Vol.2, No.2, 68-78 (2010) Natural Science
Copyright © 2010 SciRes. OPEN ACCESS
Disinfection of swimming pools with chlorine and
derivatives: formation of organochlorinated and
organobrominated compounds and exposure of pool
personnel and swimmers
Maria-Cristina Aprea, Bruno Banchi, Liana Lunghini, Massimo Pagliantini, Antonio Peruzzi,
Gianfranco Sciarra
Laboratorio di Sanità Pubblica Area Vasta Toscana Sud Est, Azienda USL 7, Siena, Italy; maaprea@tin.it, c.aprea@usl7.toscana.it
Received 12 August 2009; revised 11 September 2009; accepted 5 January 2010.
Chlorination of pool water leads to the forma-
tion of many by-products, chloroform usually
being the most abundant. The paper reports the
results of a study evaluating exposure of bath-
ers and pool employees to trihalomethanes
(chloroform, bromodichloromethane, dibromo-
chloromethane, bromoform) in four indoor
swimming pools with chlorinated water. Chlo-
roform concentrations in environmental air
samples when the pool was in use (about 9 h),
in the range 1-182 µg/m3, were greater near the
pool than in the change rooms, passageways
and offices. Chloroform concentrations in per-
sonal air samples of pool employees were in the
range 18-138 µg/m3. Urinary concentrations of
chloroform averaged (geometric means) 0.123
and 0.165 µg/l and 0.404 and 0.342 µg/l prior and
at the end of exposure during in water and out
of water activities, respectively. The significant
increase in urinary excretion of chloroform
confirms that the source of the contaminant was
pool water. Absorption of chloroform, estimated
from airborne and water concentrations, was
significantly correlated with delta chloroform
(after/before exposure) and urinary concentra-
tions of chloroform at the end of exposure. As
chloroform is a toxic and possibly carcinogenic
substance, these observations pose a problem
principally for the general population of pool
Keywords: Disinfection By-Products;
Indoor Swimming Pool; Trihalomethanes;
Biological Monitoring; Exposure; Urine
This paper is concerned with bathing complexes con-
sisting of one or more artificial pools for recreational,
educational, sporting or therapeutic activity carried out
in the water. From the health point of view, pools are
classified according to environmental and structural
characteristics and type of use. Various documents de-
fining guidelines for safe use of recreational facilities
such as pools have been published [1-3]. Chemical haz-
ards associated with frequentation of pools are summa-
rized in Figure 1.
Chemical agents in pool water depend on the type of
water used to fill the pool. Town water, for example,
may contain organic matter and by-products of disinfec-
tion from previous treatments. Among the chemical
agents derived from bathers, nitrogen compounds, espe-
cially ammonia, react with free disinfectants to form
various by-products. Nitrogen compounds may come
from skin secretions: the nitrogen content of sweat is
about 1 g/l as ammonia, amino acids, creatinine and urea.
Significant quantities of nitrogen compounds can come
from urine: urine release by bathers averages 25-30
ml/person [4] but may exceed 77.5 ml/person [5]. No
information is currently available about concentrations
of compounds from cosmetics. With regard to chemical
agents from maintenance, a considerable number of
compounds are used to keep water quality acceptable.
Disinfectants are added in order to disactivate patho-
genic microorganisms. Chlorine in one of its various
forms is the most common. Other disinfectants such as
ozone and UV radiation kill or inactivate microorgan-
isms at the time of treatment but do not have any resid-
ual effect that continues to act in the water. They are
therefore used with chlorine or bromine to provide con-
tinuous disinfection. Chlorine dioxide is not considered a
chlorine disinfectant as it acts differently without pro-
ducing residual chlorine, through conversion to chlorite
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Derived from maintenance
Disinfectancts, Flocculants, Algicides
Buffer chemicals
Derived from bathers
Urine, Perspiration, Dirt, Lotions
(sun screens, cosmetics, residues of soap)
Derived from water
By-products of water treatment
By-products of water treatment
Trihalomethanes , Haloacetic acids ,
Clorathes, Nitrogen trichloride
Figure 1. Summary of possible sources of chemical contami-
nation in swimming pools.
and chlorate ions that remain in solution. Liquid bromine
is seldom used, whereas sodium bromide and its oxidant
(hypochlorite) are more common. Disinfection with
bromine compounds is not suitable for outdoor pools
because sunlight destroys bromine residues. In all cases,
the choice is based on efficacy in the particular circum-
stances of use, as well as ease of handling and monitor-
ing. Compounds used to correct pH depend on the type
of disinfectant and its acidity/alkalinity. Alkaline disin-
fectants such as sodium hypochlorite only require addi-
tion of an acid, which is generally sodium hydrogensul-
phate, carbon dioxide or hydrochloric acid. Acid disin-
fectants such as chlorine require addition of an alkaline
substance which is generally sodium carbonate solution.
At correct doses with maintenance of pH between 7.2
and 8.0, disinfectants should not have adverse effects on
health. Flocculants such as polyaluminium chloride can
be used to facilitate removal of dissolved or suspended
substances and colloids. They trap the substances in
flocculate that can be removed by filtration.
The formation of by-products of disinfection is related
to the reaction of disinfectants with other chemical sub-
stances in the water. The most abundant by-products are
trihalomethanes, such as chloroform, the most abundant,
together with haloacetic acids of which di- and tri-
chloroacetic acids are the most abundant [6]. The pres-
ence of inorganic bromides in the water can induce for-
mation of bromine after oxidation, which can participate
in the formation of by-products such as brominated tri-
halomethanes. Use of ozone in the presence of bromides
can lead to formation of bromates that can build up in
the water if turnover is poor. Limited information is
available on ozonation and its by-products. Ozone can
react with oraganic substances to produce oxygenated
compounds such as aldehydes and carboxylic acids.
