collected alveolar air and saliva
samples must pass over or come in contact with the posterior dorsal surface
of the tongue. This is a site of bacterial plaque development that is a principal
WHITTLE et al. 255

TABLE 1. Oral/breath volatiles identified in patients with systemic disease

Pathologic condition Compound(s) Reference
Diabetes mellitus Acetone, other ketones Booth and Ostenson, 196663
Walsh, 200464
Breath methylated alkane Phillips et al., 200438
contour (BMAC)
Sleep apnea Interleukin IL-6, Carpagnano et al., 200265
8-isoprostane
H. pylori infection Nitrate, cyanide Lechner et al., 200566
Carbon dioxide Pathak et al., 199467
Sickle cell disease Carbon monoxide Sylvester et al., 200568
Methionine adenosyl- Dimethylsulfide Chamberlin et al., 199646
transferase deficiency
Asthma Leukotrienes Montuschi and Barnes,
2002;30 Hanazawa et al.,
2000;69 Cap et al., 200470
Breast cancer 2-propanol, 2,3-dihydro-1- Phillips et al., 200633
phenyl-4 (1H)-quinazoli-
none, 1-phenyl-ethanone,
heptanal
Lung carcinoma Acetone, methylethylketone, Gordon et al., 198571
n-propanol
Aniline, o-toluidine Preti et al., 198872
Alkanes, mono-methylated Phillips et al., 200328
breath alkanes, alkenes
Chronic obstructive Hydrogen peroxide Dekhuijzen et al., 199773
pulmonary disease
Nitrosothiols Corradi et al., 200131
Nitrosothiols nitric oxide Liu and Thomas, 200574
Cystic fibrosis 8-isoprostane Montuschi et al., 200075
Leukotriene B(4), interleukin- Bodini et al., 200576
8
Liver disease Hydrogen disulfide, limonene Friedman et al., 199447
Noncholestatic Hydrogen disulfide Friedman et al., 199447
Primary biliary
cirrhosis
Decompensated C 2 –C 5 aliphatic acids, Chen et al., 1970a48 ;
cirrhosis methylmercaptan Kaji et al., 197877
of the liver
(foetor hepaticus)
Ethanethiol, dimethylsulfide Tangerman et al., 199478
Uremia/kidney failure Dimethylamine, Simenhoff et al., 197779
trimethylamine
Trimethylaminuria Trimethylaminine Leopold et al., 199056
Preti et al., 199512
256 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

source of halitosis-related volatiles.43 The bad breath mixture has been studied
principally for its volatile sulfur compounds (VSCs): hydrogen sulfide,
methylmercaptan, dimethyl sulfide, carbonyl sulfide. These compounds pro-
vide much of the impact odor of halitosis; however, the breath of individuals
with halitosis does contain a variety of volatile organic odorants, not just
VSCs.44,45 Consequently, VSCs as well as many other volatiles may be indica-
tive of oral-related health issues rather than non-oral disorders and disease.
Studies addressing markers from exhaled breath may want to address the sub-
ject’s oral health status.
Several authors examining exhaled breath of patients with liver diseases
for VSCs have not addressed their subjects’ oral health.46,47 These authors
found elevated levels of VSCs in these patients versus controls. However, the
historical link of liver diseases with foul breath odor (foetor hepaticus) appears
to exclude an oral cause for breath odor in these patients. In addition, Chen
et al. used a methionine challenge to elicit VSC production in their subjects.48
Because of our basic research into the nature and origin of human body
odors, our lab has become the focal point for a large number of referrals of
people with idiopathic malodor production from either the body or oral cavity.
Regardless of presenting symptoms and to help differentiate individuals with
bad breath from other possible disorders, all individuals are examined for
odor production from the oral cavity and upper body by the same protocol
first described by Preti et al.12 A central part of this work-up is the choline
challenge test for TMAU developed by Tjoa and Fennessey.49
Trimethylaminuria was first described by Humbert,50 and is a metabolic
disorder characterized by the inability of individuals to oxidize and convert
dietary derived trimethylamine (TMA) to trimethylamine N-oxide (TMAO)
in the liver. This disorder results from an inherited autosomal recessive trait
in the gene, which codes for the flavin-containing monooxygenase enzyme 3
(FMO3). The genetic changes range from gene mutations associated with the
most severe cases to the more common single nucleotide polymorphic changes
in the FMO3 gene that may be associated with the less severe cases.51,52
Malodorous TMA is formed in the gut by bacterial metabolism of di-
etary constituents, mainly choline. In normal individuals, TMA is con-
verted/oxidized to TMAO at >95% efficiency by FMO3. Individuals that have
FMO3 metabolic capacity <90% conversion of TMA to TMAO are considered
positive for TMAU.53,54 TMAO is nonodorous, more polar and water-soluble
than TMA, and readily excreted in the urine.55 Individuals suffering from
TMAU have a reduced capacity to oxidize TMA to TMAO.
TMA is a gas at body temperature and has a foul, rotten fish odor. At low
concentrations it may be perceived as unpleasant or garbage-like. The inability
to efficiently oxidize TMA results in the sporadic production of a body odor
that is perceived as foul, unpleasant, and in its most extreme cases fish-like.
This odor is caused by excess, unmetabolized TMA present in the circulatory
system that is excreted in urine, sweat, breath, and saliva. Because there are
WHITTLE et al. 257

