Accepted Manuscript

Title: Environmental and occupational exposure to resorcinol

in Finland

Authors: Simo P. Porras, Minna Hartonen, Katriina Ylinen,

Jarkko Tornaeus, Tapani Tuomi, Tiina Santonen

PII: S0378-4274(18)30114-0
DOI: https://doi.org/10.1016/j.toxlet.2018.03.027
Reference: TOXLET 10142

To appear in: Toxicology Letters

Received date: 7-2-2018

Accepted date: 24-3-2018

Please cite this article as: Porras, Simo P., Hartonen, Minna, Ylinen, Katriina, Tornaeus,
Jarkko, Tuomi, Tapani, Santonen, Tiina, Environmental and occupational exposure to
resorcinol in Finland.Toxicology Letters https://doi.org/10.1016/j.toxlet.2018.03.027

This is a PDF file of an unedited manuscript that has been accepted for publication.
As a service to our customers we are providing this early version of the manuscript.
The manuscript will undergo copyediting, typesetting, and review of the resulting proof
before it is published in its final form. Please note that during the production process
errors may be discovered which could affect the content, and all legal disclaimers that
apply to the journal pertain.
Environmental and occupational exposure to resorcinol in Finland

Simo P. Porras*, Minna Hartonen, Katriina Ylinen, Jarkko Tornaeus, Tapani Tuomi, Tiina Santonen

Finnish Institute of Occupational Health, PO Box 40, FI-00032 Työterveyslaitos, Finland

T
* Corresponding author. Tel. +358304742105. Email address: simo.porras@ttl.fi.

R IP
Abstract

SC
Resorcinol is a suspected endocrine disruptor that affects thyroid function by inhibiting thyroxin peroxidase. It
may also have an impact on iodine uptake. Resorcinol has various uses; for example in the manufacture of rubber
products and in wood adhesives, flame retardants, UV stabilizers, and dyes. It is also used in personal care
products such as hair colorants, anti-acne preparations, and peels.
U
N
The aim of this study was to assess both environmental background exposure and occupational exposure to
A
resorcinol in Finland. We investigated occupational exposure in hairdresser work and in the manufacture of
tyres, adhesive resins and glue-laminated timber by biomonitoring total resorcinol concentration in urine
M

samples. The biomonitoring results were compared to the urinary levels of occupationally non-exposed
volunteers, and to the biomonitoring equivalent (BE), which we estimated on the basis of the EFSA’s acceptable
ED

daily intake (ADI) value for resorcinol.

Almost all the urine samples (99%) of the non-occupationally exposed volunteers contained measurable
PT

amounts of resorcinol. The urinary resorcinol data were rather scattered, and the resorcinol concentrations
among women were clearly higher than the respective concentrations among men. The reason for this
difference remains unclear. Although the two highest results exceeded the BE calculated on the basis of the
E

EFSA’s ADI, the 95th percentile of the occupationally non-exposed volunteers’ results remained well below the
CC

BE among both males and females.

According to the results, hairdressers’ exposure to resorcinol was at the same level as that of the reference
A

population of occupationally non-exposed volunteers. All hairdresser’s values remained below the BE for
resorcinol.

The urinary resorcinol levels of the industrial workers were also at the same level as those of the reference
population. We observed slight increases in the post-shift and evening samples of those working in the
manufacture of tyres and adhesive resins. The results of some workers in the tyre manufacturing company
exceeded the 95th percentile of non-occupationally exposed males, which was used as a biological guidance

1
value for occupational exposure. Moreover, in this case exposure was below the health-based biomonitoring
equivalents. All the air samples collected in the companies contained very low resorcinol concentrations.

Abbreviations: ADI, acceptable daily intake; BE, biomonitoring equivalent; ECHA, European Chemicals Agency;
EFSA, European Food Safety Authority.

Keywords: resorcinol, environmental exposure, occupational exposure, workers, biomonitoring, urine,

industrial hygiene, air samples.

T
1. Introduction

IP
Resorcinol (1,3-dihydroxybenzene, CAS number 108-46-3) is used in, for example, the manufacture of rubber

R
products, wood adhesives, flame retardants, UV stabilizers, and dyes (Schmiedel and Decker, 2000; EC, 2002). It
is also a component of many personal care products such as hair colorants, anti-acne preparations, and peels

SC
(WHO, 2006).

Resorcinol is suspected to be an endocrine disruptor via thyroid effects (WHO, 2006; CEHOS, 2012). It seems to

U
inhibit thyroxin peroxidase and may also have an impact on iodine uptake (Tukes, 2017). Resorcinol has caused
N
hypothyroidism when it has been used dermally to treat skin ulcers (Lynch et al., 2002). In addition, resorcinol
irritates the eyes and skin, and may cause sensitization through skin contact (WHO, 2006) although it seems that
A
skin sensitization cases among humans are rare. Pharmacokinetic studies show that resorcinol is absorbed by
M

oral, dermal, and subcutaneous routes, and rapidly metabolized. Excretion via urine occurs principally as
glucuronide and sulphate conjugates (WHO, 2006). Data on the excretion kinetics of resorcinol in humans is
limited, but according to rat data, the excretion of resorcinol is rapid, and > 80% of orally administered resorcinol
ED

is excreted via urine within the first 24 hours (Kim and Matthews, 1987). Resorcinol concentrations in biological
fluids have rarely been reported. Yeung et al. developed an analysis method for free resorcinol in human plasma
and urine (Yeung et al., 1981). They also measured free and conjugated resorcinol in plasma and urine after 2%
PT

resorcinol solution was continuously applied to the human skin of three subjects for four weeks (Yeung et al.,
1983). The daily resorcinol dose was 12 mg/kg bw. All plasma levels were below the detection limit (100 µg/l).
E

The 24-hour urinary concentrations after two weeks of treatment were 4800–33700 µg/l, and 1600–8400 µg/l
CC

after four weeks. The results of one control subject were below the detection limit (100 µg/l) on both occasions.

Given the widespread occurrence of resorcinol, it is surprising that no biomonitoring studies are available on
human exposure to resorcinol. The most relevant sources of consumer exposure are likely to be the above-
A

mentioned personal care products. For example, resorcinol has been detected in several hair dye products (Yazar
et al., 2009, 2012; Hamann et al., 2014). The use of resorcinol in hair dyes was evaluated some years ago in the
European Union, but due to insufficient data, risk assessment could not be performed (SCCP, 2008). Urinary
resorcinol in the general population may also be due to other sources. For example, resorcinol is a monomeric
by-product of the reduction, oxidation and microbial degradation of humic substances (WHO, 2006). Resorcinol
is a major metabolite of tannic acid (Nakamura et al., 2003; EFSA, 2014), found in foods and beverages. It is also

2
applied as an anti-browning food additive to fresh and frozen crustaceans (EFSA, 2010). The brominated flame
retardant resorcinol bis(diphenylphosphate) metabolizes to resorcinol and its conjugates (Freudenthal et al.,
2000; Ballesteros-Gómez et al., 2015). Resorcinol has also been found in cigarette smoke (WHO, 2006; Vaughan
et al., 2008). Thus, it is likely that resorcinol would be found in the urine of the general population.