Chlorine and bromine react very quickly with ammonia
forming chloramines and bromoamines. Little data is
available on the impact of UV on disinfection by-prod-
ucts when used with other chemicals, but UV does not
seem to form by-products and appears to significantly
reduce chloramine levels.
Exposure of pool personnel and bathers may occur by
ingestion of water, inhalation of aerosol or vapours and
cutaneous absorption. The quantity of water ingested by
swimmers depends on various factors, including experi-
ence, gender, age and type of activity. Estimates show
that water intake is higher in children (37 ml) than adults
(16 ml), and in men (22 ml) than women (12 ml) [7].
Bathers inhale the air in contact with the water surface.
The volume of air inhaled depends on the intensity of
physical activity and exposure time. Exposure by inhala-
tion regards substances in vapour form released by the
water and aerosols created also by swimmer-induced
splashing and stirring of the water. Concentrations at
different levels in the air above the pool depends on fac-
tors such as ventilation, the size of the building and air
circulation. Skin, including eyes and mucous membranes,
is extensively exposed to chemical agents in pool water.
The intensity of skin absorption depends on a series of
factors including contact time, water temperature and
concentration of toxic compounds.
Many by-products of disinfection have proven to be
mutagenic, genotoxic, carcinogenic, fetotoxic, hepato-
toxic, renotoxic, neurotoxic and dysmetabolic [8]. Chlo-
roform and bromodichloromethane are classified by the
International Agency for Research on Cancer (IARC) as
possible carcinogens for humans (group 2B) [9],
whereas the American Conference of Governmental In-
dustrial Hygienists (ACGIH) considers chloroform to be
carcinogenic for animals with unknown relevance to
humans (Class A3) [10].
Concentrations of trihalomethanes in pool water
[11-32] and in the air above the pool [11,14-16,18,26,
27,33,34] was examined in several studies. Some authors
investigated also the absorption of trihalomethanes dur-
ing time spent at the pool by measuring blood concentra-
tions of chloroform or those in alveolar air [14-16,29,
The aim of the present study was to assess exposure
levels of swimmers and pool personnel to chlorinated
and brominated organic compounds in public indoor
pools. Airborne concentrations were determined in dif-
ferent parts of the pool premises, and when possible, by
personal air sampling and determination of urinary ex-
cretion before and after exposure. Levels of the same
compounds were also determined in water, as well as
microclimatic and plant conditions. Absorption of chlo-
roform estimated from airborne and water concentrations
were compared with the increase in concentrations in
urine during exposure.
2.1. Microclimatic and Plant Conditions
Four public indoor pools were monitored. All used
drinking water from the town water supply. On days of
monitoring, one or more water samples were taken for
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
determination of brominated and chlorinated organic
compounds (chloroform, bromodichloromethane, di-
bromochloromethane, bromoform).
Pool 1: pool volume 470 m3 plus compensation tank
20 m3, disinfectant calcium hypochlorite 65% and occa-
sionally sodium dichloroisocyanate, air intake 18000
m3/h with aspiration of 15000 m3/h (turnover about
83%). The plant was monitored five times in 2006-2008.
On sampling days, mean relative humidity was 51%-
68%, mean air temperature 23.5-28.5°C. Mean air speed
was 0.10-0.23 m/sec. Water temperature was 28.6-29°C,
pH 7.3-7.8, nitrates 8.3-15 mg/l, isocyanic acid 40-75
mg/l, turbidity 0.2-0.3 mg/l, suspended solids 0.9 mg/l
and residual free chlorine 0.29-1.12 mg/l. Maximum
number of users per hour was 50-60 and total daily users
Pool 2: pool volume 476.3 m3 plus compensation tank
24 m3, disinfectant sodium hypochlorite, air intake
30000 m3/h (turnover about 40%). The plant was moni-
tored three times in 2007-2008. Mean relative humidity
on sampling days was 70-75%, mean air temperature
23.2-24.5°C, mean air speed 0.06-0.07 m/sec, water
temperature 28.4-28.6°C, pH 6.9-7.3, nitrates 0.7-1.1
mg/l, isocyanic acid <20 mg/l, turbidity 0.1-0.6 mg/l,
suspended solids <1 mg/l and residual free chlorine
0.81-2.59 mg/l. Maximum number of users per hour
60-100, total daily users 220-240 persons.
Pool 3: pool volume 700 m3 plus compensation tanks
60 m3 disinfectant sodium dichloro-S-triazine-trione
(Dichloro 63), air intake 40,000 m3/h without circulation.
The plant was monitored twice in 2007-2008. On sam-
pling days, mean relative humidity was 70-75%, mean
air temperature 24.5-25.5°C, mean air speed 0.03-0.05
m/sec. Water temperature in adult pool 29-30°C, pH 6.9,
nitrates 14 mg/l, isocyanic acid 20 mg/l, turbidity 0.3
mg/l, suspended solids < 1 mg/l, vinyl chloride 5-8 µg/l
and residual free chlorine 0.8-1.5 mg/l. Total number of
users per day 250-300.
Pool 4: pool volume 400 m3 plus compensation tank
24 m3, disinfectant sodium dichloro-S-triazine-trione and
sodium hypochlorite, air input not available, turnover
about 30%. The plant was monitored twice in 2007-2008.
On sampling days, mean relative humidity was 83-84%,
mean air temperature 24°C, mean air speed 0.05 m/sec,
water temperature 29-30°C, pH 7.5, nitrates 11.6 mg/l,
isocyanic acid 25 mg/l, turbidity 0.5 mg/l, suspended
solids < 1 mg/l, vinyl chloride 0.05-0.12 µg/l and resid-
ual free chlorine 1.5-1.8 mg/l. Total number of users per
day 200.
