Nephrol Dial Transplant (2006) 21: 1618–1625
doi:10.1093/ndt/gfl036
Advance Access publication 20 February 2006
Original Article
Organic contamination in dialysis water: trichloroethylene
as a model compound
Diana Poli1,2, Laura Pavone2, Pius Tansinda2, Matteo Goldoni1,2, Dante Tagliavini2,
Salvatore David2, Antonio Mutti2 and Innocente Franchini2
1
National Institute of Occupational Safety and Prevention Research Center at the University of Parma and
Department of Clinical Medicine, Nephrology and Health Sciences, University of Parma,
Via Gramsci 14, 43100 Parma, Italy
2
Abstract
Background. Routine water monitoring in a
haemodialysis centre revealed high trichloroethylene
(TCE) concentrations. The aim of this study is to
describe the measures adopted after organic contamination of dialysis water in order to avoid the
possibility of patient exposure. We also carried out
in vitro experiments to evaluate the accumulation of
TCE in various devices normally used in a dialysis
water treatment system (DWTS).
Methods. In vivo and in vitro blood and water
TCE levels were determined by means of solid
phase microextraction-gas chromatography/mass
spectrometer.
Results. High TCE concentrations were found
throughout the DWTS; acceptably low levels were
obtained only by replacing the activated charcoal,
ionic-exchange resins, microfilters and PVC pipes.
The adsorption and realising capacities of these devices
were tested in vitro, and the elimination curves made
it possible to calculate the total percentage of the
previously absorbed TCE mass released into the water.
Evidence of exposure was confirmed by the relatively
high TCE levels in the patient blood samples taken
30 days after the last exposure even if the subjects were
asymptomatic. In vivo experiments showed that the
blood gain of TCE through the low flux membrane
during the course of dialysis was about 77±10.4% of
the amount carried by dialysis fluid as calculated on
the basis of its partition coefficient value (Kb/w 3.75).
Conclusions. This study shows that, when present in
dialysis water, the lipophilic TCE contaminant can
accumulate in various devices, thus transforming them
Correspondence and offprint requests to: Diana Poli, National
Institute of Occupational Safety and Prevention Research Center
at the University of Parma, Department of Clinical Medicine,
Nephrology and Health Sciences, Via Gramsci 14, University of
Parma, 43100 Parma, Italy. Email: [email protected]
into possible sources of exposure. This highlights the
importance of periodically monitoring dialysis water
for organic substances that have a great affinity to the
blood compartment, in order to prevent occasional or
chronic patient exposure.
Keywords: dialysis; organo-halogenated compounds;
trichloroethylene; water contamination
Introduction
Patients treated with standard haemodialysis are
exposed to almost 400 l of dialysis fluid water every
week, and the fact that this is separated from the
bloodstream only by a semi-permeable membrane
explains the fact that the water needs to be chemically
and microbiologically pure.
The question of purity of dialysis fluid has been
confronted by the Association for the Advancement
of Medical Instrumentation (AAMI) [1] and European
Pharmacopoeia [2], whose guidelines consider the
principal contaminants to be the inorganic substances
that have so far been documented as toxic in
haemodialysis. There are no recommendations concerning the presence of organic compounds in dialysis
fluid, and the most widely used reference values are
those indicated for drinkable tap water.
The more recent Italian guidelines on dialysis water
and solutions [3] include organo-halogenated compounds, which come from industrial waste or water
chlorination, among the water contaminants requiring
at least annual monitoring.
Like other organo-halogenated compounds, trichloroethylene (TCE), has become an ubiquitous environmental pollutant and measurable concentrations have
been found in the urine and blood of the unexposed
population [4]. It is mainly used as an extraction solvent
ß The Author [2006]. Published by Oxford University Press on behalf of ERA-EDTA. All rights reserved.