many foods that are rich in choline (i.e., eggs, certain legumes, and organ
meats), TMAU-affected individuals, family members, friends, and physicians
are unlikely to associate the odor with food intake.
Symptoms may include foul body odor, halitosis, and/or dysguesia that can
produce social embarrassment and may only be temporarily relieved by normal
hygienic procedures.56 The main difficulties experienced by TMAU-affected
individuals are psychosocial ones that are caused by sporadic, undiagnosed
odor production.57
To enable an easier diagnosis for TMAU and to examine whether or not ele-
vated salivary levels of TMA might accompany oral symptoms in our referred
subjects, we began collecting saliva from all subjects reporting to our lab with
malodor production problems to examine this fluid for TMA.

METHODS

All procedures were approved by the Office of Regulatory Affairs at the

University of Pennsylvania and informed consent was obtained from each
subject before study participation.
We adapted the method of Mills and Walker58 that employs the tech-
niques of solid-phase microextraction (SPME) and gas chromatography-
mass spectrometry (GC/MS) to quantify salivary TMA levels.57 This adap-
tion included the use of perdeuterated trimethylamine (D 9 -TMA) as an
internal standard to aid in quantitation. We report herein on the data ob-
tained from the saliva of several subjects using SPME-GC/MS. The lin-
ear range of the method was investigated by obtaining a calibration curve
in the concentration ranges of interest. Trimethylamine concentrations (as
trimethylamine hydrochloride) were made up from a stock solution of
2 mg/mL in acidified water (pH∼1). The following concentrations were used:
2 mL each of acidified, distilled water containing TMA concentrations of
0.006 mg/mL, 0.003 mg/mL, 0.0012 mg/mL, 0.0006 mg/mL, 0.0003 mg/mL,
and 0.00012 mg/mL. TMA-HCl were combined with 2 mL of 0.002 mg/mL
perdeuterated (D 9 )-TMA-DCl and 0.5 mL of 12 M NaOH. Each solution was
equilibrated for 15 min at 50◦ C prior to exposing the SPME fiber to collect
volatiles for 15 min. Each solution was analyzed a minimum of three times for
both the calibration curve and patient samples.

Gas Chromatography/Mass Spectrometry

After collecting salivary volatiles containing the TMA, the SPME fiber was
inserted into the hot injector of the GC/MS (held at 230◦ C) and exposed for
1 min to desorb volatiles collected on the fiber. A Thermoquest/Finnigan Voy-
ager GC/MS with Xcalibur software (ThermoElectron Corp., San Jose, CA,
USA) was used for all analyses. A polar, Stabilwax column, 30 M × 0.32 mm
258 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

with 1.0- coating, (Restek Corp., Bellefonte, PA, USA) was used for separa-
tion and analysis of the volatiles extracted from the samples. The separation
of TMA from other components was done isothermally by holding the col-
umn at 50◦ C for 5 min. TMA elutes within the first 5 min; consequently, we
rapidly increased the column temperature at 40◦ C/min to the final temperature
of 220◦ C to bake-off undesired volatiles. The column was recycled back to
the starting temperature of 50◦ C for the next analysis. The injection port was
set at 230◦ C. Helium carrier gas was used at a constant column flow rate of
2.5 mL/min throughout the analysis.
Data acquisition and operating parameters for the mass spectrometer were
set as follows: scan rate 2/sec; scan range m/z 41 to m/z 440; ion source tempera-
ture 200◦ C; ionizing energy 70eV. Identification of structures/compounds was
performed using both the NIST ’02 library, as well as a manual interpretation
of mass spectra compared with those reported in the literature.