2.2. Study Population
In the four bathing complexes, six lifeguards and four
instructors were monitored: the former worked at the
poolside and the latter in the water. Thirty-one bathers
underwent biological monitoring (15 swimmers at dif-
ferent levels of expertise, four competitive swimmers
and 12 persons enrolled water gym sessions). All filled
in a questionnaire about personal details, weight, height,
smoking and drinking habits, occupation. This informa-
tion was used in the statistical analysis of the results.
Before enrolment in the study, all subjects gave their
informed consent.
2.3. Personal and Environmental Air
Personal air sampling during the work-shift was per-
formed for poolside personnel by means of radial diffu-
sion air samplers for chloroform assay (Radiello®). In
the measurements conducted in 2008, parallel active
sampling with carbon vials was carried out at a flow rate
of 100 ml/min to determine chloroform, bromodichloro-
methane, dibromochloromethane and bromoform. Dou-
ble sampling was carried out to assay bromine com-
pounds, for which the manufacturer of diffusion sam-
plers does not provide equivalent rates.
Fixed (environmental) sampling was carried out about
1.5 m from the pavement at the edge of the pools (3 or 4
samples per bathing complex per day), in the changing
rooms, offices and passages between the changing rooms
and the pool. Sampling lasting 24 h was carried out with
diffusion samplers (Radiello®) and others lasting about
9 h (when the pool was in use) were done using active
carbon vials at an air flow of 100 ml/min. The same
contaminants as for personal samples were assayed. In
Pool 3, the 9-h sampling was divided into two periods of
4.5 h (morning and afternnon) in order to detect any
changes in concentrations of the contaminants in the
various areas. In this case the data was used as such and
after calculation of the weighted mean concentration
over the whole period the pool was open.
For the analytical determination, samples were added
with carbon disulfide containing deuterated benzene as
internal standard and left in contact with the solvent for
30 min. The extract was injected in the GC/MS appara-
tus (EI-SIM electronic impact, single ion monitoring).
The analytical limit of detection (LOD) was 0.1 µg/
2.4. Estimation of Absorption of Chloroform
Absorption of chloroform was estimated for staff and
bathers, summing the fractions derived from direct in-
gestion of water, inhalation of aerosols and vapours, and
transcutaneous absorption. For absorption by ingestion,
the results of Evans et al. 2001 [7] were used. According
to the latter, water intake averages 22 ml/h for men 12
ml/h for women. Knowing the concentration of chloro-
form in pool water, it was possible to calculate the quan-
tity ingested, assuming 100% absorption. The fraction
derived from ingestion was assumed to be zero for pool-
side staff.
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
The fraction derived from inhalation was estimated on
the basis of time spent in the water, lung ventilation and
median concentration of airborne contaminant at the
poolside, corrected by a factor of 1.8 because the meas-
urements were made 1.5 m from the pavement instead of
20 cm from the water’s surface [11]. A lung retention of
59% was assumed, as proposed by Kuo et al. [37]. Lung
ventilation was assumed to be 15 l/min for males and 12
l/min for females for tasks involving little exertion (life-
guard) and 30 l/min for males and 25 l/min for females for
activity in the water. For poolside personnel, the concen-
tration found by personal air sampling was used.
The fraction derived from skin absorption was esti-
mated on the basis of time spent in the water, chloroform
concentration in pool water, body surface area estimated
on the basis of weight and height using the Du Bois
formula [38] and 80% contact of the skin with water.
The permeation constant of skin was assumed to be 0.2
cm/h as proposed by Kuo et al. [37]. The fraction de-
rived from skin absorption was assumed to be zero for
poolside personnel.
2.5. Urine Sampling
Spot samples of urine were obtained from personnel and
bathers before and after exposure. Chloroform, bro-
modichloromethane, dibromochloromethane and bro-
moform were determined in all samples. The determina-
tion was performed analyzing the head space in GC/MS
EI-SIM using deuterated benzene as internal standard.
The LOD was 0.050 µg/l.
The data, expressed in µg/l was used as such and as dif-
ferences between concentrations before and after exposure.
2.6. Statistical Analysis
Statistical analysis was done using Stat View 5.0, Power
PC Version (SAS Institute Inc.). Values below the ana-
lytical limit of detection (LOD) were analyzed as half
the LOD when at least half the data was over the LOD.
Values above LOD but not quantifiable (<LOQ) were
analyzed as the mean of LOD and LOQ. Parametric tests
were used (analysis of variance, regression analysis,
Student’s t test for paired and unpaired data) and the
level of significance chosen was ά = 0.05.
Concentrations of trihalomethanes in pool water on
sampling days are shown in Table 1. Concentrations of
contaminants in the water depended on the type of dis-
infectants used, on any impurities in the water used to
fill the pool and on water characteristics. Brominated
compounds seemed to be associated more with use of
sodium hypochlorite than with compounds such as cal-
cium hypochlorite or Dichloro 63, irrespective of isocy-
anates. Chloroform was confirmed to be the most abun-
dant trihalomethane, and was only equal in concentration
to bromodichloromethane and dibromochloromethane in
pool 2, disinfected with sodium hypochlorite alone.
Table 1. Concentrations of trihalomethanes (µg/l) in water of four public pools.
Chloroform BDCM DBCM Bromoform
Pool 1
Mean ± SD
Pool 2
Mean ± SD
Pool 3
Mean ± SD
Pool 4
Mean ± SD
BDCM= Bromodichloromethane. DBCM= Dibromochloromethane. GM = geometric mean.
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Table 2. Descriptive statistics of concentrations of chloroform (µg/m3) detected in environmental air samples in four public pools.