There are only a few, mainly rather old occupational studies available on resorcinol exposure. Some old
epidemiological studies measuring thyroid parameters suggest that occupational resorcinol exposure is not
sufficient to cause clear thyroid effects (Lynch et al., 2002). In a resorcinol production plant in the United States,
measured air concentrations were in the range of 0.6–66 mg/m3 (Flickinger, 1976). Rubber workers were

T
exposed to resorcinol in concentrations of less than 0.3 mg/m3 during an average working day (Gamble et al.,

IP
1976), and in the tyre industry, air concentrations were typically under 0.1 mg/m3 (EC, 2002). Roberts et al.
studied the hypothyroidism of textile workers in the United Kingdom (Roberts et al., 1990). The air concentration

R
of resorcinol at the inlet of a local exhaust ventilation of a stenter machine was <20 µg/m3. Lind et al. have
studied hairdressers’ occupational skin exposure to hair dye compounds containing resorcinol. They found

SC
resorcinol in 21 hand wash samples of 33 hairdressers taken after the application of hair dye, and 20 out of 29
hair dressers’ hand wash samples taken after cutting newly-dyed hair (Lind et al., 2005). The same group has

U
also tested the permeability of the protective gloves used in hairdressing. The authors concluded that all the
tested disposable gloves provided considerable protection against the permeation of resorcinol, if they are
N
properly used (Lind et al., 2007). Hydrogen peroxide seems to have no effect on the permeability of protective
A
gloves to resorcinol in hairdressing (Lind et al., 2014). To our knowledge, no occupational biomonitoring studies
have been reported in the literature.
M

The aim of this study was to assess both environmental background and occupational exposure to resorcinol.
We investigated occupational exposure in hairdresser work, as well as in the manufacturing of tyres, adhesive
ED

resins, and glue-laminated timber. Human biomonitoring was the main method for assessing exposure, but we
also carried out some air measurements. We compared the urinary resorcinol levels of potentially occupationally
PT

exposed workers to those of the non-exposed reference population.

E

2. Materials and methods

CC

2.1 Study population

We investigated the background urinary resorcinol levels of non-occupationally exposed volunteers who worked
A

at the Finnish Institute of Occupational Health (FIOH). Altogether 103 occupationally non-exposed volunteers
from the Helsinki, Kuopio and Tampere regions (representing the regions in which FIOH is located) participated
in the study. After dilute urine samples were discarded (urine relative density < 1.010), one spot urine sample
remained (first morning void) from 101 volunteers (aged 24–65; 73 women and 28 men). Ten volunteers were
smokers (two men and eight women); three did not report smoking information.

3
The hairdressers were recruited via the Finnish hairdressers’ trade association. A total of 106 hairdressers from
all over Finland volunteered for the study. Our final number of volunteers who delivered urine sample was 77
(73%), and they were aged 21–67. Of these, 74 were women and three were men. Sixteen were smokers (one
man and 15 women); one person did not report smoking information.

Potential companies for the study were contacted via telephone or email. Three companies took part in the
study: a tyre company, a manufacturer of adhesive resin, and a producer of glue-laminated timber. The tyre
manufacturer used pure resorcinol as one of the production components. Pure resorcinol was also used in the
manufacture of adhesive resin, and the same adhesive resin was used in the production of glue-laminated

T
timber. These three companies were visited before the sample collection campaign began in order to identify

IP
the most significant work tasks with respect to resorcinol exposure. Altogether 46 workers aged 21–61
participated in the study (five women and 41 men). Nine men were smokers, and three individuals did not report

R
smoking information.

SC
The study design was given to all participants in written form, and each gave their written consent. The study
was approved by the local Coordinating ethics committee (HUS Joint Authority, Helsinki, Finland) (hairdresser
study: reference 346/13/03/00/2014; non-occupationally exposed volunteers and industrial workers: reference
347/13/03/00/149).
U
N
A
2.2 Workers’ urine sample collection
M

The participants were given instructions to collect either three (hairdressers) or five (industrial workers) spot
urine samples. The three samples were as follows: (i) First-morning-void sample after at least one day off. Since
ED

resorcinol has a short half-life (Kim and Matthews, 1987; WHO, 2006), we expected that after exposure, all
resorcinol would be excreted into the urine within one day. Accordingly, this sample represents the individual
background level of the participant. (ii) Post-shift sample immediately after work shift, and (iii) next-morning
PT

(first-morning-void) sample.

The five samples were as follows. (i) First-morning-void after a 2‒4-day holiday. This sample represents the
E

individual background level of the participant. The samples collected during day shifts were (ii) pre-shift (first-
CC

morning-void), (iii) post-shift (collected immediately after the shift), (iv) post-shift evening (collected around 7–
8 pm), and (v) next-morning (first-morning-void). Equivalent sample collection times were set for other work
shifts. In most cases, at least one work shift preceded the sample collection day.
A

All sample collections took place in 2015. Total resorcinol was measured (free + conjugated) unless otherwise
noted. The urine collection vessels and storage vessels were made of polypropylene, polystyrene, polyethylene
or glass. If not immediately analysed, the urine samples were stored as frozen until analysis (-20 °C).

2.3 Chemicals and materials

4
The resorcinol and derivatisation reagent 2,3,4,5,6-pentafluorobenzoyl chloride (99%) were purchased from
Sigma-Aldrich (Helsinki, Finland). The internal standard (ISTD) was D6-resorcinol (Sigma-Aldrich). The methanol
(MeOH) and n-hexane were of HPLC grade (Sigma–Aldrich), whereas analytical grade isooctane was used (Merck,
Darmstadt, Germany). We used anhydrous sodium acetate (puriss. p.a., Merck) and hydrochloric acid (≥ 37%,
Sigma–Aldrich) to prepare 1 M acetate buffer (pH 4.5). The potassium hydroxide (KOH), sodium chloride (NaCl)
and formic acid (Suprapur, 98–100%) were from Merck. The ß-Glucuronidase/Arylsulphatase, used as a
deconjugation agent, was obtained from Roche Diagnostic. We further purified the distilled water using ELGA’s
Maxima system (ELGA, Helsinki, Finland).

T
We prepared stock solutions of analytical and internal standards in methanol and stored them in a refrigerator

IP
for a maximum of three months. Working solutions (six different levels) were prepared in methanol at least
every fortnight. Calibration was conducted by spiking the urine matrix with working solutions.

R
We chose isolute ENV+ 100 mg/3 ml (Norlab, Vantaa, Finland) cartridges for solid phase extraction. Other

SC
cartridges tested were Oasis HLB (1cc Vac RC 60 mg; Waters, Helsinki, Finland) and Strata-X 100 mg/3 ml
(Phenomenex, Værløse, Denmark). Extractions were performed using a Varian VAC Elute SPS 24 system (Palo
Alto, USA) equipped with a Millipore pump (Molsheim, France).