Gas Chromatography with Flame Photometric Detection to Measure Oral

Volatile Sulfur Compounds
We used gas chromatography with flame photometric detection (GC/FPD)
to detect and quantify the amount of volatile sulfur compounds in the oral
cavity of all subjects. The instrument used for these analyses was a Finni-
gan 9001 fitted with a flame photometric detector. We employed the method
of Tonzetich,59 albeit with modifications previously reported by Kostelc
et al. and Preti et al.60,61 Analysis of the subject’s mouth air was performed
in duplicate. Each analysis employed 10 mL of the subject’s mouth air pulled
into the chromatograph via an atmospheric sampling loop using a gas tight
syringe.
FMO3 Gene Sequence Analysis

Genomic DNA was extracted from patient peripheral blood samples using
the QIAamp DNA Mini Kit (Qiagen, Valencia, CA, USA). FMO3 coding ex-
ons 2-9 and flanking intron sequences were amplified by PCR using primers
and conditions reported by Dolphin et al.62 PCR amplification products were
purified from 1.5% agarose gels prior to sequencing. All samples were submit-
ted to the DNA Sequencing Facility at the University of Pennsylvania School
of Medicine for analysis. Sequencing reactions were performed in an ABI
GeneAmp
9700 thermal cycler, resolved with an ABI 3730 DNA sequencer,
and analyzed using ABI Sequencing Analysis software v 5.1 (ABI, Applied
Biosystems, Incorporated, Foster City, CA, USA).
RESULTS
We have seen and tested more than 300 individuals in our laboratory using
the protocol outlined in TABLE 2. One hundred two of these have been diagnosed
with some form of TMAU using the choline challenge test.49 The presenting
WHITTLE et al. 259

TABLE 2. Diagnostic protocol for referred individuals seen at the Monell Chemical Senses
Center
1. First morning voided urine (base line for choline challenge test).
2. Individual presents in fasted condition (no cologne, cosmetics, fragrances, teeth and
tongue plaque brushing); questionnaire and interview.
3. Organoleptic evaluation of individual’s body odor and breath: 3 judges, scale of 1–10.
4. Collect axillary odors by placing a 4× 4-inch cotton pad in individual’s axillae.
5. Analysis of individual’s mouth air by GC/FPD for volatile sulfur compounds (VSCs):
H 2 S, CH 3 SH, (CH 3 ) 2 S.
6. Swab tongue for bacterial plaque and use for collection and analysis of volatiles.
7. Determination of resting whole mouth saliva flow rate using the methods described in
Christensen and Navazesh, 1982.80
8. Determine and collect two, whole mouth-stimulated saliva samples using flavorless
gum base. One of these is collected into a vial with 0.2 mL of 6N HCl for analysis of
precholine challenge TMA. Other volatiles are examined by SPME-GC/FPD and
SPME-GC/MS.
9. Lung air collection (10 L) for analysis of volatiles by GC/FPD.
10. Analyze tongue plaque volatiles by SPME-GC/MS and SPME-GC/FPD.
11. Administer 5 g of choline in 12–16 oz of juice for choline challengea .
12. Remove axillary pads and perform organoleptic evaluation.
13. For the next 24 h individual must:
(A) Collect urine in three 8-h aliquots
(B) Collect two stimulated whole mouth saliva samples during each 8-h time period
using vials with 0.2 mL of 6N HCl.
Next day:
Individual returns saliva and urine samples.
Blood sample taken for genotyping of FMO3 gene.
a All urine samples collected during choline challenge testing were analyzed for trimethylamine and
triemthylamine oxide in the laboratory of Dr. Paul Fennessey and Ms. Susan Tjoa,49 at the University
of Colorado Health Sciences Center.

symptoms for the TMAU-positive individuals are illustrated in FIGURE 1. A

majority of individuals had oral symptoms, complaining of either bad breath
or bad taste, many in conjunction with body odors.
In addition, as illustrated in FIGURE 2, the majority of our TMAU-positive
individuals were females. However, regardless of gender, in our experience,
TMAU was the largest cause for undiagnosed body odor. The large number of
TMAU-positive individuals who listed oral complaints led us to hypothesize
that salivary concentrations of TMA were responsible for these symptoms.
During most individuals’ visits, saliva was collected in conjunction with
urine as described in TABLE 2. This has resulted in an archive of more than
270 frozen (at –10◦ C) saliva samples. However, only a small number of the
saliva samples collected have been examined thus far for TMA levels: six
individuals each from TMAU-positive and -negative categories. We report
these preliminary results here.
TABLE 3 lists the mean salivary concentrations of TMA measured in each
8-h interval for six TMAU-positive and six TMAU-negative subjects. Clearly,
the saliva of the positive subjects contains far more TMA than the saliva from
260 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

FIGURE 1. Presenting symptoms and the numbers of subjects reporting each symptom,
in the subjects’ own words. “Body odor” is the most common presenting symptom found in
the literature pertaining to these patients. The symptoms were not always confirmed during
subject evaluation using the protocol in TABLE 2.