For Permissions, please email: [email protected]
Trichloroethylene dialysis water contamination
in chemical processing and may be emitted from
industrial plants in the form of vapour and/or in
aqueous effluent [4–6]. TCE is classified in the
International Agency for Research on Cancer class
2A as ‘probably’ carcinogenic to humans [6] and its
carcinogenicity seems to be mainly mediated by its
metabolites in liver and kidney [7,8]. It is also known to
induce hepatic damage, polyneuropathy, trigeminal
neuropathy and dermatitis [9].
In municipal water, the maximum concentration
accepted by the Italian Health Ministry is 10 mg/l for
both tri- and tetra-chloroethylene and 30 mg/l for the
total halomethane [10].
Case report
During the course of a routine 6-monthly chemical
monitoring of reverse osmosis (RO) water in a dialysis
centre, we found a TCE concentration of 60 mg/l, well
above the acceptable limits. In addition to monitoring
the substances indicated for water dialysis purity,
organo-halogenated compounds are periodically
tested by the Emilia Romagna Regional Agency for
Environmental Protection (ARPA), and anomalous
TCE concentrations had never been previously found.
ARPA did not show us their previous monitoring
schedules, but assured us that the TCE level in
tap water had never exceeded the 10 mg/l reference
limits. Other organo-halogenated compounds, such
as trichloromethane, trichloroethane, dichlorobromomethane, tetrachloroethylene, tribromomethane and
carbon tetrachloride were detected at very low concentrations (<0.4 mg/l) in dialysis fluid, and the tap
water TCE level at the entry to the haemodialysis
circuit was 1.4 mg/l, below the maximum reference
value.
1619
Haemodialysis treatment was immediately interrupted and the patients were switched to other local
centres until the TCE values normalized. Analyses
made on the following days confirmed that there were
high TCE concentrations at various points in the
dialysis water treatment system (DWTS), levels that
had never been seen in our centre before.
The replacement of all the contaminated devices
(activated charcoal, microfilters, ionic-exchange resins
and PVC tubes) allowed the normalization of the
water TCE values, and the patients were able to return
to the centre after 1 month of DWTS checks. The aim
of this study is to describe the measures that should
be adopted after organic substance contamination of
dialysis water, with TCE being considered a model
compound:
(1) cleaning the system in order to avoid the possibility
of patient exposure;
(2) evaluating the contamination of the devices
normally used in a DWTS by means of in vitro
experiments;
(3) quantifying the affinity of TCE (as a lipophilic
organic contaminant) towards the blood compartment by studying the blood/water partition coefficient under both static and dynamic conditions
(during dialysis treatment).
Subjects and methods
Water circuit monitoring
The dialysis water treatment and distribution circuit in our
centre is shown in Figure 1. It has a single-pass design and
consists of a pretreatment system (tap water chlorination
by means of the injection of sodium hypochlorite, microfilters, softener and activated charcoal) and single-stage RO
Fig. 1. Dialysis water treatment system (DWTS) showing the TCE levels recorded in water samples taken from different points in the
system.
1620
treatment followed by the use of microfilters and a UV lamp.
This type of circuit is very common in small dialysis centres
in Italy.
We collected 2 ml fluid sample in 4 ml screw cap vials
containing 1 g of dried NaCl, and froze them until analysis.
The TCE concentrations were measured in water samples
drawn at the following points in the DWTS: circuit entry
(tap water), after activated carbon filtering, after RO, at the
feeding machine point (affluent water) and the circuit
terminal. The tests were repeated after the progressive
replacement of the potentially contaminated circuit components as follows: activated charcoal, microfilters, ionicexchange resins, chlorination tanks and PVC tubes.
Finally, TCE levels were also measured in fresh dialysis
fluid.
D. Poli et al.
urea clearance 238 ml/min (range 222–249 ml/min) and
creatinine clearance 206 ml/min (range 194–249 ml/min),
with a blood flow of 300 ml/min and dialysate flow of
500 ml/min.