U
N
2.4 Biological monitoring
A
2.4.1 Analytical equipment and conditions
M

HP 6890 gas chromatography was coupled to a HP 5973 mass selective detector operated in negative chemical
ionization (NCI) mode. We used helium as the carrier gas in constant flow mode with a flow rate of 1 ml/min.
ED

We introduced a 2 µl aliquot of sample extracts into a 5%-phenyl-95%-dimethylpolysiloxane column (HP-5MS,

30 m × 0.25 mm I.D. × 0.25 µm film thickness) in pulsed splitless mode at 290 °C with a head pressure of 172
kPa@1min. The oven temperature ramp was as follows: 100 °C (1 min)-15 °C/min-300 °C (10 min).
PT

For negative chemical ionization) we used methane (AGA/Linde Group, Finland as a reagent gas at a flow rate of
2 ml/min. The transfer line, ionization chamber and quadrupole were held at 300 °C, 150 °C and 150 °C,
E

respectively. The monitored ions for the derivatised target analytes were 498 and 499 for resorcinol; 502 and
CC

503 for D6-resorcinol.

We measured the specific gravity of the urine samples using a UG-1 refractometer (Atago, Tokyo, Japan). The
A

resorcinol concentrations were normalized to a specific gravity of 1.021, which is the average urine specific
gravity value of the Finnish adult population (see previous paper for details (Porras et al., 2014)). We measured
urinary creatinine concentrations using a colorimetric method. Concerning workers, no data were excluded due
to specific gravity/creatinine concentrations that were either too high or too low. When specific gravity was <
0.010, we used the 0.010 value in normalization; when specific gravity was > 1.030, we used the 1.030 value.
For a creatinine concentration below the limit of quantitation (LOQ), we used LOQ.

5
2.4.2 Sample preparation

The sample preparation procedure was modified from a previous study (Heinälä et al., 2017). The internal
standard (D6-resorcinol) was added to 1 ml of urine. For deconjugation, 50 µl of ß-glucuronidase/arylsulphatase
and 750 µl of sodium acetate buffer (pH 4.5) were added, and urine samples were held at 40 °C for 30 minutes.
To ensure that the target analytes were in non-ionized form, we added 200 µl of formic acid. The sample was
transferred to a conditioned (2 ml MeOH, 2 ml water) Isolute ENV+ cartridge, subsequently washed with 2 ml of
water and then dried for 40–45 minutes until the colour of the cartridge changed from dark to light
brown/orange. Elution was performed with 4 ml of MeOH. The eluate was evaporated to dryness under nitrogen

T
flow on a heating unit at 40 °C. The derivatisation procedure consisted of adding 1 ml 25% NaCl, 50 µl of 2M

IP
KOH, 100 µl of 5% pentafluorobenzoyl chloride (in hexane) and 2 ml of hexane. Samples were vortexed for two
minutes and centrifuged for five minutes (2500 rpm). The hexane layer was transferred to a clean test tube. We

R
repeated the extraction with 2 ml of hexane. We then removed the hexane layer and combined it with the first
fraction. The combined layers were evaporated to dryness and dissolved in 100 µl of isooctane.

SC
2.4.3 Method validation

U
We conducted method validation using control urine samples to which we added a known amount of resorcinol
(35 µg/l). Resorcinol recovery ranged from 76% to 90%, with an average of 84% (n = 8). Within-day repeatability
N
was 3.2% (n = 8) and long-term repeatability for a three-month period 11.4% (n = 46). LOQ was 4 µg/l and it was
A
set to the point at which the amount of added resorcinol was distinguished from the background signal with
sufficient confidence. Calibration ranged from 10 µg/l to 120 µg/l, and the method was linear up to 200 µg/l.
M

2.4.4 Statistical Analysis

ED

In the present work, we applied the non-parametric Wilcoxon signed-rank test, independent sample t test, and
Mann-Whitney U test for statistical comparisons of urinary specific gravity-normalised data. P-values of < 0.05
PT

were considered statistically significant, and p-values of < 0.001 highly significant. Statistical analyses were
conducted using IBM© SPSS© Statistics software, version 23 (IBM Corporation, NY, United States). In statistical
E

calculations, we replaced urinary resorcinol values below LOQ with values of LOQ/2 (Hornung and Reed, 1990).
CC

2.5 Air measurements

A

2.5.1 Sample collection

We collected air samples at the workplaces to measure resorcinol concentrations during the production
processes. Stationary and/or personal breathing zone (PBZ) samples were collected into XAD-7 OVS tubes (SKC-
226-57, SKC Inc.). We used calibrated pumps in sample collection, with an airflow of 1 l min. All PBZ samples
were collected outside of respiratory protective equipment (RPE) when in use. After collection, the samples
were stored in a freezer (−20°C) until analysis.

6
2.5.2 Sample preparation

Before analysis, the adsorbent materials were transferred to glass test tubes. The upper section – glass fibre
filter, XAD resin and polyurethane foam (PUF) – were moved to a test tube, and the lower section – XAD resin
and PUF – to another tube. We added 2 ml methanol with ISTD (D6-resorcinol) and extracted the mixture in an
ultrasonic bath for 15 minutes. After extraction, the mixture was filtered through an 0.45 µm filter (Millex HV
PVDF) into a glass vial.

2.5.3 Analysis

T
The analysis was performed using a liquid chromatography tandem-mass spectrometry (LC-MS/MS) technique

IP
The analytical conditions were as follows: Thermo Scientific instrumentation, Waters XTerra MS C18 3.5 µm, 2.1
mm x 100 mm column; flow rate of 0.3 ml/min; Injection volume of 10 µl; mobile phase: (A) H2O 95%/MeOH

R
5%/formic acid 0.1%, (B) H2O 5%/MeOH 95%/formic acid 0.1%; and gradient elution: (A) 90% (1 min) – (90%–
30%, 6 min) – (30%–90%, 0.5 min). The ions monitored for identification and quantification were m/z 111 for

SC
resorcinol, and m/z 115 for D6-resorcinol. The LOQ depends on air volume.

2.6 Calculation of biomonitoring equivalent for resorcinol U

N
Since there are no existing biological limits or guidance values for resorcinol, we used the methodology described
A
by Angerer et al. (Angerer et al., 2011) to estimate the BE for resorcinol. As a starting point, we used the
M

acceptable daily intake (ADI) value of 0.12 mg/kg bw of the European Food Safety Authority (EFSA) (EFSA, 2010).
Under EU REACH legislation, the registrant of resorcinol has also given limit values for external resorcinol intake.
The EFSA’s ADI is based on the no observable adverse effect level (NOAEL) of 50 mg/kg, observed in a long-term
ED

study of rats. The application of an assessment factor of 100 covered inter- and intra-species uncertainties, and
an additional factor of 3 covered the steep dose-response curve for acute neurotoxic effects. The REACH-derived
no effect levels (DNELs) are based on a subchronic rat study with a NOAEL of 80 mg/kg (Tukes, 2017) but the
PT

applied assessment factors and their justification has not been published. The available ADI and DNEL values are
presented in Table 1.
E
CC

BE was calculated using the following formula

A

D×BW×Fue
𝐶𝑠𝑠 = ,
V24

Css×V24
which is derived from the basic formula of 𝐷 = (Angerer et al., 2011). Here, D is the external dose as
Fue ×BW

mg/kg bw, Css is the urinary level of the substance at steady state, V24 is the estimated average 24-hour urinary
volume, and FUE is the mass of metabolite excreted to the urine during the 24-hour per mass of parent compound
ingested.