TMAU-negative individuals, but there is a great deal of variation in the data,

particularly from the positive subjects. Two of the positive subjects account
for much of this variation because they demonstrated more than a 100-fold
increase in salivary TMA levels from their precholine challenge base line to
their highest levels in the 2nd or 3rd 8-h interval: Positive male 3 went from a
base line of 25 ng/mL to 4,812 ng/mL in the 3rd, 8-hour segment; and positive
male 6 went from a base line of 46.7 ng/mL to 4, 511 ng/mL in the 2nd, 8-h
segment. Each of these subjects had a very low conversion of TMA to TMAO,
indicative of one or more genetic mutations present in the gene for FMO3:
male 3 had 25% TMAO conversion; male 6 had only 11% TMAO conversion.
We also measured the volatile sulfur compounds in the oral cavity associated
with each of these subjects. Results are summarized in TABLE 4. In the subjects
chosen for these analyses, we found that, on average, TMAU-negative subjects
had greater concentrations of each VSC measured; these concentrations were
also converted to parts per billion (ppb) levels and are presented in TABLE 4. The
difference in VSC levels between the TMAU-negative and positive individuals
is a result of the subjects chosen for these analyses: in our clinical experience,
both TMAU-positive and -negative individuals may have halitosis. However,
the two conditions, TMAU and chronic halitosis caused by bacterial tongue
plaque, are independent of each other.
WHITTLE et al. 261

FIGURE 2. The gender distribution of TMAU-positive subjects seen in our laboratory

is shown in this figure. We do not know whether or not the larger number of females
affected is due to hormonal differences in the regulation of the FMO3 gene. Females report
symptoms before and after menopause.

DISCUSSION

Our findings regarding the presenting symptoms of TMAU-affected individ-

uals are in contrast to results found in most of the medical literature. Articles
discussing TMAU suggest to the reader that all sufferers have a fishy body
odor presentation. In our population, all of whom have been seen in person,
the fish odor presentation was present in only about 10% of individuals who
are TMAU-positive. Further to this point, these individuals emitted a strong
fish odor recognizable at social distances only after choline challenge. Conse-
quently, the assumption that the individual with TMAU will always smell “like
fish” is incorrect and is often the reason that many TMAU-affected individuals

TABLE 3. Mean salivary trimethylamine concentrations: precholine and for each 8-h
period (ng/mL)
Precholine 1st 8 hour 2nd 8 hour 3rd 8 hour
Negative (n = 6) 37.41 48.37 61.07 68.40
SE 15.27 19.74 24.93 27.92
Positive (n = 6) 31.2 595.61 799.20 1736.57
SE 12.74 243.10 742.39 823.88
262 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

TABLE 4. Volatile sulfur compounds (VSCs) in the mouth air of referred subjects: ng/10
mL of mouth air
COS H2S CH 3 SH (CH 3 ) 2 S
TMAU-negative subjects: N = 5a
Mean 1.08 2.17 1.97 0.21
SE 0.67 1.38 1.04 0.10
Mean parts per billion (ppb) of total VSC in the mouth air of TMAU-negative
subjects = 453

COS H2S CH 3 SH (CH 3 ) 2 S

TMAU-positive subjects: N = 5a
Mean 0.416 0.83 0.0 0.0
SE 0.415 0.57 0.0 0.0
Mean ppb of total VSC in the mouth air of TMAU-positive subjects = 104
a Instrument malfunction caused one subject in each group not to be sampled for VSC levels.
ABBREVIATION: COS = carbonyl sulfide; H 2 S = hydrogen sulfide; CH 3 SH = methylmercaptan;
(CH 3 ) 2 S = dimethylsulfide.