Poolside Change rooms and offices Passageways
9 h (a) 24 h (b) 9 h (a) 24 h (b) 9 h (a) 24 h (b)
N 26 40 15 23 3 5
N <LOD 0 0 7 2 1 0
Mean ± SD 85±50 52±30 11±12 8±7 34±29 25±9
Median 65 46 4 5 50 20
GM 70 43 5 6 14 24
Min-Max 21-182 12-127 1-34 1-29 1-52 18-36
Poolside Change rooms and offices Passageways
9 h (a) 24 h (b) 9 h (a) 24 h (b) 9 h (a) 24 h (b)
N 9 15 3 5 3 5
N <LOD 0 0 1 0 1 0
Mean ± SD 124±41 81±24 14±11 7±3 34±29 25±9
Median 128 78 18 6 50 20
GM 118 77 7 6 14 24
Min-Max 66-182 39-127 1-22 3-10 1-52 18-36
Poolside Change rooms and offices Passageways
9 h (a) 24 h (b) 9 h (a) 24 h (b) 9 h (a) 24 h (b)
N 9 9 6 6 0 0
N <LOD 0 0 3 0 - -
Mean ± SD 35±15 25±10 5±6 5±1 - -
Median 32 23 2 5 - -
GM 33 23 3 5 - -
Min-Max 21-62 12-39 1-14 4-7 - -
Poolside Change rooms and offices Passageways
9 h (a) 24 h (b) 9 h (a) 24 h (b) 9 h (a) 24 h (b)
N 4 8 3 6 0 0
N <LOD 0 0 0 0 - -
Mean ± SD 132±11 56±14 29±4 18±8 - -
Median 130 53 27 16 - -
GM 131 55 29 17 - -
Min-Max 120-147 39-76 26-34 11-29 - -
Poolside Change rooms and offices Passageways
9 h (a) 24 h (b) 9 h (a) 24 h (b) 9 h (a) 24 h (b)
9 ore (a) 24 ore (b) 9 ore (a) 24 ore (b) 9 ore (a) 24 ore (b)
N 4 8 3 6 0 0
N <LOD 0 0 3 2 - -
Mean ± SD 61±4 26±12 - 4±4 - -
Median 63 26 - 3 - -
GM 61 24 - 2 - -
Min-Max 55-64 14-40 - 1-10 - -
(a) Sampling conducted for 9 h in the presence of bathers; (b) Sampling conducted for 24 h in the presence and absence of bathers.
Chloroform concentration detected in environmental
samples are summarized in Table 2, where the poolside,
change room, office and passageway data is presented
separately. The table also shows 9-h and 24-h sampling
data separately. The sampling site and bathing complex
significantly affected chloroform concentrations when
the pools were open and over 24 h.
For 9-h sampling, the model explained 76% of the
variance, with bathing complex explaining 31% and
sampling site 45%, whereas for 24-h sampling the cor-
responding percentages were 70%, 28% and 42%.
Twenty-four-hour values were less than those measured
when the pools were open, especially at the poolside,
indicating that movement of the water increased airborne
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Figure 2. Concentrations of chloroform in morning, after-
noon and 24-h environmemtal samples at pool 3.
Confirming this, Figure 2 shows chloroform concen-
trations found in environmental samples at pool 3: sam-
pling when the pool was open in the morning and after
noon gave values much higher than 24-h data at the
poolside, whereas those obtained in changing rooms and
offices did not vary.
Chloroform concentrations measured at the poolside
when the pool was in use showed a statistically signifi-
cant correlation (p<0.0001) with chloroform concentra-
tions in the water and with the number of pool users per
day (N). Multiple regression analysis showed that 74%
of the variance was explained according to the following
CHCl3 air (µg/m3) = -101.6 + 1.70 CHCl3 water (µg/l) + 0.57 N
Temperature did not contribute significantly to the re-
Chloroform concentrations in personal air samples in
the three bathing complexes where they were obtained
are summarized in Table 3. As expected, they were
lower than at the poolside because these personnel did
not spend the whole shift beside the pools but also spent
time in offices etc. where airborne concentrations of the
contaminant were often less.
With regard to airborne concentrations of brominated
compounds, bromoform was never detected whereas
bromdichloromethane (BDCM) and dibromochloro-
methane (DBCM) were only detectable in poolside sam-
ples of pools 2 and 4, where significant concentrations of
the same contaminants were also found in the water
(Table 4). The amount of data was therefore insufficient
for multiple regression analysis between airborne and
water concentrations. However, it can be said that the
chemicophysical characteristics of contaminants strongly
affect dispersal dynamics. In other words, for a given
water concentration, chloroform passes much more read-
ily into the vapour phase (vapour pressure 21.2 kPa at
20°C) than bromodichloromethane (vapour pressure 6.6
kPa at 20°C) and other brominated compounds that have
even lower vapour pressures.
Concentrations of BDCM, DBCM and bromoform
were undetectable in all urine samples. Descriptive sta-
tistics of chloroform concentrations detected in urine
before and after exposure are summarized in Table 5
which also shows differences in concentration (delta)
between the two times. The data is separated for persons
carrying on activity in the water, for whom inhalation,
ingestion and cutaneous absorption are likely, and per-
sonnel working out of the water, for whom only respira-
tory exposure is likely.
A quick look at Table 5 shows that delta after/before
was higher for subjects carrying on activity in the water,
confirming the hypothesis of skin and digestive absorp-
tion. In Table 6, the difference in concentrations af-
ter/before exposure is shown in a differentiated manner
depending on the type of activity and/or the pool fre-
quented. Despite the small number of data items avail-
able the table shows that for a given activity, delta af-
ter/before depended on the bathing complex and there-
fore on water and airborne concentrations of this con-
Table 3. Descriptive statistics of concentrations of chloroform (µg/m3) detected in personal air samples.