7
Since there are no human data on FUE in humans, we based FUE on rat data, showing that > 80% of orally
administered resorcinol is excreted in the urine within 24 hours (Kim and Matthews, 1987). We used a value of
1.7 l/day as the 24-hour urinary volume (V24) and 70 kg as the average body weight in the calculations.

3. Results and discussion

3.1 Environmental exposure

Environmental exposure to resorcinol was investigated in 101 volunteers with no occupational exposure. Table

T
2 presents the results. The second column of Table 2 shows the original data with no adjustments. The third

IP
column shows the data normalized to a specific gravity of 1.021 (see Materials and methods). Resorcinol
concentrations adjusted to creatinine excretion are shown in the last column. Normalized expression was used

R
throughout the work unless otherwise specified.

SC
All but one sample of the non-occupationally exposed reference population contained measurable amounts of
resorcinol. The normalized geometric mean urinary concentration was 66 µg/l. Median concentration was 42

U
µg/l, 95th percentile 1856 µg/l, and the data range was from below LOQ (4 µg/l) to 9996 µg/l. The difference
between the minimum and maximum urinary resorcinol concentrations was more than three orders of
N
magnitude.
A
M

The highest urinary resorcinol concentrations (9996 µg/l and 7365 µg/l) came from two women who reported
ED

having used hair design products. To test the possible effect of such products on urinary resorcinol
concentration, these two women were invited to a small-scale follow-up study in which they were asked to
collect two urine samples: (i) after at least a one-day break from using the hair design products, and (ii) after
PT

normal use of the products. If the hair design products affect the urinary resorcinol concentration, the second
sample should contain more resorcinol than the first one. This was, however, not the case. The follow-up results
E

of the woman with the initially highest urinary resorcinol concentration (9996 µg/l) were 32 µg/l and 34 µg/l,
CC

respectively. The follow-up results of the other woman were 2363 µg/l and 102 µg/l, respectively. Possible
contamination was excluded as a cause of high concentrations, since free resorcinol corresponded to only 0.5%
of the total resorcinol concentration.
A

Resorcinol has also been found in tobacco smoke (WHO, 2006; Vaughan et al., 2008). Thus, we tested the effect
of smoking on urinary resorcinol concentrations. The geometric mean of the urinary resorcinol concentration
among the smokers was 61 µg/l (n = 10), whereas the respective concentration among the non-smokers was 67
µg/l (n = 91). The median concentrations were also relatively close to each other (58 µg/l and 41 µg/l,
respectively) but the 95th percentile and the maximum among non-smokers were about 5 and 20 times higher,

8
respectively. In fact, among the 20 highest urinary resorcinol concentrations, only one result was from a smoker
(concentration 487 µg/l).

As the seven highest urinary resorcinol concentrations were all from women, we tested the effect of sex (Table
3). The geometric mean concentration of the women was 84 µg/l, which was more than twice that of the
respective mean of the men (35 µg/l). The same was true for median concentrations. The 95th percentile of
women (2072 µg/l) was more than three times higher than the respective value of the men (587 µg/l). The
scatter of the women’s data (range <LOQ–9996 µg/l) was much larger than the scatter of the men’s data (range
6–1065 µg/l). The difference between women and men was statistically significant (Mann-Whitney U: p = 0.012;

T
independent sample t test: p = 0.026).

R IP
SC
3.2 Occupational exposure

We studied the occupational resorcinol exposure of hairdressers (Table 4), and industrial workers in three

U
different factories (Tables 5–7). In these plants, either pure resorcinol or resorcinol-containing products were
used.
N
A
Hairdressers
M

Three urine samples were collected from 77 hairdressers: after the weekend off work or at least one day off
work (first morning void), immediately after the work shift (post-shift), and the next morning (first morning void).
ED

Altogether we had 231 samples, and almost all of them (98%) contained measurable amounts of resorcinol. The
geometric mean and median concentrations of the after-holiday samples (87 µg/l and 78 µg/l, respectively) were
higher than the respective concentrations of the post-shift and next-morning samples. The highest amounts of
PT

resorcinol were present in the next-morning samples, whereas the lowest values were in the post-shift samples.
The difference between the after-holiday samples, and both the post-shift and next-morning samples was
E

statistically significant (Wilcoxon: p ≤ 0.01).

CC

We tested the effect of smoking on the results. On all three sampling occasions, the geometric means of the
A

smokers (n = 16) were slightly higher than that of the non-smokers (n = 60). However, statistically there was no
difference between the data (independent sample t test: p ≥ 0.520; Mann-Whitney U: p ≥ 0.606).

Only three hairdressers were men, so the effect of sex could not be statistically compared. The geometric mean
of hairdresser women (n =74) was 89 µg/l, 51 µg/l, and 54 µg/l for the after-holiday, post-shift, and next-morning
samples, respectively. The respective means for men were 78 µg/l, 45 µg/l, and 46 µg/l.

9
The geometric mean concentration of the hairdresser’s after-holiday samples (87 µg/l, Table 4) was higher than
that of the non-occupationally exposed (66 µg/l, Table 2), but in turn, the respective means of the post-shift (51
µg/l) and next-morning (54 µg/l) samples were lower than this. The 95th percentile of the urinary concentration
data of the general population is usually used to set a non-health based biological guidance value (BGV), which
can then be used to identify occupational exposure from background exposure (SCOEL, 2013). Similarly, in
environmental health, the 95th percentile can be used as a statistically set reference value for the general
population. The 95th percentile of the urinary resorcinol data of the non-occupationally exposed was 1856 µg/l
(Table 2), but because there was a statistically significant difference between the background urinary levels of
women and men, we calculated separate guidance values for males and females. The respective 95th percentiles

T
for women and men were 2072 µg/l and 587 µg/l (Table 3).

IP
In the after-holiday samples, one sample from the female participant contained 2286 µg/l resorcinol, which is

R
above the 95th percentile of the female data. The sample was collected after a five-day holiday. The urinary
resorcinol concentration of one male after-holiday sample (1725 µg/l) exceeded the 95th percentile of the male

SC
data. This time the sample was collected after a two-day holiday. One post-shift sample of a man contained 1541
µg/l resorcinol, which is above the 95th percentile value. In two women’s next-morning samples, the resorcinol

U
concentration exceeded the 95th percentile (2698 µg/l and 3029 µg/l). One next-morning sample from a man
(1559 µg/l) contained more resorcinol than the 95th percentile. Taking into account the fact that the values
N
exceeding the 95th percentiles were distributed within all sampling times, that there was no general trend for
A
increasing levels after the shift and that there was a huge variability in the background levels in the
occupationally non-exposed population, these single increases are unlikely to reflect occupational exposure.
M
ED

Tyre manufacturing plant

Table 5 presents the urinary resorcinol concentrations of workers in the tyre manufacturing plant. Five urine
PT

samples were collected: first morning void after a weekend holiday (2 days), pre-shift, post-shift, evening, and
next-morning. All samples contained measurable amounts of resorcinol. There was a clear trend in the geometric
means of the workers: the geometric mean increased from pre-shift samples and reached its maximum in the
E

evening. The geometric mean of the next-morning samples was about the same as that of the pre-shift samples.
CC

No such trend was observed in the median concentrations.