are sent from one clinical specialist to another: quite often they are sent to a
psychiatrist since their reported symptoms are thought to be subjective.
The choline challenge test for TMAU provides a recognized means for diag-
nosing this disorder. The diagnosis currently relies upon a 24-h urine collection,
divided into three 8-h aliquots.49 In our initial attempt to extend this diagnosis
to saliva, we collected whole mouth-stimulated saliva to determine salivary
TMA levels. Our hypothesis that the oral symptoms of many TMAU-affected
individuals appears to be supported by the preliminary results presented here,
although the numbers of subjects analyzed is still small. The data in TABLE 3
show much larger variation in the salivary TMA concentrations of TMAU-
positive versus TMAU-negative individuals. TMAU is known to be caused
by a “spectrum” of genetic changes to the gene that codes for FMO3;51 con-
sequently, this variation may be due, in part, to differences in the genotype
of each of the TMAU-positive individuals. This is supported by our clinical
observations regarding the odor of different individuals as well as genotyping
data. Two of the six TMAU-positive individuals whose saliva was analyzed
presented with overt fish odor from their upper body and oral cavity after
(∼22 h) choline challenge. As noted above, each of these male subjects had a
low conversion of TMA to TMAO (<25%) and documented mutations in their
FMO3 gene (data not shown).51

ACKNOWLEDGMENTS

This research was supported, in part, by the NIH Institutional Training Grant
2T32DC00014 as well as unrestricted funds from the Monell Chemical Senses
Center.
WHITTLE et al. 263

REFERENCES
1. WYSOCKI, C.J. & G. PRETI. 2004. Facts, fallacies, fears, and frustrations with human
pheromones. Anat. Rec. 281A: 1201–1211.
2. WYSOCKI, C.J. & G. PRETI. 2000. Human body odors and their perception. Jpn. J.
Taste Smell Res. 7: 19–42.
3. LIDDELL, K. 1976. Smell as a diagnostic marker. J. Postgrad. Med. 52: 136–138.
4. PHILLIPS, M. 1992. Breath tests in medicine. Sci. Am. 267: 74–79.
5. PENN, D. & W.K. POTTS. 1998. Chemical signals and parasite-mediated sexual
selection. Trends Ecol. Evol. 13: 391–396.
6. GEIST H.C.I. 1957. Halitosis in ancient literature. Dent. Abst. 2: 417–418.
7. CASPARY, W.F. 1978. Breath tests. Clin. Gastroenterol. 7: 351–74.
8. WATSON, J.T. 1998. A historical perspective and commentary on pioneering devel-
opments in gas chromatography/mass spectrometry at MIT. J. Mass Spect. 33:
103–108.
9. CAO, W. & Y. DUAN. 2006. Breath analysis: potential for clinical diagnosis and
exposure assessment. Clin. Chem. 52: 800–811.
10. RAHMAN, I. & F. KELLY. 2003. Biomarkers in breath condensate: a promising
new non-invasive technique in free radical research. Free Rad. Res. 37: 1253–
1266.
11. MIEKISCH, W., J.K. SCHUBERT & G.F.E. NOELDGE-SCHOMBURG. 2004. Diagnostic
potential of breath analysis—focus on volatile organic compounds. Clin. Chim.
Acta 347: 25–39.
12. PRETI, G., H.J. LAWLEY, C.A. HORMANN, et al. 1995. Non-oral and oral aspects of
oral malodor. In Bad Breath Research Perspectives. M. Rosenberg, Ed.: 149–174.
Ramot Publishing. Tel Aviv, Israel.
13. KHARITONOV, S.A. & P.J. BARNES. 2002. Biomarkers of some pulmonary diseases
in exhaled breath. Biomarkers 7: 1–32.
14. MUKHOPADHYAY, R. 2004. Don’t waste your breath. Anal. Chem. 76: 273A–276A.
15. WOLFRAM, M., J.K. SCHUBERT & G.F.E. NOELDGE-SCHOMBURG. 2004. Diagnostic
potential of breath analysis—focus on volatile organic compounds. Clin. Chim.
Acta 347: 25–39.
16. KHARITONOV, S.A. & P.J. BARNES. 2001. Exhaled markers of pulmonary disease.
Am. J. Resp. Crit. Care Med. 163: 1693–1722.
17. MARCZIN, N., S.A. KHARITONOV & S.M.H. YACOUB. 2003. Disease Markers in
Exhaled Breath. Dekker. New York.
18. BARALDI, E., N.M. AZZOLIN, S. CARRA, et al. 1998. Effect of topical steroids on
nasal nitric oxide production in children with perennial allergic rhinitis: a pilot
study. Resp. Med. 92: 558–561.
19. HENRIKSEN, A.H., M. SUE-CHU, T. LINGAAS HOLMEN, et al. 1999. Exhaled and
nasal NO levels in allergic rhinitis: relation to sensitization, pollen season, and
bronchial hyperresponsiveness. Eur. Resp. J. 13: 1301–1306.
20. FRANKLIN, P.J., S.W. TURNER, G.L. HALL, et al. 2005. Exhaled nitric oxide is reduced
in infants with rhinorrhea. Pediat. Pulmonol. 39: 117–119.
21. DUPONT, L.J., F. ROCHETTE, M.G. DEMEDTS & G.M. VERLEDEN. 1998. Exhaled
nitric oxide correlates with airway hyperresponsiveness in steroid-naive patients
with mild asthma. Am. J. Resp. Crit. Care Med. 157: 894–898.
22. CHATKIN, J.M.A.K., P.E. SILKOFF, P. MCCLEAN, et al. 1999. Exhaled nitric oxide as
a noninvasive assessment of chronic cough. Am. J. Resp. Crit. Care Med. 159:
1810–1813.
264 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