N 17 8 6 3
N<LOD 0 0 0 0
Mean ± SD 68±35 89±35 61±17 25±10
Median 61 95 61 20
GM 58 82 59 23
Min-Max 18-138 37-138 34-84 18-36
Table 4. Descriptive statistics of concentrations di bromodichloromethane (BDCM) and dibromochloromethane (DBCM) (µg/m3)
detected in poolside air samples of the two pools in which at least 50% of the data was detectable.
N 9 9 4 4
N<LOD 2 3 0 4
Mean ± SD 16±10 8±6 10±3 -
Median 18 8 9 -
GM 10 5 9 -
Min-Max 1-27 1-17 7-13 -
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Table 5. Descriptive statistics of concentrations of chloroform (µg/l) detected in urine.
Type of exposure (a) Start of exposure
End of exposure
(after) Delta after/before
N In water
Out of water
N<LOD In water
Out of water
Mean ± SD In water
Out of water
Median In water
Out of water
GM In water
Out of water
Min-Max In water
Out of water
(a) Exposure in water: swimmers, competitive swimmers, water gym participants, instructors; Exposure out of water: lifeguards and attendants.
Table 6. Differences between concentrations at the end and start of exposure (µg/l) detected in urine during activity in and out of the
water by pool and specific activity.
N Mean ± SD Median GM Min-Max
Competitive swimming (a) 4 0.065±0.047 0.074 - 0-0.113
Total swimmers
Pool 1
Pool 4
Water gym course
Pool 3
Pool 4
Pool 1
Pool 3
Pool 1
Pool 3
Pool 4
(a) only pool 3.
Table 7. Absorbed chloroform (µg) estimated on the basis of concentrations in water and in poolside air while the pool was open.
N Mean ± SD Median GM Min-Max
Total doses 41 358.9±301.9 173.6 267.2 109.4-1248.4
Doses ingested 35 1.4±1.8 0.35 0.66 0.12-5.82
Transcutaneous doses 35 194.6±217.5 85.5 108.4 25.4-844.0
Respiratory doses 41 191.5±132.3 160.1 158.7 89.9-716.7
% ingested dose 35 0.3±0.2 0.2 0.2 0.1-0.7
% cutaneous dose 35 42.5±17.1 45.7 39.2 17.1-78.6
% respiratory dose 35 57.2±17.2 54.1 54.3 17.2-77.9
Table 7 shows estimated absorbed doses of chloro-
form and the percentages of the total constituted by di-
gestive, skin and inhalatory doses. Our estimates pro-
duced absorption values up to about 1.25 mg for swim-
mers or instructors in the water for long periods (2-3 h).
For those carrying on activity in the water, the ingested
percentage of the total dose was negligible compared to
respiratory and skin doses that were 57% and 43%, re-
Estimates of absorbed dose were analysed by linear
regression model with chloroform delta after/before ex-
posure and urinary concentrations of chloroform at the
end of exposure. The results are shown in Figures 3 and
4. Both regressions were highly significant and the vari-
ance explained by the model was 53% and 71%. The
intercept with the ordinate was very close to zero in Fig-
ure 3, as expected, and at 0.084 mg/l in Figure 4. In the
latter case, the intercept should indicate the urinary con-
centration of chloroform not due to time spent in the
pool (pre-exposure value) and indeed it was close to the
median for urinary chloroform at the beginning of ex-
posure for subjects carrying on activity in the water,
shown in Table 5 (this data was also the most numerous).
The better correlation obtained in Figure 4 between
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Figure 3. Linear regression analysis between estimated ab-
sorbed doses and delta chloroform after/before exposure (y =
0.00069 x +0.0004, r2 = 0.534, significant p<0.0001).
Figure 4. Linear regression analysis between estimated ab-
sorbed doses and chloroform concentrations in urine at the end
of exposure (y = 0.0011 x + 0.084, r2 = 0.706, significant
estimated absorbed dose of chloroform and urinary ex-
cretion of chloroform at the end of exposure with respect
to the delta for chloroform can probably be ascribed to
the further factor of variability due in the second case to
the pre-exposure value of chloroform. The slope of the
two regressions was very low, probably due to the fact
that chloroform eliminated in urine at the end of expo-
sure is only a limited part of the total absorbed, whereas
a greater fraction is presumably eliminated by exhalation
and stored in body fat.
Concentrations of trihalomethanes reported in pool water
vary from study to study but the results are not dissimilar
to ours: Sandel [12] examined data from 114 home pools
in the USA, obtaining a mean concentration of chloro-
form of 67.1 μg/l and a maximum of 313 μg/l. Most
other available data on trihalomethanes in pool water is
summarized in Table 8. In the pools monitored by us,
formation of brominated trihalomethanes seemed preva-
lently associated with the use of sodium hypochlorite for
water treatment. Ignoring bromide ions in the water used
to fill these pools (town water in the case of pools 3 and
4), the presence of these compounds is presumably due
to bromide impurities in the treatment reagents. This
evidence makes it important to use only high purity re-
agents to treat town water.
Water-air transport of trihalomethanes depends on a
number of factors that include concentrations in pool
water, temperature and water disturbance and splashing
by bathers. In our study, the concentration of chloroform
detected at the poolside showed a good correlation with
chloroform concentrations in pool water and with the
number of swimmers present. Air and water tempera-
tures were excluded from the regression model because
they did not seem to have a significant effect on envi-
ronmental concentrations of chloroform. Trihalomethane
concentrations at different levels in the air above the
pool should also depend on factors such as ventilation,
size of pool building and air circulation. Most of the data
available on concentrations of trihalomethanes in air
above the pools is summarized in Table 9, which shows
that measurements taken 20 cm above the water were on
average 1.8 times higher than those taken 150 cm above
the water.