One month after TCE exposure, blood TCE concentrations were measured in seven patients: 2 ml of pre-filter
blood (pre-B) were collected from an artero-venous fistula
in 4 ml screw cap vials containing 1 g of dried NaCl, and
frozen until analysis. The blood TCE determinations were
repeated every 2 weeks until the results were comparable with
those
observed
in
the
unexposed
population
(<0.015–0.090 mg/l) [4].
Pre- and post-filter blood (pre-B and post-B) and dialysate
(pre-D and post-D) TCE levels were also measured in
order to evaluate TCE blood/water partition during dialysis
treatment. The samples were collected as described above.
In vitro experiment
Water chlorination for TCE in situ production. After
the addition of 2 mg/l of sodium hypochlorite (NaClO) to
distilled and tap water, the samples were stored at room
temperature for four days and then analysed as described in
the section, ‘Analytical procedures’, subsequently.
TCE absorption and releasing by DWTS devices. Two
inert tubes were completely filled with activated charcoal
(113 g) and ionic-exchange resins (196 g) each, and a standard
microfilter was placed in a specific container. These devices,
together with one 31-cm long PVC pipe with an internal
diameter of 24 mm, were plugged at both ends, filled with
1 mg/l of a TCE water solution, and stored at room
temperature for 24 h, when the TCE concentrations were
measured.
After removing the solution, the activated charcoal, ionicexchange resins and filter were washed with distilled water
for 190 min (flow: 0.1 l/min for activated charcoal and ionicexchange resins; 2 l/min for filter); water samples were
collected from each at the fixed timepoints of 0, 3, 5, 10, 15,
20, 110 and 190 min, and analysed to study the elimination
kinetics. In the case of the PVC pipe, the elimination study
was carried out under static conditions because of its low
absorption capacity; the pipe was re-filled with distilled water
and analysed after resting intervals of 10, 110 and 190 min.
Patients
We examined all of the 15 patients who regularly undergo
haemodialysis at our centre: four females and 11 males,
with a mean age of 78 years (range 65–94) and a median
body mass index (BMI) of 26 (range 19–30). Four were
diabetic, including three with a painful peripheral neuropathy
and one with restless-leg syndrome; two patients (including
one with chronic HBV-related liver disease) normally showed
slight fluctuations in cholestasis (-glutamyl-transferase
and/or alkaline phosphatase) and transaminases (alanine
aminotransferase and aspartate aminotransferase); two had
anti-HCV antibodies but no signs of chronic liver disease;
and four had residual renal function (creatinine clearance
8–12 ml/min). Leucocyte count, was within our laboratory
range values (4800–10 800 cell/mm3).
All the patients were treated with bicarbonate haemodialysis and low-flux membranes. The dialyser characteristics
were: mean UF rate 6.5 ml/min (range 5.5–7.8 ml/min),
Analytical procedures
Chemicals. The TCE and 1-chlorobutane (used as internal
standard), came from Sigma Aldrich (Milan, Italy); the
standard stock solutions prepared in HPLC-grade methanol
and stored at 20 C were stable for at least 1 month. The
ultra-pure sodium hypochlorite solution (available chlorine
10–13%) also came from Sigma Aldrich (Milan, Italy).
Sample preparation and extraction. Before analysis,
1-chlorobutane was added as internal standard to each
sample of dialysis water and heparinized blood (final
concentration 2 mg/l). The samples were extracted for 30 min
at room temperature by means of solid phase microextraction
(SPME), and then analysed by gas chromatography/mass
spectrometry (GC/MS), as previously described [11]. The
calibration samples were prepared in a TCE-free blood and
dialysis fluid pool (linear range 0–100 mg/l).