23. DUPONT, L.J.F., M.G. DEMEDTS & G.M. VERLEDEN. 1999. Prospective evaluation
of the accuracy of exhaled nitric oxide for the diagnosis of asthma. Am. J. Resp.
Crit. Care Med. 159: 861.
24. DUPONT, L.J., M.G. DEMEDTS & G.M. VERLEDEN. 2003. Prospective evaluation
of the validity of exhaled nitric oxide for the diagnosis of asthma. Chest 123:
751–756.
25. THOMAS, S.R., S.A. KHARITONOV, S.F. SCOTT, et al. 2000. Nasal and exhaled nitric
oxide is reduced in adult patients with cystic fibrosis and does not correlate with
cystic fibrosis genotype. Chest 117: 1085–1089.
26. TRACEY, W.R., C. XUE, V. KLINGHOFER, et al. 1994. Immunocytochemical detection
of inducible NO synthase in human lung. Am. J. Physiol. 266: L722–L727.
27. HO, L.P., J.A. INNES & A.P. GREENING. 1998. Exhaled nitric oxide is not elevated
in the inflammatory airways diseases of cystic fibrosis and bronchiectasis. Eur.
Resp. J. 12: 1290–1294.
28. PHILLIPS, M., R.N. CATANEO, B.A. DITKOFF, et al. 2003. Volatile markers of breast
cancer in the breath. Breast J. 9: 184–191.
29. PHILLIPS, M., K. GLEESON, J.M. HUGHES, et al. 1999. Volatile organic compounds
in breath as markers of lung cancer: a cross-sectional study. Lancet 353: 1930–
1933.
30. MONTUSCHI, P. & P.J. BARNES. 2002. Exhaled leukotrienes and prostaglandins in
asthma. J. Allergy Clin. Immunol. 109: 615–620.
31. CORRADI, M., P. MONTUSCHI, L.E. DONNELLY, et al. 2001. Increased nitrosothiols
in exhaled breath condensate in inflammatory airway diseases. Am. J. Resp. Crit.
Care Med. 163: 854–858.
32. BALINT, B., S.A. KHARITONOV, T. HANAZAWA, et al. 2001. Increased nitrotyrosine
in exhaled breath condensate in cystic fibrosis. Eur. Resp. J. 17: 1201–1207.
33. PHILLIPS, M., R.N. CATANEO, B.A. DITKOFF, et al. 2006. Prediction of breast can-
cer using volatile biomarkers in the breath. Breast Cancer Res. Treat. 99: 19–
21.
34. OLOPADE, C.O., M. ZAKKAR, W.I. SWEDLER & I. RUBINSTEIN. 1997. Exhaled pentane
levels in acute asthma. Chest 111: 862–865.
35. PAREDI, P., S.A. KHARITONOV, D. LEAK, et al. 2000. Exhaled ethane, a marker of
lipid peroxidation is elevated in chronic obstructive pulmonary disease. Am. J.
Resp. Crit. Care Med. 162: 69–73.
36. PHILLIPS, M., R.N. CATANEO, A.R.C. CUMMIN, et al. 2003. Detection of lung cancer
with volatile markers in the breath. Chest 123: 2115–2123.
37. KANOH, S., H. KOBAYASHI & K. MOTOYOSHI. 2005. Exhaled ethane: an in vivo
biomarker of lipid peroxidation in interstitial lung diseases. Chest 128: 2387–
2392.
38. PHILLIPS, M., J.P. BOEHMER, R.N. CATANEO, et al. 2004. Heart allograft rejection:
detection with breath alkanes in low levels (the HARDBALL study). J. Heart
Lung Transplant. 23: 701–708.
39. TONZETICH, J. 1995. Preface. In Bad Breath Research Perspectives. M. Rosenberg,
Ed.: xi–xviii. Ramot Publishing. Tel Aviv, Israel.
40. SCULLY, C., M. EL-MAAYTAH, S.R. PORTER & J. GREENMAN. 1997. Breath odor:
etiopathogenesis, assessment and management. Eur. J. Oral Sci. 105: 287–
293.
41. FELLER, L. & E. BLIGNAUT. 2005. Halitosis: a review. SADJ. 60: 17–19.
42. MESSADI, D.V. 1997. Oral and non-oral sources of halitosis. J. Calif. Dent. Assoc.
25: 127–131.
WHITTLE et al. 265