In our study, concentrations of airborne trihalome-
thanes depended on where the measurements were taken
(poolside, change rooms and offices, corridors) as found
in other studies: Fantuzzi et al. [34] studied total triha-
lomethane concentrations in five Italian indoor pools,
finding mean concentrations in poolside air of 58.0 ±
22.1 μg/m3 and 26.1 ± 24.3 μg/ m3 at the reception.
Absorption of trihalomethanes during time spent at
the pool was investigated by comparing urinary excre-
tion of chloroform before and after exposure. Levels
observed at the start of exposure were slightly less than
detected in the general population. A study conducted in
Italy in 1994 [39] found median concentrations of 194
ng/l in the rural population (115 subjects) and 490 ng/l in
the urban population (87 subjects). Most previous stud-
ies on absorption of trihalomethanes at swimming pools
measured blood concentrations of chloroform or those in
alveolar air. Strähle et al. [35] compared concentrations
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
of trihalomethanes in blood of swimmers with those in
pool water and air. The results, summarized in Table 10,
demonstrate that inhalation is probably the main route of
absorption of volatile components, since concentrations
in water of indoor pools are greater than those of outdoor
pools, while concentrations in ambient air are higher
indoors, as are blood concentrations. Good ventilation of
pool premises should therefore significantly reduce ex-
posure. Erdinger et al. [29] confirmed that exposure is
prevalently respiratory, showing a ratio of 3:1 with re-
spect to skin absorption. Aggazzotti et al. [14-16,36]
showed that exposure in chlorinated pools can cause an
increase in trihalomethane concentrations in plasma and
alveolar air, but the latter declines soon after leaving the
pool. Plasma concentrations of chloroform were detect-
able in 100% of the 127 samples analyzed, showing a
mean concentration of 1.06 µg/l, whereas BDCM, de-
tectable in only 25 samples, showed a mean of 0.14 μg/l
and DBCM, detectable in only 17 samples, showed a
mean of 0.1 μg/l.
Table 8. Concentrations of trihalomethanes in pool water (µg/l).
Chloroform BDCM DBCM Bromoform
country Mean Range Mean Range Mean Range Mean Range
Type of poolRef.
Poland 35.9-99.7 2.3-14.7 0.2-0.8 0.2-203.2 Indoor [13]
19-94 [14]
93.7 9-179 [15] Italy
33.7 25-43 2.3 1.8-2.8 0.8 0.5-10 0.1 0.1
37.9 Indoor [17]
4-402 1-72 <0.1-8 <0.1-1 Outdoor USA
3-580 1-90 0.3-30 <0.1-60 Indoor
14.6 2.4-29.8 Indoor
43 14.6-111 Outdoor
198 43-980 22.6 0.1-150 10.9 0.1-140 1.8 0.1-88 Indoor [20]
0.5-23.6 1.9-16.5 <0.1-3.4 <0.1-3.3 Indoor
3.6-82.1 1.6-17.3 <0.1-15.1 <0.1-4.0 Outdoor
94.9 40.6-117.5 4.8 4.2-5.4 1.8 0.78-2.6 Indoor [22]
80.7 8.9 1.5 <0.1 Indoor
74.9 11.0 3.0 0.23 Outdoor
3-27.8 Indoor [24]
1.8-28 Indoor [11]
8-11 Indoor [25]
14 0.51-69 2.5 0.12-15 0.59 0.03-4.9 0.16 <0.03-8.1 Indoor
30 0.69-114 4.5 0.27-25 1.1 0.04-8.8 0.28 <0.03-3.4 Outdoor
3.8 6.4 max Indoor [28]
7.1-24.8 Indoor [29]
Denmark 145-151 Indoor [30]
Hungary 11.4 <2-62.3 2.9 <1-11.4 Indoor [31]
UK 121.1 45-212 8.3 2.5-23 2.7 0.67-7 0.9 0.67-2 Indoor [32]
Table 9. Concentrations of trihalomethanes in air above the pool surface (µg/m3).
Chloroform BDCM DBCM Bromoform
country Mean Range Mean Range Mean Range Mean Range
Type of poolRef.
214 66-650 19.5 5-100 6.6 0.1-14 0.2 [15]
140 49-280 17.4 2-58 13.3 4-30 0.2 [14] Italy
169 35-195 20 16-24 11.4 9-14 0.2
Indoor (a)
Canada 597-1630 Indoor [33]
65 9.2 Indoor (a)
36 5.6 3.8 Indoor (b)
5.6 0.21 1.2 Outdoor (a)
2.3 Outdoor (a)
3.3 0.33-9.7 0.4 0.08-2.0 0.1 0.02-0.5 <0.03 Outdoor (a)
1.2 0.36-2.2 0.1 0.03-0.16 0.05 0.03-0.08 <0.03 Outdoor (b)
39 5.6-206 4.9 0.85-16 0.9 0.05-3.2 0.1 <0.03-3.0 Indoor (a)
30 1.7-136 4.1 0.23-13 0.8 0.05-2.9 0.08 <0.03-0.7 Indoor (a)
<0.1-1 <0.1 <0.1 <0.1 Indoor (c)
USA <0.1-260 <0.1-10 <0.1-5 <0.1-14 Outdoor (c)[18]
(a) 20 cm above water surface; (b) 150 cm above water surface; (c) 200 cm above water surface.
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
Table 10. Comparison of concentrations of trihalomethanes (THM) in blood of swimmers after 1 h of exercise, in pool water and in
ambient air of indoor and outdoor pools [35].
THM (mean - range)
indoor pools outdoor pools
Blood of swimmers (μg/l) 0. 48(0.23-0.88) 0.11(<0.06-0.21
Pool water (μg/l) 19.6(4.5-45.8) 73.1(3.2-146)
air 20 cm above water surface (μg/m3) 93.6(23.9-179.9) 8.2(2.1-13.9)
air 150 cm above water surface (μg/m3) 61.6(13.4-147.1) 2.5(<0.7-4.7)
Absorptions estimated by us confirmed that the quan-
tity of chloroform taken up by inhalation was a major
portion of the total dose for bathers and instructors, be-
ing 57% compared to 43% absorbed through the skin.