A

Ten out of eleven workers in the tyre manufacturing plant were men. Figure 1 shows the results of men
compared to the 95th percentiles among the general population. As in all the data (see Table 5), the geometric
mean reaches its maximum in the evening (113 µg/l). The maximum concentrations of the post-shift and evening
samples were above the 95th percentile of the urinary resorcinol concentration of the non-occupationally
exposed population (1856 µg/l, dashed horizontal line in Figure 1). On all sampling occasions, the maximum
concentrations were above the 95th percentile of the urinary resorcinol concentration of non-occupationally

10
exposed men (587 µg/l, dotted horizontal line in Figure 1). We observed no increasing trend in the resorcinol
concentrations of the female worker.

The resorcinol concentration(s) of three male workers exceeded the 95 th percentile concentration of the non-
occupationally exposed men. The post-shift urine sample of one calender operator contained 602 µg/l of
resorcinol, which is slightly above the 95th percentile 587 µg/l. The post-shift sample (2296 µg/l) and evening-
sample (3575 µg/l) of one maintenance worker exceeded the 95th percentile. The concentration in the next-
morning sample was already clearly below the 95th percentile (331 µg/l). Interestingly, the after-holiday sample
also contained relatively high amounts of resorcinol (1498 µg/l). The third male worker’s job was filling the

T
chemical silos. His post-shift sample was missing, but the evening sample (638 µg/l) and the next-morning

IP
sample (824 µg/l) contained more resorcinol than the 95th percentile concentration of the non-occupationally
exposed men. The pre-shift sample contained 847 µg/l of resorcinol, which is also above the 95th percentile. This

R
might indicate exposure during the previous working day. His after-holiday sample contained 565 µg/l of
resorcinol.

SC
U
Three stationary air samples collected from the factory: (i) in an area in which tyre mass was produced, (ii) in an
area in which raw-material was fed, and (iii) at a weighing point. In this case, stationary air samples were selected
N
over PBZ samples because the workers were mobile and did not stay in one place for long. No detectable levels
A
of resorcinol were found in any of the air samples (all results were below LOQ 0.02 mg/m3).
M

Phenolic resin manufacturer

ED

Table 6 presents the urinary resorcinol concentrations of the workers of the phenolic resin manufacturing plant.
All workers (one woman and four men) were reactor operators, dosing resorcinol. Here too, we collected five
urine samples from each participant. All samples contained measurable amounts of resorcinol. There was a clear
PT

trend in the geometric means of the samples: the geometric mean reached its maximum in the post-shift
samples and decreased thereafter. The trend in median concentrations was slightly different: the median
E

increased from the post-shift samples to evening samples. The median of the next-morning samples was still
CC

about double that of the pre-shift samples. The highest urinary resorcinol concentrations were from a woman.
However, all urinary resorcinol concentrations were less than 500 µg/l, which is slightly below the 95th percentile
of the non-occupationally exposed men (587 µg/l).
A

We collected three air samples from the factory: two stationary samples from the reactor room in which
resorcinol-containing resin was manufactured, and one PBZ sample of the worker operating the reactor. No
detectable levels of resorcinol were observed in the stationary samples (all results below LOQs 0.02 mg/m3 and

11
0.14 mg/m3). Small amounts of resorcinol were detected in the PBZ sample outside the worker’s respirator (0.08
mg/m3). This work task lasted 134 minutes. When this is averaged for an eight-hour work shift (assuming an
inhalation exposure of 0.01 mg/m3 during the rest of the work-shift), the level corresponds to 0.03 mg/m3. These
levels are very low compared to any available occupational exposure limits (OEL) for resorcinol, which seem to
vary between 10 mg/m3 and 46 mg/m3 (IFA, 2018) and also clearly below the DNELworker given in Table 1.

Manufacturing of glue-laminated timber

T
Table 7 presents the urinary resorcinol concentrations of six male workers at the plant that manufactures glue-

IP
laminated timber. Workers either operated a glue application machine, a drying oven, or finished the products.
Five urine samples were collected from each participant. All samples contained measurable amounts of

R
resorcinol. There was a slight trend in the geometric means of the urinary resorcinol concentrations: they
increased from pre-shift samples to evening samples, and decreased again in the next-morning samples. We

SC
observed no such trend in the median concentrations, even though also in this case, the highest median
concentration was in the evening samples. The maximum concentration in the samples collected after work

U
shifts was 264 µg/l, which is more than two times lower than the 95th percentile of the non-occupationally
exposed men (587 µg/l, Table 3). The highest urinary resorcinol concentration of 561 µg/l was measured in one
N
after-holiday sample (collected after a two-day holiday).
A
M

Four air samples were collected from the factory: two PBZ samples from operators of the glue application
machine, and two stationary samples; one from the glue application machine and one from the nearby drying
ED

oven. The resorcinol concentration of all samples was below the LOQ of 0.01 mg/m3.
PT

Comparison to calculated biomonitoring equivalent

The biomonitoring equivalent (BE) calculated using the EFSA’s ADI value of 0.12 mg/kg as a starting point was:
E

0.12×70×0.80
𝐶𝑠𝑠 (𝐵𝐸) = =3.95 mg/l
1.7
CC

Taking into account the fact that the EFSA’s ADI is the most conservative value given for resorcinol, health risks
at urinary levels below the calculated BE are unlikely. It should, however, be noted that this is a very crude
A

estimation and some uncertainties arise from the fact that there is no specific data on the excretion kinetics of
resorcinol in humans, which caused us to base the FUE on rat data. If humans excreted less resorcinol in their
urine, for example, only 50% instead of rats’ 80% excretion, the BE level would be 2.5 mg/l. Even though in
some occupationally and non-occupationally exposed female population samples the concentrations were
high, up to 3–10 mg/l, it should be also noted that urinary resorcinol levels are highly variable, and these

12
highest levels may represent peak urinary concentrations, whereas BE represents the steady state
concentration.