Gas chromatography/mass spectrometry (GC/MS)
analysis. The analyses were carried out on a Hewlett
Packard HP 6890 gas chromatograph coupled with an HP
5973 Mass Selective Detector (Hewlett Packard, Palo Alto,
CA, US). Separation was performed on an HP-5MS column
(30 m 0.25 mm i.d., 0.25 mm film) using H2 as carrier gas
(1 ml/min). Quantitative analysis was performed in selected
ion monitoring mode by monitoring the signals of the
following ions (dwell time in parentheses): 95 (60), 97 (90),
130 (50) and 132 (60) for TCE and 41 (90), 43 (120) and
56 (60) for internal standard. The chromatographic run was
complete in 6 min.
Statistical analysis
The significance (P<0.05) of the differences in blood TCE
levels at different times was calculated using Friedman test
followed by Dunn’s post-hoc comparisons and Prism 3.0
software (Graphpad, San Diego, CA, US).
Results
Water circuit monitoring
The TCE levels recorded in tap water were between
1.1–1.6 mg/l, but were progressively higher after the
Trichloroethylene dialysis water contamination
1621
activated carbon filter (37 mg/l), the RO membrane
(60 mg/l) and at the end of circuit (107 mg/l); there was
a wide range of values (18–86 mg/l) at different feeding
machine points (Figure 1). Replacing the activated
charcoal lowered the TCE level in the RO water to
3 mg/l, but the values remained high (75 mg/l) at the end
of the circuit. We then substituted the ionic-exchange
resins, microfilters and PVC pipes and obtained
acceptably low TCE concentrations (max value
0.5 mg/l) throughout the DWTS. Despite the above
changes, and even after complete circuit cleaning, the
pre-filter dialysis fluid still showed relatively high TCE
levels (range: 12.5–20.1 mg/l), and normal values were
only reached after the replacement of the ultrafilters
planted in the rear of the machines.
elimination (lag) phase of 10 min (Figure 2B), after
which a large amount of TCE (47.6%) was quickly
released (t1/2<1 min) and the remainder (52.3%) was
slowly eliminated (t1/2 ¼ 56.4 min). Finally, the releasing capacity curve for activated charcoal was a monoexponential function (Figure 2C) with t1/2 of 68.1 min.
Applying the integrating function to the elimination
curves made it possible to calculate the percentage of
total TCE released by each device in relation to the
previously absorbed mass: 7.3% for the activated
charcoal, 62.0% for the ionic-exchange resins, and
89.8% for the microfilter. The total percentage (61.8%)
released from the PVC pipe was obtained by calculating
the amount of TCE released after the resting times of
10, 110, 190 min (Figure 2D).
In vitro experiments
Patient exposure
We investigated two hypotheses to explain TCE
contamination:
All of the 15 patients who attended the centre were
clinically evaluated at the time of the water contamination and after the re-opening of the dialysis centre.
No relevant signs related to the TCE intoxication
emerged even after a complete physical examination,
and the patients never complained of any new
symptoms; there were no neurological disturbances
such as somnolence, or facial or peripheral paresthesias. After the re-opening of the dialysis centre, no
significant alterations in routine blood tests were
observed and, in particular, the hepatotoxicity indices
(alanine aminotransferase, aspartate aminotransferase,
-glutamil-transferase, alkaline phosphatase) were
within the normal range.
Only seven patients agreed to repeat the blood
collection during dialysis sessions in a follow-up study
of 30–60 days after the re-opening of the centre. Blood
samples collected 30 days after the last exposure,
showed higher TCE concentrations than those measured in the unexposed population (Figure 3, which
also shows the time course of the depletion of blood
TCE concentrations) [4]. The 30 and 45 day levels were
significantly higher than those recorded at 60 days
(P<0.01 and P<0.05, respectively) when the TCE
levels were similar to those expected in the unexposed
population.
The use of a kinetic curve made it possible to
extrapolate an initial value of 1.48 mg/l for the blood
TCE concentration, about 15 times higher than the
maximum reported in the general unexposed population (0.090 mg/l) [4].