These estimates are based on the results of other studies
in the literature, from which we obtained lung retention
and skin penetration. They are undoubtedly associated
with errors because respiratory dose is greatly affected
by lung ventilation which was assumed by us without
any precise indications about the real volume of air in-
haled and without considering differences between sub-
jects due to physical exertion and age. As far as we are
aware, no similar estimates have been reported in the
literature and therefore our data forms an excellent basis
for further research, including epidemiological studies.
The data should be implemented in this way to make it
more representative. The good correlation observed with
urinary concentrations at the end of exposure and with
delta after/before exposure confirms that even if our es-
timates were not quantitatively exact, they are highly
indicative of exposure.
This study shows that concentration of trihalomethanes
in pool water vary as a consequence of the type of disin-
fectants used and of the impurities in the treatment re-
agents. Trihalomethanes are lost from the surface of the
water and are found in the air above the pool. Water-air
transport depends on a number of factors that include
concentrations in pool water, temperature and water dis-
turbance by bathers. The sampling site and bathing com-
plex significantly affect air concentrations. Absorption
of trihalomethanes for workers and swimmers, during
time spent at the pool, evaluated by urinary excretion of
the same compounds before and after exposure, is higher
for subjects carrying on activity in the water, confirming
the importance of skin and digestive absorption, al-
though inhalation is on average the major portion of the
total absorbed dose.
The results show that even “healthy” places like pools
can pose chemical agent management problems that are
far from simple. Since the aim of water treatment is to
control biological risk for users, it is senseless to cate-
gorically condemn swimming pools or water chlorina-
tion. “The risks to health from these by-products at the
levels at which they occur in pool water are extremely
small in comparison with both the risks associated with
inadequate disinfection and the enormous health benefits
(including relaxation and exercise) associated with pool
use” [1]. It is therefore to be hoped that more attention
be paid to the design and management of pools, as well
as to correct behaviour, in order to improve our living
and occupational environments.
This study considers all the aspects related to the tri-
halomethanes exposure in indoor swimming pool disin-
fected with chlorine and derivatives, and the results can
be generalized and applied in similar situations.
[1] WHO (2006) Guidelines for safe recreational water en-
vironments. Swimming pools and similar environments,
World Health Organization, 2.
[2] Istituto Superiore di Sanità, (2007) Piscine ad uso natato-
rio: aspetti igienico-sanitari e gestionali per l’appli-
cazione della nuova normativa. Rapporti ISTISAN 07/11.
[3] ISPESL (2005) Quaderni per la salute e la sicurezza “Le
piscine”. Osservatorio Nazionale Epidemiologico sugli
ambienti di vita. Istituto Superiore Prevenzione e
Sicurezza sul Lavoro.
[4] Gunkel, K. and Jessen, H-J. (1988) The problem of urea
in bathing water. Zeitschrift für die Gesamte Hygiene, 34,
[5] Erdinger, L., Kirsch, F. and Sonntag, H-G, (1997) Potas-
sium as an indicator of anthropogenic contamination of
swimming pool water. Zentralblatt für Hygiene und
Umweltmedizin, 200(4), 297-308.
[6] WHO (2006) Guidelines for Drinking-water Quality, first
addendum to third edition. Recommendations, World
Health Organization, 1.
[7] Evans, O., Cantú, R., Bahymer, T.D., Kryak, D.D. and
Dufour, A.P. (2001) A pilot study to determine the water
volume ingested by recreational swimmers. Paper pre-
sented to 2001 Annual Meeting of the Society for Risk
Analysis, Seattle, Washington, 2-5 December 2001.
[8] Meek, M.E., Beauchamp, R., Long, G., Moir, D., Turner,
L. and Walker, M. (2002) Chloroform: exposure estima-
tion, hazard characterization, and exposure-response
analysis, J Toxicol Environ Health B Crit Rev, 5(3),
[9] International Agency for Research on Cancer (1999)
IARC Monographs on the evaluation of carcinogenic risk
to humans. Some chemicals that cause tumors of the
kidney or urinary bladder in rodents and some other sub-
M. C. Aprea et al. / Natural Science 2 (2010) 68-78
Copyright © 2010 SciRes. OPEN ACCESS
stances, IARC Monographs, 73.
[10] American Conference of Governmental Industrial Hy-
gienists (2007). Threshold limit values for chemical sub-
stances and physical agents and biological exposure in-
dices, ACGIH. Cincinnati OH, USA.
[11] Jovanovic, S., Wallner, T. and Gabrio, T. (1995) Final
report on the research project “Presence of haloforms in
pool water, air and in swimmers and lifeguards in out-
door and indoor pools”. Stuttgart, Landesgesundheitsamt
[12] Sandel, B.B. (1990) Disinfection by-products in swim-
ming pools and spas. Olin Corporation Research Center
(Report CNHC-RR-90-154) (available from Arch Chem-
ical, Charleston).
[13] Biziuk, M., Czerwinski, J. and Kozlowski, E. (1993)
Identification and determination of organohalogen com-
pounds in swimming pool water, Int J Environ Anal
Chem, 46, 109-115.
[14] Aggazzotti, G., Fantuzzi, G., Righi, E., Tartoni, P.L., Cas-
sinadri, T. and Predieri, G. (1993) Chloroform in alveolar
air of individuals attending indoor swimming pools, Arch
Environ Health, 48, 250-254.
[15] Aggazzotti, G., Fantuzzi, G., Righi, E. and Predieri, G.
(1995) Environmental and biological monitoring of
chloroform in indoor swimming pools, J Chromatogr A,
710(1), 181-190.