4. Discussion

To the authors’ knowledge, no published general population data are available on the urinary resorcinol
concentrations. Due to the widespread occurrence of resorcinol (Schmiedel and Decker, 2000; EC, 2002; WHO,
2006), we hypothesized that the general population is likely to be exposed to resorcinol. The present study

T
confirms this hypothesis. Almost all the urine samples (99%) of the non-occupationally exposed volunteers

IP
contained measurable amounts of resorcinol. The scatter of the urinary resorcinol data was rather wide – the
data range was from below the detection limit of 4 µg/l to almost 10 000 µg/l (95th percentile 1856 µg/l) (Table

R
2). Smoking had no effect on urinary resorcinol, as the highest concentrations were detected among non-
smokers. In contrast, there was a statistically significant gender difference in the urinary resorcinol

SC
concentrations (Table 3). The reasons for this difference remain unclear, although it could be speculated that
the use of personal care products could be one possible reason.

U
The occupational exposure measurements in this study indicate that occupational resorcinol exposure is mainly
N
at the same level as that of the non-occupationally exposed population. However, we did observe some
occupational exposure in the manufacturing of tyres among a calender operator, when filling the silos and during
A
maintenance work. Both the post-shift and evening samples of three male workers contained more resorcinol
M

than the 95th percentile of the data of non-occupationally exposed men, and this may indicate occupational
exposure to resorcinol. However, in the case of two workers, the concentrations were still below the maximum
concentration of the non-occupationally exposed men. The resorcinol concentration of the post-shift and
ED

evening samples of the maintenance worker, on the other hand, was clearly above the maximum of the non-
occupationally exposed men, and this was probably due to occupational exposure. Even though the air
PT

concentrations were not measured during the maintenance work, the very low air concentrations measured
elsewhere in the factory suggest exposure via dermal contact.
E

The urinary resorcinol data of the workers of the phenolic resin manufacturing plant also had clear trends (Table
6): the median increased from the after-holiday samples to evening samples and decreased thereafter. The
CC

geometric mean reached its maximum in the post-shift samples, after which it decreased. However, all urinary
resorcinol concentrations were less than 500 µg/l, remaining below both the 95th percentile of non-
occupationally exposed whole population and the 95th percentile of the non-occupationally exposed males. The
A

only measurable resorcinol air concentration was detected in this factory (PBZ sample of the reactor operator),
but it was well below any OEL for resorcinol. Thus, in this case, although the biomonitoring levels showed some
increase in resorcinol exposure, which is in line with occupational exposure, the increase was not enough to
exceed the 95th percentiles of the reference population.

13
In the case of the workers in the plant that manufactures glue-laminated timber, we observed no clear trends
in the urinary resorcinol concentrations. The maximum resorcinol concentration of 561 µg/l was detected in one
after-holiday sample collected after a two-day holiday. All samples collected after work shifts contained a
maximum of 264 µg/l of resorcinol. Four air samples collected in the factory contained no measurable amounts
of resorcinol.

The urinary resorcinol concentrations of the hairdressers were at the same level as those of the reference
population. In some of the hairdressers’ samples, the resorcinol concentration exceeded the 95th percentile of
the data of non-occupationally exposed people. However, there were no trends showing that the post-shift

T
values would have been higher than the after-holiday values. In addition, in two after-holiday samples the

IP
resorcinol concentration exceeded the 95th percentile of the non-exposed reference population. Thus, it is
unlikely that the reason for such high concentrations is due to occupational exposure. This data suggest that

R
hairdressers are not exposed to resorcinol more than the non-occupationally exposed population.

SC
Resorcinol is listed as an endocrine disruptor due to its ability to disrupt thyroid function by inhibiting thyroxin
peroxidase and affecting iodine uptake (Tukes, 2017). In animals, these effects seem to occur at rather high
doses and quantitative information on these effects in humans is limited. The health-based values available are

U
therefore based on animal data. The lowest health-based limit value set for resorcinol is ADI of 0.12 mg/kg bw
N
set by EFSA (EFSA, 2010). The DNEL values set for resorcinol by the industry under EU’s REACH regulation are
approximately 3–7 times higher (see Table 1). Current OELs in different countries are even higher (between 10
A
mg/m3 and 46 mg/m3, which represents 1.4 mg/kg bw to 6.6 mg/kg bw in eight-hour occupational exposure
M

assuming 100% absorption via inhalation). When we compare the urinary levels measured in the current study
to the calculated BE of 4 mg/l for resorcinol based on EFSA’s ADI, we note that most of the measurements,
including all 95th percentile levels, remain below this level. However, some samples from the reference
ED

population (samples from two occupationally non-exposed females) and some samples from the occupationally
exposed population were either close to this or even exceeded this level. Considering the uncertainties related
PT

to the estimation of the BE level (especially related to the excretion kinetics of resorcinol, see results), this might
raise some concern, especially for the reference population. However, it should be noted that the EFSA’s ADI
seems to be rather conservative, and the high uncertainty factor has been used in its setting (EFSA, 2010). In
E

addition, it should also be noted that the BE estimated for resorcinol represents a steady state urinary
CC

concentration, whereas the highest levels measured in spot urine samples may represent peak levels. Thus,
according to the current knowledge on the health effects of resorcinol, health risks based on the currently
measured urinary levels are considered unlikely. Highly variable and high peak levels among non-occupationally
A

exposed females especially may, however, warrant further studies in larger population groups. It seems that
resorcinol levels may show significant day-to-day variation. In the case of the two highest urinary concentrations
among the non-occupationally exposed individuals, new samples were collected a few weeks later, and the
results were well below the BE.

As is typical with occupational exposure studies, the number of participants is often relatively low. Apart from
the hairdressers, this was also the case in our study, which was limited by both the small number of companies

14
involved and the small number of participants in the companies. This was because the number of employees
working with resorcinol in each company was limited. Obviously, the survey did not cover all companies using
resorcinol in Finland. Although the results obtained in this study help fill the gap in exposure data, further data
are still needed to obtain a broader overview of occupational exposure to resorcinol. In addition, the sources of
the background resorcinol levels among general population need to be clarified, and more data on resorcinol
levels in other countries are needed. One interesting observation in this study was also the gender difference,
showing higher resorcinol levels among females. This also merits further study.

T
Conclusions

IP
Detectable and highly variable urinary resorcinol levels were observed among the non-occupationally exposed

R
reference population, with females showing the highest levels. In tyre and phenolic resin manufacturing, slightly
increasing urinary resorcinol levels were seen, suggesting occupational exposure. The increase was clearest in

SC
the tyre manufacturing industry. The most likely source of exposure was dermal contamination. Hairdressers
were not occupationally exposed to resorcinol.

U
The urinary resorcinol levels measured in this study are unlikely to cause any significant health risk according to
N
our current knowledge on the dose responses of resorcinol. However, highly variable urinary levels and high
peak levels, especially among the non-occupationally exposed reference population warrant further study.
A
Funding
M

The Finnish Work Environment Fund (Project No. 114078) and the Ministry of Social Affairs and Health (Project
No. 417914012) provided financial support for this research.
ED

Conflict of interest
PT

The authors declare no conflicts of interest.

E
CC

Acknowledgements

We would like to acknowledge all the companies and individuals who participated in the study, as well as the
A

Finnish hairdressers’ trade association for their help with recruitment. We thank Moona Teljomaa and Jenna
Nordström for the creatinine measurements, and Raija Vastapuu for her assistance in preparing the air samples.