43. LOESCHE, W. & C. KAZOR. 2002. Microbiology and treatment of halitosis. Peri-
odontol. 2000 28: 256–279.
44. KOSTELC, J.G., G. PRETI, P.R. ZELSON, et al. 1980. Salivary volatiles as indicators
of periodontitis. J. Periodontal Res. 15: 185–192.
45. PAYNE, R.K., J.N. LABOWS & X. LIU. 1999. Released oral malodors measured by
solid phase microextraction/gas chromatography mass spectrometry (HS-SPME-
GC-MS). ACS Symp. Series 76: 373–386.
46. CHAMBERLIN, M.E., U. TSUNEYUKI, S.H. MUDD, et al. 1966. Demyelination of the
brain is associated with methionine adenosylation I/II deficiency. J. Clin. Invest.
98: 1021–1027.
47. FRIEDMAN, M.I., G. PRETI, R.O. DEEMS, et al. 1994. Limonene in expired lung air
of patients with liver disease. Digest. Dis. Sci. 39: 1672–1676.
48. CHEN, S., L. ZIEVE & V. MAHADEVAN. 1970a. Mercaptans and dimethyl sulfide in
the breath of patients with cirrhosis of the liver. J. Lab. Clin. Med. 75: 628–635.
49. TJOA, S. & P.V. FENNESSEY. 1991. The identification of trimethylamine excess in
man: quantitative analysis and biochemical origins. Anal. Biochem. 197: 77–82.
50. HUMBERT, J.R., K. B. HAMMOND, W.E. HATHAWAY, J.G. MARCOUX & D. O’BRIEN.
1970. Trimethylaminuria: the fish-odour syndrome. Lancet 2: 770–771.
51. CASHMAN, J.R., K. CAMP, S.S. FAKHARZADEH, et al. 2003. Biochemical and clinical
aspects of the human flavin-containing monooxygenase form 3 (FMO3) related
to trimethylaminuria. Curr. Drug Metab. 4: 151–170.
52. LATTARD, V., J. ZHANG, Q. TRAN, et al. 2003. Two new polymorphisms of the FMO3
gene in Caucasian and African-American populations: comparative genetic and
functional studies. Drug Metab. Disp. 31: 854–860.
53. ZHANG, A.Q., S. MITCHELL & R. SMITH. 1995. Fish odour syndrome: verification
of carrier detect test. J. Inherit. Metab. Dis. 18: 669–674.
54. ZSCHOCKE, J., D. KOHLMUELLER, E. QUAK, et al. 1999. Mild trimethylaminuria
caused by common variants in FMO3 gene. Lancet 354: 834–835.
55. CASHMAN, J.R. & J. ZHANG. 2002. Interindividual differences of human flavin-
containing monooxygenase 3: genetic polymorphisms and functional variation.
Drug Metab. Disp. 30: 1043–1052.
56. LEOPOLD, D.A., G. PRETI, M.M. MOZELL, et al. 1990. Fish-odor syndrome present-
ing as dysosmia. Arch. Otolaryngol. Head Neck Surg. 116: 354–355.
57. WALKER, V. 1993. The fish odour syndrome: the problems are psychosocial. Br.
Med. J. 307: 539.
58. MILLS, G.A., V. WALKER & H. MUGHAL. 1999. Quantitative determination of
trimethylamine in urine by solid-phase microextraction and gas chromatography-
mass spectrometry. J. Chromat. B. 723: 281–285.
59. TONZETICH, J. 1971. Direct gas chromatographic analysis of sulphur compounds
in mouth air in man. Arch. Oral Biol. 16: 587–597.
60. KOSTELC, J.G., G. PRETI, P.R. ZELSON, et al. 1984. Oral odors in early experimental
gingivitis. J. Periodontal Res. 19: 303–312.
61. PRETI, G., L. CLARK, B.J. COWART, et al. 1992. Non-oral etiologies of oral malodor.
J. Periodontol. 63: 790–796.
62. DOLPHIN, C.T., J.H. RILEY, R.L. SMITH, et al. 1997. Structural organization of the
human flavin-containing monooxygenase 3 gene (FMO3), the favored candidate
for fish-odor syndrome, determined directly from genomic DNA. Genomics 46:
260–267.
63. BOOTH, G. & S. OSTENSON. 1966. Acetone in alveolar air and the control diabetes.
Lancet 2: 1102–1105.
266 ANNALS OF THE NEW YORK ACADEMY OF SCIENCES