In order to understand the affinity of TCE for
dialysis fluid and blood as a model for the transfer
of a lipophilic organic contaminant to the blood
compartment, we studied the blood/water partition
coefficient after equilibrium was reached under static
conditions.
The experiments were carried out in airtight vials
(20 ml) filled with 10 ml of water (dialysis fluid) or blood
in order to calculate the water/air (Kw/a) and blood/air
(Kb/a) partition coefficients which were respectively
0.24 and 0.90 at room temperature. The blood/water
partition coefficient (Kb/w) was calculated indirectly
(a) excessive tap water chlorination leading to the
in situ production of TCE;
(b) the TCE, accumulated as an environmental
pollutant at different points in the DWTS and
was subsequently released.
In situ TCE production was evaluated by adding
NaClO solution to distilled and tap water with a
previous concentration of 0.89 mg/l. After storing
the samples at room temperature for 4 days, some
halomethanes (chloroform and bromodichloromethane) were detected in the tap water but the TCE
concentration remained unchanged; the TCE level in
the distilled water remained undetectable (data not
shown).
To verify the second hypothesis, we tested the
TCE absorption and releasing capacity of the ionicexchange resins, microfilters, activated charcoal and
PVC water pipes in the DWTS. The devices were
contaminated under static conditions with an 1 mg/l
TCE water solution for 24 h, as described in the section
‘Subjects and methods’. After measuring the TCE
concentrations at 24 h, we calculated the amount
absorbed by ionic-exchange resins (82.8%), microfilter
(93.0%), activated charcoal (99.8%) and PVC
pipe (2.1%).
After removing the solution, ionic-exchange resins,
filter and activated charcoal were washed with distilled
water, and water samples were collected from all the
devices at fixed timepoints in order to study their
elimination kinetics; in the case of the PVC pipe, the
elimination study was carried out under static conditions because of its low absorption capacity. The
elimination curve for the ionic-exchange resins was a
tri-exponential function (Figure 2A): about 38% of the
total TCE amount absorbed was quickly eliminated
(t1/2<1 min), whereas the highest fraction (57%) was
eliminated with a slower kinetic (t1/2 ¼ 8.8 min) and
only 5% was released with a t1/2 of 162 min. The filter
showed a bi-exponential function following a constant
1622
D. Poli et al.
Fig. 2. Elimination kinetics after exposure to 1 mg/l TCE water solution for 24 h: ionic-exchange resins (A), microfilter (B), activated
charcoal (C) and PVC pipe (D). The elimination study of the PVC pipe was conducted under static conditions.
using the following equation:
ðKb=w Þ ¼ Kb=a =Kw=a ¼ 3:75
and corresponded to a partition of 73.4% of the total
TCE in blood fluid after reaching equilibrium.
In order to evaluate the transfer from dialysis fluid
to the bloodstream under dynamic condition, we
measured TCE levels in pre- and post-D and pre- and
post-B during treatment, after the water TCE concentration in the centre had normalized. The highly
sensitive SPME-GC/MS technique for organohalogenated compounds [10] allows the detection of
very low TCE concentrations, such as the amount
lost in water after contact with the bloodstream.
The final decrease in TCE concentration in post-D
(77±10.4%) and the consequent increase in post-B
(64.5±32.1%), reflected with a good approximation
the results expected from the partition coefficient value
(Kb/w) previously calculated under static conditions.
Discussion
Haemodialysis patients have higher risks of intoxication in the case of dialysis water contamination because
of the almost direct contact between water dialysis and
the blood compartment, and the patients’ limited or no
urinary excretion. Periodic chemical tests to exclude
the contamination of dialysis water by inorganic and
organic pollutants are therefore strictly necessary.