15
References

Angerer, J., Aylward, L.L., Hays, S.M., Heinzow, B., Wilhelm, M., German Human Biomonitoring Commiss, 2011.
Human biomonitoring assessment values: Approaches and data requirements. International Journal of Hygiene
and Environmental Health 214, 348-360. http://dx.doi.org/10.1016/j.ijheh.2011.06.002

Ballesteros-Gómez, A., Van den Eede, N., Covaci, A., 2015. In vitro human metabolism of the flame retardant
resorcinol bis(diphenylphosphate) (RDP). Environ Sci Technol 49, 3897-3904.
http://dx.doi.org/10.1021/es505857e

T
CEHOS, 2012. Evaluation of 22 SIN List 2.0 substances according to the Danish proposal on criteria for endocrine

IP
disrupters. Danish Centre on Endocrine Disrupters. http://orbit.dtu.dk/files/51554411/SINreportandAnnex.pdf

EC, 2002. Study on the scientific evaluation of 12 substances in the context of endocrine disruptor priority list of

R
actions. Produced by WRc-NSF, UK, for the European Commission. European Commission.

SC
http://ec.europa.eu/environment/chemicals/endocrine/pdf/wrc_report.pdf

EFSA, 2010. EFSA Panel on Food Additives and Nutrient Sources added to Food (ANS); Scientific Opinion on the

U
use of Resorcinol as a food additive. EFSA J 8, 1411. http://dx.doi.org/10.2903/j.efsa.2010.1411
N
EFSA, 2014. EFSA Panel on Additives and Products or Substances used in Animal Feed (FEEDAP); Scientific
Opinion on the safety and efficacy of tannic acid when used as feed flavouring for all animal species. EFSA J 12,
A
3828. http://dx.doi.org/10.2903/j.efsa.2014.3828
M

Flickinger, C.W., 1976. The benzenediols: catechol, resorcinol and hydroquinone - a review of the industrial
toxicology and current industrial exposure limits. American Industrial Hygiene Association journal 37, 596-606.
ED

http://dx.doi.org/10.1080/0002889768507526

Freudenthal, R.I., McDonald, L.J., Johnson, J.V., McCormick, D.L., Henrich, R.T., 2000. Comparative metabolism
and toxicokinetics of C-14-resorcinol bis-diphenylphosphate (RDP) in the rat, mouse, and monkey. Int J Toxicol
PT

19, 233-242. http://dx.doi.org/10.1080/10915810050202042

Gamble, J.F., McMichael, A.J., Williams, T., Battigelli, M., 1976. Respiratory function and symptoms: an
E

environmental-epidemiological study of rubber workers exposed to a phenolformaldehyde type resin. American

CC

Industrial Hygiene Association journal 37, 499-513. http://dx.doi.org/10.1080/0002889768507508

Hamann, D., Yazar, K., Hamann, C.R., Thyssen, J.P., Liden, C., 2014. p-Phenylenediamine and other allergens in
A

hair dye products in the United States: a consumer exposure study. Contact Dermat 70, 213-218.
http://dx.doi.org/10.1111/cod.12164

Heinälä, M., Ylinen, K., Tuomi, T., Santonen, T., Porras, S.P., 2017. Assessment of occupational exposure to
bisphenol A in five different production companies in Finland. Annals of work exposures and health 61, 44-55.
http://dx.doi.org/10.1093/annweh/wxw006

16
Hornung, R.W., Reed, L.D., 1990. Estimation of average concentration in the presence of nondetectable values.
Applied Occupational and Environmental Hygiene 5, 46-51.
http://dx.doi.org/10.1080/1047322X.1990.10389587

IFA, 2018. GESTIS - International limit values for chemical agents (Occupational exposure limits, OELs). Institut
für Arbeitsschutz der Deutschen Gesetzlichen Unfallversicherung (IFA), Germany. http://limitvalue.ifa.dguv.de/

Kim, Y.C., Matthews, H.B., 1987. Comparative metabolism and excretion of resorcinol in male and female F344
rats. Fund Appl Toxicol 9, 409-414. http://dx.doi.org/10.1016/0272-0590(87)90023-6

T
Lind, M.-L., Boman, A., Sollenberg, J., Johnsson, S., Hagelthorn, G., Meding, B., 2005. Occupational dermal

IP
exposure to permanent hair dyes among hairdressers. Annals Occup Hyg 49, 473-480.
http://dx.doi.org/10.1093/annhyg/mei030

R
Lind, M.-L., Johnsson, S., Liden, C., Meding, B., Boman, A., 2014. The influence of hydrogen peroxide on the

SC
permeability of protective gloves to resorcinol in hairdressing. Contact Dermat 72, 33-39.
http://dx.doi.org/10.1111/cod.12296

U
Lind, M.-L., Johnsson, S., Meding, B., Boman, A., 2007. Permeability of hair dye compounds p-phenylenediamine,
toluene-2,5-diaminesulfate and resorcinol through protective gloves in hairdressing. Annals Occup Hyg 51, 479-
N
485. http://dx.doi.org/10.1093/annhyg/mem028
A
Lynch, B.S., Delzell, E.S., Bechtel, D.H., 2002. Toxicology review and risk assessment of resorcinol: thyroid effects.
M

Regulat Toxicol Pharmacol 36, 198-210. http://dx.doi.org/10.1006/rtph.2002.1585

Nakamura, Y., Tsuji, S., Tonogai, Y., 2003. Method for analysis of tannic acid and its metabolites in biological
ED

samples: Application to tannic acid metabolism in the rat. J Agric Food Chem 51, 331-339.
http://dx.doi.org/10.1021/jf020847+

Porras, S.P., Heinälä, M., Santonen, T., 2014. Bisphenol A exposure via thermal paper receipts. Toxicology Letters
PT

230, 413-420. http://dx.doi.org/10.1016/j.toxlet.2014.08.020

Roberts, F.P., Wright, A.L., O'Hagan, S.A., 1990. Hypothyroidism in textile workers. J Soc Occup Med 40, 153-156.
E

http://dx.doi.org/10.1093/occmed/40.4.153
CC

SCCP, 2008. Opinion on resorcinol. Scientific Committee on Consumer Products, European Commission.
http://ec.europa.eu/health/ph_risk/committees/04_sccp/docs/sccp_o_124.pdf
A

Schmiedel, K.W., Decker, D., 2000. Resorcinol. Ullmann's Encyclopedia of Industrial Chemistry. Wiley-VCH Verlag
GmbH & Co. KGaA.
SCOEL, 2013. Methodology for the Derivation of Occupational Exposure Limits - Key Documentation (version 7).
Scientific Committee on Occupational Exposure Limits (SCOEL).
http://efcc.eu/Document.aspx?di=1225&fn=2013

17
Tukes, 2017. Substance Evaluation Conclusion as required by REACH Article 48 and evaluation report for
Resorcinol. Finnish Safety and Chemicals Agency (Tukes), Finland.
https://echa.europa.eu/documents/10162/fedfa3b0-f8a2-66b4-2a08-7f686df46994