64. WALSH, J. 2004. A breath of fresh air in diabetes detection. Spectroscopy 19: 50.
65. CARPAGNANO, G.E., S.A. KHARITONOV, O. RESTA, et al. 2002. Increased 8-
isoprostane and interleukin-6 in breath condensate of obstructive sleep apnea
patients. Chest 122: 1162–1167.
66. LECHNER, M., A. KARLSEDER, D. NIEDERSEER, et al. 2005. H. pylori infection
increases levels of exhaled nitrate. Helicobacter 10: 385–390.
67. PATHAK, C.M., D.K. BHASIN, D. PANIGRAHI & R.C. GOEL. 1994. Evaluation of
14
C-urinary excretion and its comparison with 14 CO2 in breath after 14 C-urea
administration in Helicobacter pylori infection. Am. J. Gastroenterol. 89: 734–
738.
68. SYLVESTER, K.P. 2005. Exhaled carbon monoxide levels in children with sickle
cell disease. Eur. J. Pediat. 164: 162–165.
69. HANAZAWA, T., S.A. KHARITONOV, W. OLDFIELD, et al. 2000. Nitrotyrosine and
cysteinyl leukotrienes in breath condensates are increased after withdrawal of
steroid treatment in patients with asthma. Am. J. Resp. Crit. Care Med. 161:
A919.
70. CAP, P., J. CHLADEK, F. PEHAL, et al. 2004. Gas chromatography/mass spectrometry
analysis of exhaled leukotrienes in asthmatic patients. Thorax 59: 465–470.
71. GORDON, S.M., J.P. SZIDON, B.K. KROTOSZYNSKI, et al. 1985. Volatile organic
compounds in exhaled air from patients with lung cancer. Clin. Chem. 31: 1278–
1282.
72. PRETI, G., J.N. LABOWS, J.G. KOSTELC, et al. 1988. Analysis of lung air from
patients with bronchogenic carcinoma and controls using gas chromatography-
mass spectrometry. J. Chromat. B: Biomed. Appl. 432: 1–11.
73. DEKHUIJZEN, P.N., K.K. ABEN, I. DEKKER, et al. 1996. Increased exhalation of
hydrogen peroxide in patients with stable and unstable chronic obstructive pul-
monary disease. Am. J. Resp. Crit. Care Med. 154: 813–816.
74. LIU, J. & P.J. THOMAS. 2005. Exhaled breath condensate as a method of sampling
airway nitric oxide and other markers of inflammation. Med. Sci. Monitor 11:
MT53–MT62.
75. MONTUSCHI, P., S.A. KHARITONOV, G. CIABATTONI, et al. 2000. Exhaled 8-
isoprostane as a new non-invasive biomarker of oxidative stress in cystic fibrosis.
Thorax 55: 205–209.
76. BODINI, A., C. D’ORAZIO, D. PERONI, et al. 2005. Biomarkers of neutrophilic
inflammation in exhaled air of cystic fibrosis children with bacterial airway
infections. Pediat. Pulmonol. 40: 494–499.
77. KAJI, H., M. HISAMURA, N. SATO & M. MURAO. 1978. Gas chromatographic de-
termination of volatile sulfur compounds in the expired alveolar air in the hep-
atopathic subjects. J. Chromat. B: Biomed. Appl.
78. TANGERMAN, A., M.T. MEUWESE-ARENDS & J.B. JANSEN. 1994. Cause and compo-
sition of foetor hepaticus. Lancet 343: 730.
79. SIMENHOFF, M.L., J.F. BURKE, J.J. SAUKKONEN, et al. 1977. Biochemical profile of
uremic breath. N. Engl. J. Med. 247: 132–135.
80. CHRISTENSEN, C. & M.A. NAVAZESH. 1982. A comparison of whole mouth resting
and stimulated salivary measurement procedures. J. Dent. Res. 61: 1158–1162.