One such test in our centre revealed TCE contamination in RO water and, in order to be able to resume
Trichloroethylene dialysis water contamination
Fig. 3. Blood TCE values in seven patients, 30 days after the
accident and the corresponding time course depletion. The
horizontal bars represent the median values. **P<0.01, *P<0.05
vs 60-days values (Friedman test followed by Dunn’s post-hoc
comparisons). The background TCE levels were taken by Skender
et al. [4].
haemodialysis sessions, it was necessary to determine
the main source(s) of TCE, clean up the DWTS and
prevent further contamination.
A recent article [12] has described organohalogenated water contamination in a dialysis centre
as principally due to chloroform (CHCl3), although the
source of the contamination was not identified. The
authors suspected that the in situ production of CHCL3
was due to excessive water chlorination because it is
known that chlorination induces the formation of
disinfection by-products, especially trihalomethanes
[13,14]. On the other hand, no clear indications have
been published concerning TCE production after water
treatment, and so we checked possible TCE generation
after water chlorination; however, our experiments
excluded this hypothesis. The origin/source of the
contamination is still unclear. Although ARPA assured
us that the TCE levels in tap water has never been
higher than the accepted limit (10 mg/l), we considered
the possibility that the TCE accumulation in our
DWTS may have been due to intermittent and/or
undetected high TCE concentrations in the municipal
water supply, because other episodes of contamination
by organo-halogenated compounds have been reported
in the area in which our centre is located, even if the
corresponding tap water concentrations have always
been considered within safe limits.
Activated carbon filter usually used to remove
dissolved organic contaminants such as chlorine,
chloramines and possibly organo-halogenated compounds from tap water [15,16]. Our experiments
confirmed its high capacity to absorb TCE (99.8%),
but also showed a low and slow releasing capacity that
amounted to about 7% of the retained quantity. It is
likely that when its absorption capacity is exceeded
due to an acute or chronic overload, the compounds
may be released in effluent water. Daily monitoring of
the chlorine concentration in the activated charcoal
effluent and consequently, the periodic replacement of
1623
carbon filter are therefore necessary to control optimal
functioning.
Nevertheless, we found that water TCE levels
remained high after the replacement of the activated
charcoal filter, and so we had to examine the other
devices. The ionic-exchange resins and microfilters
showed greater releasing capacity and faster kinetics
than the activated charcoal. It is interesting to note
that water stagnation can induce the accumulation of
organic substances, such as colloids and particulates,
and that the presence of such substances induces
rapid filter membrane saturation which could facilitate
TCE contamination and eventually lead to a breakthrough. None of our DWTS devices therefore offers
the possibility of producing TCE-free water on a longterm basis, thus indicating that all the elements of the
water preparation and distribution system may be
sources of TCE contamination; their programmed
replacement would thus seem to be a safe and necessary
means of preventing the accumulation and/or release
of contaminants.
Monitoring the patients’ blood TCE levels at three
different times showed the progressive depletion of
blood TCE, and confirmed the previous exposure;
however, the small number of patients and their
relocation to other dialysis centres made it impossible
to relate the time course of the depletion to their
individual characteristics, such as BMI, residual renal
function or chronic obstructive pulmonary disease.
Studying the elimination curves allowed us to extrapolate a blood TCE concentration of 1.48 mg/l at time
zero, which is about 15 times higher than the maximum
reported in the general unexposed population
(0.090 mg/l); however, it is necessary to note that this
is only a rough estimate because the first part of the
elimination was lost. Nevertheless, blood levels of
1.48 mg/l during exposure is consistent with the maximum level measured in the DWTS (107 mg/l), which
was about 10 times higher than the concentration
allowed in tap water (10 mg/l).
However, it is unlikely that blood TCE concentrations had reached acute toxic levels because clinical
symptoms due to acute TCE intoxication in humans
have been described in the literature only in the case
of much higher blood TCE levels (>4 mg/l) than those
observed in our patients [17], which would explain why
they had no symptoms. In terms of chronic toxicity,
there are studies that have quantified safe oral and
airborne TCE exposures in relation to cancer [18], but
there are no published papers reporting a correlation
between blood TCE levels and cancer risk in humans.