Vaughan, C., Stanfill, S.B., Polzin, G.M., Ashley, D.L., Watson, C.H., 2008. Automated determination of seven
phenolic compounds in mainstream tobacco smoke. Nicot Tobacco Res 10, 1261-1268.
http://dx.doi.org/10.1080/14622200802123146

WHO, 2006. Concise International Chemical Assessment Document 71: Resorcinol. World Health Organization.
http://www.who.int/ipcs/publications/cicad/cicad71.pdf

T
IP
Yazar, K., Boman, A., Liden, C., 2009. Potent skin sensitizers in oxidative hair dye products on the Swedish market.
Contact Dermat 61, 269-275. http://dx.doi.org/10.1111/j.1600-0536.2009.01612.x

R
Yazar, K., Boman, A., Liden, C., 2012. p-Phenylenediamine and other hair dye sensitizers in Spain. Contact Dermat

SC
66, 27-32. http://dx.doi.org/10.1111/j.1600-0536.2011.01979.x

Yeung, D., Kantor, S., Nacht, S., Gans, E.H., 1983. Percutaneous absorption, blood levels, and urinary excretion

U
of resorcinol applied topically in humans. Int J Dermatol 22, 321-324. http://dx.doi.org/10.1111/j.1365-
4362.1983.tb02149.x
N
Yeung, D., Nacht, S., Gans, E.H., 1981. High-performance liquid chromatographic detection of free resorcinol in
A
plasma and in urine J Chromatogr 224, 513-518.
M
ED
E PT
CC
A

18
 T
R IP
SC
U
Figure 1. Total urinary resorcinol concentrations of male workers in tyre manufacturing plant (n = 10).
N
Concentrations were normalized to a specific gravity of 1.021. The horizontal dashed line and dotted line are the
A
95th percentile of the data of the non-occupationally exposed population (1856 µg/l), and the 95th percentile of
the data of non-occupationally exposed men (587 µg/l), respectively.
M
ED
E PT
CC
A

19
Table 1. Available health-based limit values for resorcinol.

Target population Exposure Limit value Source

route

General population Oral 0.12 mg/kg bw (ADI) (EFSA, 2010)

Workers Inhalation 5.6 mg/m3 (ca. 0.8 mg/kg bwa) (DNELworker) ECHAb (REACH registration dossier)

Workers Skin 40 mg/kg bw (DNELworker) ECHA (REACH registration dossier)

General population Oral 0.4 mg/kg bw (DNEL) ECHA (REACH registration dossier)

T
IP
General population Skin 20 mg/kg bw (DNEL) ECHA (REACH registration dossier)

a
Calculated assuming inhalation volume of 10 m3 during 8-h work shift and 100% absorption via inhalation.

R
b
echa.europa.eu

SC
U
N
A
M
ED
E PT
CC
A

20
able 2. Total urinary resorcinol concentrations of non-occupationally exposed reference population (n = 101).

Total resorcinol concentration in urine

a
µg/l (original) µg/l (normalized)b µg/g creatinine
Arithmetic mean 309 405 336
Geometric mean 50 66 53
Minimum < LOQ < LOQ < LOQ
Median 34 42 35
th
75 percentile 176 190 162

T
95th percentile 1217 1856 1476

IP
Maximum 9520 9996 7751
LOQ, limit of quantitation (4 µg/l).

R
a
Original results with no adjustments.

SC
b
Results normalized to specific gravity of 1.021 (see Materials and methods).

U
N
A
M
ED
E PT
CC
A

21
Table 3. Total urinary resorcinol concentrations of non-occupationally exposed women (n = 73) and men (n =
28). Original results and creatinine-adjusted results are given for comparison.

Total resorcinol concentration in urine

µg/l (original)a µg/l (normalized)b µg/g creatinine
Women Men Women Men Women Men
AM 381 123 513 124 432 86
GM 61 30 84 35 69 26
Minimum < LOQ 4 < LOQ 6 < LOQ 5

T
Median 40 16 55 22 43 17

IP
75th 185 84 277 107 343 62
percentile

R
95th 1601 607 2072 587 1740 407

SC
percentile
Maximum 9520 1217 9996 1065 7751 726
AM, arithmetic mean; GM, geometric mean; LOQ, limit of quantitation.

a
Original results with no adjustments.
U
N
b
Results normalized to specific gravity of 1.021.
A
M
ED
E PT
CC
A

22
Table 4. Total urinary resorcinol concentrations of hairdressers (n = 77) split into different sampling times.
Concentrations were normalized to a specific gravity of 1.021.

Total resorcinol concentration in urine (µg/l)

After-holiday Post-shift Next-morning
Arithmetic mean 272 163 232
Geometric mean 87 51 54
Minimum 4 < LOQ < LOQ
Median 78 44 44

T
75th percentile 280 126 135

IP
95th percentile 1239 828 1408
Maximum 2283 1818 3029

R
LOQ, limit of quantitation.

SC
U
N
A
M
ED
E PT
CC
A

23
Table 5. Total urinary resorcinol concentrations of workers in tyre manufacturing plant (n = 11) split into different
sampling times. Concentrations were normalized to a specific gravity of 1.021.

Total resorcinol concentration in urine (µg/l)

After-holiday Pre-shift Post-shift Evening Next-morning
Arithmetic mean 234 301 358 523 176
Geometric mean 76 87 108 124 88
Minimum 10 8 13 11 10
Median 60 91 82 72 114

T
75th percentile 117 158 226 424 174

IP
95th percentile 1031 1258 1534 2253 578
Maximum 1498 1595 2296 3575 824

R
SC
U
N
A
M
ED
E PT
CC
A

24
Table 6. Total urinary resorcinol concentrations of workers in phenolic resin manufacturing plant (n = 5) split
into different sampling times. Concentrations were normalized to a specific gravity of 1.021.

Total resorcinol concentration in urine (µg/l)

After-holiday Pre-shift Post-shift Evening Next-morning
Arithmetic mean 47 75 159 162 108
Geometric mean 25 41 91 85 66
Minimum 5 6 10 11 7
Median 19 46 111 121 93

T
75th percentile 89 153 184 123 119

IP
95th percentile 106 155 373 423 231
Maximum 110 155 420 498 259

R
SC
U
N
A
M
ED
E PT
CC
A

25
Table 7. Total urinary resorcinol concentrations of workers in plant manufacturing glue-laminated timber (n = 6)
split into different sampling times. Concentrations were normalized to a specific gravity of 1.021.

Total resorcinol concentration in urine (µg/l)

After-holiday Pre-shift Post-shift Evening Next-morning
Arithmetic mean 110 43 65 67 43
Geometric mean 33 31 35 47 35
Minimum 10 11 11 17 11
Median 21 37 30 50 39

T
75th percentile 28 43 36 79 42

IP
95th percentile 428 97 207 159 88
Maximum 561 115 264 183 103

R
SC
U
N
A
M
ED
E PT
CC
A

26