[16] Aggazzotti, G., Fantuzzi, G., Righi, E. and Predieri, G.
(1998) Blood and breath analyses as biological indicators
of exposure to trihalomethanes in indoor swimming
pools, Sci Total Environ, 217 (1-2), 155-163.
[17] Copaken, J. (1990) Trihalomethanes: Is swimming pool
water hazardous? In: Jolley RL, Condie LW, Johnson.
[18] Armstrong, D.W. and Golden, T. (1986) Determination of
distribution and concentration of trihalomethanes in
aquatic recreational and therapeutic facilities by elec-
tron-capture GC, LC-GC, 4, 652-655.
[19] Eichelsdörfer, D., Jandik, J. and Weil, L. (1981) Forma-
tion and occurrence of organic halogenated compounds
in swimming pool water. A.B. Archiv des Badewesens, 34,
[20] Lahl, U., Bätjer, K., Duszeln, J.V., Gabel, B., Stachel, B.
and Thiemann, W. (1981) Distribution and balance of
volatile halogenated hydrocarbons in the water and air of
covered swimming pools using chlorine for water disin-
fection, Water Research, 15, 803-814.
[21] Ewers, H., Hajimiragha, H., Fischer, U., Böttger, A. and
Ante, R. (1987) Organic halogenated compounds in
swimming pool waters, Forum Städte-Hygiene, 38,
[22] Puchert, W., Prösch, J., Köppe, F-G. and Wagner, H.
(1989) Occurrence of volatile halogenated hydrocarbons
in bathing water. Acta Hydrochimica et Hydrobiologica,
17, 201-205.
[23] Puchert, W. (1994) Determination of volatile halogenated
hydrocarbons in different environmental compartments
as basis for the estimation of a possible pollution in West
Pommerania. Dissertation, Bremen, University of Bre-
[24] Cammann, K., Hübner, K. (1995) Trihalomethane con-
centrations in swimmers’ and bath attendants’ blood and
urine after swimming or working in indoor swimming
pools. Arch Environ Health, 50(1), 61–65.
[25] Schössner, H., Koch, A. (1995) Investigations of trihalo-
genmethane-concentrations in swimming pool water.
Forum Städte-Hygiene, 46, 354–357.
[26] Stottmeister, E. (1998) Disinfection by-products in Ger-
man swimming pool waters. Paper Presented to the 2nd
International Conference on Pool Water Quality and
Treatment, 4 March 1998, School of Water Sciences,
Cranfield University, Cranfield, UK.
[27] Stottmeister, E. (1999) Occurrence of disinfection by-
products in swimming pool waters. Umweltmedizinischer
Informationsdienst, 2, 21–29.
[28] Erdinger, L., Kirsch, F., Hoppner, A., Sonntag, H.G.
(1997) Haloforms in hot spring pools. Zentralblatt für
Hygiene und Umweltmedizin, 200, 309–317.
[29] Erdinger, L., Kuhn, K.P., Kirsch, F., Feldhues, R., Frobel,
T., Nohynek, B., Gabrio, T. (2004) Pathways of triha-
lomethane uptake in swimming pools. International
Journal of Hygiene Environmental Health, 207(6), 1–5.
[30] Kaas, P. and Rudiengaard, P. (1987) Toxicologic and
epidemiologic aspects of organochlorine compounds in
bathing water. Paper Presented to the 3rd Symposium on
Problems of Swimming Pool Water Hygiene, Reinhards-
[31] Borsányi, M. (1998) THMs in Hungarian swimming pool
waters. Budapest, National Institute of Environmental
Health, Department of Water Hygiene (unpublished).
[32] Chu, H. and Nieuwenhuijsen, M.J. (2002) Distribution
and determinants of trihalomethane concentrations in in-
door swimming pools. Journal of Occupational Envi-
ronmental Medicine, 59(4), 243–247.
[33] Lévesque, B., Ayotte, P., LeBlanc, A., Dewailly, E.,
Prud’Homme, D., Lavoie, R., Allaire, S., Levallois, P.
(1994) Evaluation of dermal and respiratory chloroform
exposure in humans. Environmental Health Perspectives,
102(12), 1082–1087.
[34] Fantuzzi, G., Righi, E., Predieri, G., Ceppelli, G., Gobba,
F., Aggazzotti, G. (2001) Occupational exposure to triha-
lomethanes in indoor swimming pools. Science of Total
Environment, 264(3), 257–265.
[35] Strähle, J., Sacre, C., Schwenk, M., Jovanovic, S., Gabrio,
T., Lustig, B. (2000) [Risk assessment of exposure of
swimmers to disinfection by-products formed in swim-
ming pool water treatment. Final report on the research
project of DVGW 10/95, Landesgesundheitsamt Ba-
den-Württemberg, Stuttgart.
[36] Aggazzotti, G., Fantuzzi, G., Tartoni, P.L., Predieri, G.
(1990) Plasma chloroform concentration in swimmers
using indoor swimming pools. Archives of Environmental
Health, 45A(3), 175–179.
[37] Kuo, H.W., Chiang, T.F., Lo, I.I., Lai, J.S., Chan, C.C.,
Wang, J.D. (1998). Estimates of cancer risk from chloro-
form exposure during showering in Taiwan. Science of
Total Environment, 218(1), 1-7.
[38] Du Bois, D. and Du Bois, E. (1916). A formula to esti-
mate the approximate surface if height and weight be
known. Clinical Calorimetry, tenth paper. Archives of
International Medicine, 863-871.
[39] Brugnone, F., Perbellini, L., Giuliari, C., Cerpelloni, M.,
Soave, C. (1994) Blood and urine concentrations of
chemical pollutants in the general population. Med Lav.
85(5), 370-389.