The presence of organic contaminants in dialysis
water may be a serious problem, especially in the case
of lipophilic substances such as TCE, because of the
ease with which they pass into the blood compartment.
Calculating the blood/water partition coefficient under
static conditions and a low concentration, we found
that the affinity of TCE for blood was about four times
greater than its affinity for water, which is similar to
the distribution calculated during dialysis treatment.
This seems to indicate that the transfer of low TCE
1624
concentrations during dialysis treatment is faster than
fluid flow rates because of the low molecular weight of
TCE and its high affinity for the bloodstream.
However, in the presence of high TCE concentrations
such as those measured during our DWST contamination, the time necessary for TCE to diffuse from dialysis
fluid to the blood compartment may be longer, and the
blood compartment may not be able to remove TCE
completely from dialysis fluid. Furthermore, the blood
compartment could be saturated by high TCE concentrations in dialysis fluid, and so the dynamic passage
of TCE from water to blood may deviate from the
distribution law that is valid under static equilibrium
conditions. We were, therefore, unable to establish
a mass balance between the blood compartment
and dialysis fluid at time zero, as we did when the
TCE concentrations in both water and blood were
normalized.
On the basis of the results obtained at low TCE
concentrations, we can make some speculations concerning the direct application of the upper tap water
TCE limit to dialysis fluid. Given the similar molecular
weight of TCE and creatinine, we can assume that a
haemodialysis filter has the same mass transfer area
coefficient for both compounds; however, under the
same dialysis conditions, the equivalent blood flux for
TCE is about four times than that of creatinine because
of its Kb/w. It can be estimated from the Michaels’
equations [19] that a haemodialysis filter with a
creatinine clearance of 206 ml/min, has a TCE clearance
of 280 ml/min, and so 67 l of water can be completely
purified of TCE in a 4-days dialysis session, which
corresponds to a daily exposure of 33.5 l. Given the
upper tap water TCE concentration fixed by the Italian
Health Ministry (10 mg/l), we can assume that daily
oral exposure to 20 mg of TCE should be considered
safe for a mean water consumption of 2 l per day. This
amount of TCE can be received by a patient during
a standard dialysis session if the fluid TCE concentration is 0.59 mg/l (20 mg/33.5 l), a value that is 17 times
lower than the acceptable limit, thus showing the
amplification of exposure during the course of dialysis.
The recent Italian guidelines on dialysis water and
solutions for dialysis [3] include organo-halogenates
(and thus TCE) in the list of water contaminants, and
recommend at least annual dialysis water monitoring;
however, on the basis of our calculations, dialysate
TCE levels should be at least 17 times lower than the
limit fixed for drinkable water. Any attempt to apply
the organic contaminant limits for drinkable tap water
to dialysis fluid should therefore be made very
cautiously on the basis of the direct exposure to high
water volumes during a dialysis session. However, it is
not our intention to propose fixed limits, but only to
draw attention to the importance of defining more
adequate limits for environmental and/or industrial
pollutants in dialysis fluid.
In conclusion, our study demonstrates that lipophilic
contaminants in dialysis water can be accumulated in
DWTS devices, thus transforming them into possible
sources of intoxication. It also demonstrates the ease
D. Poli et al.
with which organic contaminants such as TCE pass
the haemodialysis filter. Thus implying the possible
risk of exposure of haemodialysis patients even in the
case of low xenobiotic concentrations. This highlights
the importance of periodically monitoring the
dialysis water levels of organic substances such as
TCE, which show considerable affinity with the blood
compartment.
Acknowledgements. We thank Ing. D. Graziani for his cooperation during the study.
Conflict of interest statement. The authors declare that they have no
competing interests.
References
1. Association for the advancement of medical instrumentation
and American national standards institute, Hemodialysis
systems, 1993 AAMI Standard and Recommendation practice.
AAMI/DS-1 RD 62. Water treatment equipment for hemodialysis applications. Association for the Advancement of
Medical Instrumentation. Arlington, VA, USA; 2000: 2–32
2. European Pharmacopoeia, 3rd edn, Hemodialysis solution,
concentrated water for diluting, Supplement 2001: Monograph
1997: 1167 corrected 2000
3. Giornale Italiano di Nefrologia, Anno 22, n. 3, 2005pp. 246–273. Linee guida su acque e soluzione per dialisi
4. Skender L, Karacic V, Bosner B, Prpic-Majic D. Assessment
of urban population exposure to trichloroethylene and
tetrachloroethylene by means of biological monitoring. Arch
Environ Health 1994; 49: 445–451
5. National Toxicology Program, Rep Carcinog 2002; 10: 244–247
6. IARC, International Agency for Research on Cancer.
Dry cleaning, some chlorinated solvents and other industrial
chemicals. IARC Monogr. Eval. Carcinog. Risks Hum. 63,
Lyon, 1995, p.75 and 159
7. Brunig T, Bolt HM. Renal toxicity and carcinogenity of
trichloroethylene: key results, mechanism, and controversies.
Crit Rev Toxicol 2000; 30: 253–285
8. Bernauer U, Birner G, Dekant W, Henschler D.
Biotransformation
of
trichloroethene:
dose-dependent
excretion of 2,2,2-trichloro-metabolites and mercapturic acids
in rats and human after inhalation. Arch Toxicol 1996; 70:
338–346
9. Kaneko T, Wang TY, Sato A. Assessment of the health effects
of trichloroethylene. Ind Health 1997; 35: 301–324
10. D.Lgs 2 Febbraio 2001 n 31 Attuazione della direttiva 98/83/
CE relativa alla qualità delle acque destinate al consumo
umano
11. Poli D, Manini P, Andreoli R, Franchini I, Mutti A.
Determination of dichloromethane, trichloroethylene and
perchloroethylene in urine samples by headspace solid phase
microextraction gas chromatography-mass spectrometry.
J Chromatogr B Analyt Technol Biomed Life Sci 2005; 820:
95–102
12. Formica M, Vallero A, Forneris G et al. Organohalogenated
contamination of a dialysis-water treatment plant. G Ital Nefrol
2002; 19: 479–482
13. Cho DH, Kong SH, Oh SG. Analysis of trihalomethanes in
drinking water using headspace-SPME technique with gas
chromatography. Water Res 2003; 37: 402–408
14. Kim H, Shim J, Lee S. Formation of disinfection by-products
in chlorinated swimming pool water. Chemosphere 2002; 46:
123–130
15. Weiss LG, Danielson BG, Fellstrom B, Wikstrom B.
Aluminium removal with hemodialysis, hemofiltration
Trichloroethylene dialysis water contamination
and charcoal hemoperfusion in uremic patients after desferrioxamine infusion. A comparison of efficiency. Nephron 1990;
54: 325–329
16. Muirhead N, Mitton R. Use of bone char as an
adsorbent in preparation of water for dialysis. ASAIO J
1992; 38: M334–M337
17. Boyes WK, Evans MV, Eklund C, Janssen P, Simmonds JE.
Duration adjustment of acute exposure guideline level values
1625
for trichloroethylene using a physiologically-based pharmacokinetic model. Risk Anal, 2005, 25: 677–686
18. Bogen KT, Gold LS. Trichloroethylene cancer risk: simplified
calculation of PBPK-based MCLs for cytotoxic end points.
Regul Toxicol Pharmacol 1997; 25: 26–42
19. Michaels AS. Operating parameters and performance criteria
for hemodialyzer and other membrane separation devices.
Tr Am Soc Artif Inter Organs 1966; 12: 387–392
Received for publication: 22.4.05
Accepted in revised form: 23.1.06